ChapterPDF Available

Spermatozoal Phylogeny of the Vertebrata.

January 1999

In book: The Male Gamete: From Basic Science to Clinical Applicatioend (pp.303-331.)
Chapter: 29
Publisher: Cache River Press
Editors: Claude Gagnon

Authors:

Barrie G. M. Jamieson

University of Queensland

Generalized or hypothetical ancestral tetlapod spermatozoon.

…

Comparison ol the paraxonemal rod in Ascaphus with the accessory axonemal libers and undulating membrane of r urodele and a hylid fiog, as seen in transverse sections of the flagel1um. Only about one half of the length of the undulating membrane is shown in the latter two species. In the urodele, the major fiber is much larger than illdicated relative to the minor t'iber. Presence of a fiber on each side of the axoneme, at doublets 3 and 8, in urodeles is probably the plesionorphic lissamphibian (and sarcopterygian) condition-The juxta axonemal fiber at 3 is usually absent in urodeles and the density in this position in Plethodon may not be homologous witb this; the axial tiber is hypertrophied. The two fibers on each side of the axonen]e are reduced in Alura to one tiber, at 3, though both may persisl as an occasional variation in the same testis-The anuran fiber at 3 is subdivided by the undulating membrane into a juxta-axonemal (minor) fiber and an axjal (major) fiber. In the hylid, litol/4 graciletlta, the minor fiber is particularly well developed, being unusually large relative to the major fiber. It is deduced that ),n Ascaphus, as a paedomorphic condition. lhe undulatjng membrane has been greatly reduced in length. and broadened, but that homologs of the juxta-axonemal and axial libers remain well developed. (From Jamieson et al. 1993.)

…

Figure J6. A marsupial spernatozoon exernplified by La go rc he ste s ,l i r,r'/ r& r. rufous hare wallaby. LS, showlDg lnarsupial feature ol compression of lhe nucleus in a plane at right angles to axoneme and subsequent rotatton of the nucleus so that il lies along the midpiece. (Frorn S Johnston, L Smith, F Carrick and BGM Jamieson, unpublished data.)

…

Figure l7 (dght). The eutherian spermatozoon exenplified by thc Macroscelidae. clephant shrews. All arc Mactosrclides proboscideus, except F, whjch is petrotlro,tlus tetra(t0ad\lus. A-Bt LS of acrosome and nucleus. C-D: proximal and more distal TS oI acrosome E: TS of nucleus, which is envelope.l by the acrosome for its entire length-Fr LS of e[tire midpiece. showing cytoplasmic droplet and helical anangenent of mitochondlia. with adjacenr nuclcus ind librous sheath of principal piece. G-H: LS of basc of nucleus and anteriol region of midpiece in planes approximately at right angles. I-J: TS of midpiece, showing axoneme with periphelal fibers. K: same' through cytoplasmic droplet L: LS of base of nucleus and antenor regron of midpiece, showing pariof cytoplasmic droplet. M: LS ofjunction of midpiece and principal piece, showing annulus ar the commencenenr of fhe tibious sheath. N: iS of fibrous sheath lurther distally. Ol TS ol principal piece where the peripheral tibers at 3 and ll have each been replaced by an inward keel fiom the longiludinal column oi the fibrous sheath P: TS of principal pieces ol which some, nore distal. have lost the peripher;l fibers. Q: Ts of two endpieces. with disrupted arangemeit ot' microtubules (odginal )

…

Figures - uploaded by Barrie G. M. Jamieson

Content may be subject to copyright.

Content uploaded by Barrie G. M. Jamieson

Content may be subject to copyright.

29 Sp e

rrn a t o z o a I Phy loge nY

oif the Vertebrata

Spermatozoa

of Sarcopterygii

Spermatozoal

Synapomorphies

of the

Tetrapoda

Spermatozoa

of the Lissamphibia

The Spermatozoa

of Gymnophiona

The Spermatozoa

of Anura

The Spermatozoa

of the Amniota

The Spermatozoa

of Aves

Spermatozoa

of the Squamata

Spermatozoa

of Mammalia

B.G.M.

Jamieson

University

of Queensland

304 / CneprBn29

he ultrastructure

of the spermatozoa

of fish, from

agnathans

to Dipnoi, and its phylogenetic sig-

nificat.rce

has been

reviewed

by Jamieson

(1991) and Mzittei (1991) and the evolution of tetrapod

sperm.

with particular reference

to amnlotes.

was

dls-

cussed br Jamieson

(lg95a) In this

chapter.

spermato-

zoal ultrastructure and phylogeny from the

Sarcopterygii (lobed-finned fish and their descendants)

will be examined

and an attempt

will be made

to deduce

the spermatozoal

synapomorphies

which distinguish

the

maior constltuent

groups.

Spermatozoa

of SarcoPterYgii

Extant sarcopterygians

consist

of the Actinistia (con-

taining a single

species,

the coelacanth,

Latimeria

chaluinae), the Dipnoi (containing the three lungfish

genera

Neoceratodus,

Protopterus

and

Lepidosiren)' and

ihe Tetrapoda

(containing the Lissamphibia and the-

Amniota). Altemative classifications

and phylogenies

these

three major groups are discussed

by Jamieson

(1991).

Despite

some

equivocal

results,

recent

molecular

studies

appear

to confirm the finding (for example,

Meyer,

Wilion 1990;

Meyer, Dolven 1992) that

lungfish-

es are

the sister

group of the Tetrapoda

though

the closest

relatives of Latimeria (Yokobori et al- 1994; Zardoya'

Meyer 1996.

l9q7a.

bl. All endorse

the sarcopterygian

stalts

of Latimeria demonstrated

by Hillis et al. (1991)'

The spermatozoa

of the sarcopterygian

fish have

been deslribed ultrastructurally for the actinistian

Latimeria chalumnae

(Mattei et al. 1988);

for the

Australian lungfish, Neoceratodus

foruterl (Jamieson

1995a;

Jespersen

1971) and

in this account

(Figs lA-R);

for Protopterus

(Boisson 1963;

Boisson et al'-

1967,

Purkerson et al. 1974); and for Lepidosiren (Matos,

Azevedo 1989); tetrapod sperm

literature is briefly

summarized

in Jamieson

(1995a)

and

in this chapter.

Symplesiomorphies

of Sarcopterygian

Sperm

.- Ai deduced

fiom the sperm

of sarcopterygian

fish

(Latimeria and the Dipnoi) and a survey of tetrapod

sperm,

the following features

appear

plesi.omorphic

for

the spermatozoa

of Sarcopterygn

Sarcopterygian

fish sperm

have a very long, slenderly

conical.acrosome

vesicle (Figs. lA-E) but lack the sub-

acrosomal

cone that is basic to tetrapods.

There is only

one endonuclear

canal in the coelacanth,

Latimeria

chalumnae,

but this contains two or three perforatoria

(Jamieson 1991, 1995a',

Mattei et al. 1988)'

In Neoceratodus

there are two or three (Figs. lC-F, H'

I) or sometimes

four perforatoria initially in one canal

but

posteriorly

tFig.

lHtin as

many

canals

there

are perforatoria

(Jamieson

1995a;

Jespersen

1911). Neoceratodas

is exceptional in sarcopterygians

in that the perforatoria re-emerge

from the nucleus-

their posterior ends

(Fig. 1I). The number

of endonucle-

ar canals and of enclosed

perforatoria is one in basal

Lissamphibia,

in the caiman (though

poorly substantiat-

ed by micrographs), tinamou, rhea and non-passerrnes

(for'exampG, galliforms), but in the Chelonia and

irocodylui iohistoni therc are two or three canals

and

in Sphinodon there are two. There are three

endonucle-

canals

in the sperm

of the

sturgeon.

Acipenser

sturio'

in the Actinopeiygii (ray-finned fish). It is therefore-

probable

that the presumed

common ancestor

Lissamphibia

and amniotes

possessed

more than one

per{oratorium and possibly more than one endonuclear

ianal. A single

canal

occurs

in the Lissamphibia,

except

where lost in more advanced

Anura (Jamieson et al'

1993).

and appears

basic

to all amniotes

above

turtles

and

Sphenodoi;

that is. in birds.

squamates

and mam-

mals lJamieson, Healy 1992). ln Acipenser the canals

are spiralled around each other as they are in turtles,

Sphinodon and Crocodylus

iohnstoni- The spiral

arrangement.

or at least

the

presence of one

or more

endoiuclear canals, may well be a synapomorphy for

the Osteichythes, a monophyletic clade including the

Actinopterygii, Sarcopterygii

and, within the latter,

ttre Tetiapodi. The canals

are absent

(presumed lost) in

the highlt simplified sperm

of holosteans

(a paraphylet-

ic group) and

Neopterygii (Jamieson 1991)

ihe nucleus

is long in the sarcopterygian

fish

(Neoceratodus,

Fig. 1A). This may be a plesiomorphic

retention from osteichthyan fish, as it is also long in

Acipenser, a basal actinoptery

gian, and in

Chondricthyes

(Jamieson

199

1, 1995a).

Further elonga-

tion

in sarcopterygian

fish appears

apomoryhic

A simple midpiece,

as in Neoceratodus

(Figs lG'

J-L), and in Acipenser,

with some of the mitochondria

in a cytoplasmic

collar, is presumably

plesiomorphic

for

the Sarcopterygii.

The location of a putatively

mito-

chondrial sleeve,

usually incomplete, lateral to the

ntrllclets

in L,otimeria is clearly apomorphic

A 9+2 axoneme is plesiomorphic for the

Sarcopterygii.

Whether the lateral fins (Figs. lM-O) are

a plesiomorphy held over from osteichthyan

ancestors,

and basal

lo Aclinopterygii.

or are a new.

homoplasic

development

is debatable

Spermatozoal

Synapomorphies

of the

Sarcopterygii

If one accepts

the validity of the Sarcopterygii

defined above,

at least

four synapomorphies

for the

group can be proposed on the basis

of sperm

ultrastruc-

iure. These

pertain to the perforatoria, the nucleus,

the

retronuclear

body and

the structure

of the flagellum

The great length of the nucleus in Actinistia (acro-

some and nucleus 25-26 pm long) and Dipnoi may be

___- I

an initial synapomorphy

of the Sarcopterygtt.

The nucleus

teaches

a length of 70 1tm

in Neocerolodtts

fbnrl?rl (Jespersen

1971),

the longest

recorded

in fish

sperm

(Jarnieson

1991).

The extension.

to the nucleus,

of rod-like

structules,

the perforatoria, is a new development

in the

Sarcopterygii

and thus constitutes

a synapomorphy,

and

an autapomorphy,

for the group. The portions of these

within the nucleus

lie in one or more endonuclear

canals.

Pelforatoria, or at least

endonuclear

canals

indi-

cating

the existence

of these,

are

present

in lampreys,

which an acrosomal

filament is extruded on reaction

(Afzelius et al. l9-57); the cladistian Polypterus

sene-

galzs. in which there is an axial endonuclear

canal

but a

perforatorium remains

to be identified

(Mattei 1970)l

the chondrostean

Acipenser stellotus,

in which an acro-

some

reaction

involving subacrosomal

material has been

demonstrated,

although

the role of the material

in the

three

endonuclear

canals

is uncertain

(Cherr,

Clark

1984r Detlaf, Ginzburg 1963); the dipnoan

N e o c e rat o

(lLt

s lbr,rreri

(Jamieson

995a;

Jespersen

1971):

urodele

amphibians;

prirnitive fiogs. including

Ast'tuthus

(Jamieson

l995al Jamieson

et al. 1993);

and

amniotes,

including

"reptiles"

and non-passerine

birds,

of which

the most

basal are

the Chelonia

(Furieri 1970;

Healy,

Jamieson

19921

Hess et al. 1991; Jamieson

1995a: Jamieson,

Healy 1992) and

the sphenodontidan

Sphenotl.on

putr(totLts

(Healy, Jamieson 1992. 1994:

Jamieson.

Healy 1992).

With the exception

of lampreys.

Acipenser and Pollpterus, therefbre in the sarcoptery-

gians, the pedbratorial rods extend anteriol to the nucle-

(Figs.

1C-F.

for Neoceratodus).

It is probable, in view

of the presence

of rods and endonuclear

canals

Lntimerio and the ceratodo ntifortn Neocerotodus, that

their absence

in the lepidosireniforms

Protopterlts

(Boisson

1963: Mattei 1970) and Lepidosiren

(Matos,

Azevedo

1989)

is secondary.

A large dense

body, between

the nucleus

and centri-

oles and telmed the retronuclear body, has been

described

for Protopterus eruleclens

by Boisson

(1963:

Fig. 2) and P. aethiopicus

by Purkerson

et al. (1974).

It has

been homologized

with a srnaller structure

which,

though

postmitochondrial, oliginates

behind the nucle-

u.s,

in Neoceratoda.s

by Jespersen

(1971)

and

Jarnieson

(1991;

Fig. lG); and a postnuclear

structure,

termed by

Mattei

et al. (1988)

the

"paracentriolar

body,"

Lotimeria

chalmnnae

by Jarnieson

(1991).

Its cross stri-

ation in P oetlliopicLrs

has led to its being compared

with the striated

columrs of mammalian spefm

(Purkerson

et al. 1974).

It is tentatively

considered

homologous

with the neck region of urodele and

anuran

sperm

(Jamieson

1991. 1995a).

being. in urodeles,

most

strongly

developed in ambystomatoids,

plethodontids

(Fig.

6C) and salamandroids

(Figs.

6D, E) (Baker

1962,

1963, 19661

Furieri 1962;

Jamieson 1995a;

Picheral

VpnrsenAre

Spenu

Puvt-ocsxv

/ 305

1967,1979;

Picheral

et al. 1966;

Werner

et al. 1972)

and

weakly developed

in clyptobranchs

(Baker 1963).

The retronuclear

body is hele considered

to be a

synapornorphy.

and autapomorphy.

of the Sarcopterygii.

In Neoceratothrs,

lateral

rin-like prolongations of the

sperm

flagellum

are

present at doublets

3 and 8

(Jamieson

l995ar Jespersen

1971),

(Figs.

1L-O,

Q),

except

at the

endpiece

(Fig. 1R). Shortly

behind

the dis-

tal centriole.

within the mitochondrial

collar'

(Fig. 1L)'

and

behind this

(Fig. lM), for a short

distance,

each fin

is supported

by a large

dense

juxta-axonemal

rod and

a smaller

lateral Iod within its free extremity (Jamieson

1995a).

The fin becomes

more extensive

behind this

short anterior

region but the rods are reduced

in size

(Figs. 1N, O) and soon only the lateral

fibel persists

(Figs.

1O.

Q). Such

lateral

prolongations, though

ques-

tionably with supporting

rods, in Latimeria were appro-

priately termed undulating

membranes

by Tuzet and

Millet (1959).

Lateral lins (also

at doublets

3 and

8) in

many actinopterygian

fish

(Jamieson

1991;Mattei

1988)

could conceivably

have been

pfecursory to actinistian

and

dipnoan

undulating

membranes

but homoplasy

cannot be ruled out as lateral axonemal

fins occur also

in some

echinoderms

and protostomes

(Jamieson

1995a).

Two bilateral

elements

which also occur

doublets

3 and 8 in Chondlichthyes

were presumably

convergently

acquired.

It is here accepted,

proposed

by Jamieson

(1995a).

that dipnoan

axonemal fins are homologous

with the undulating membrane of lissamphibian

sperm.

It is thus

proposed that

presence

o{'two undulating

membranes

is a sarcopterygian

synapomorphy.

As supporting

rods are not reported

for ProtopterLts

(Boisson

1963: Purkerson

et al. 1974) or Lepidosiren

(Matos,

Azevedo

1989). nol for Latimeria

(Mattei

et a]

1988), it is possible

that

presence of such rods

is a

ceratodontiform-amphibian

synapomoryhy.

Even if lateral axonemal

fins are sarcopterygian

symplesiomorphies

carried over from an osteichthyan

ancestor,

their elaboration in dipnoans and amphibians

is considered synapomorphic.

Spermatozoal

Synapomorphies

of the Tetrapoda

A generalized

tetrapod

spermatozoon

manifesting

the shared

features of basal lissamphibians

(fbr example,

Ast'ttph.us)

and basal

amniotes

(Chelonia,

Sphenodort)

illustrated

in Figure 3. As it shares features

of basal,

extant tetfapods,

delineation of this hypothetical

ancestral tetrapod

sperm

is not unduly speculative.

It should

be borne

in mind that ancestral

tetrapods,

being non-anniote, are classifiable

as Amphibia.

Their descendants

are the Lissamphibia

and the

Amniota.

306 / CHnPrsn29

l.Opm

ilffi"

The tetrapod sperm is derived relative to that of

sarcoplerygian

fish (Actinistia

and Dipnoi) in a remod-

eling of the acrosome complex and nucleus.

This involves

developmenl

of a subacrosomal cone and.

presumably to house this, correlated reshaping of the

proximal

end of the nucleus.

A cone of subacrosomal material, not seen in sar-

copterygian hsh, is developed,

in addition to the basal

sarcopterygian

perforatorial

rods. in the Lissamphibia in

Ascaphus

(Fig. a) and discoglossoids, and in the

amniotes in Chelonia, Sphenodon

(Fig. 11), crocodiles

(Fig. 12), squamates

(Fig. 15A) and monotremes.

Its function

is not known but it may be

perforatorial.

suggested by the fact that it is present

in the absence of

perforatorial rods in monotremes. Birds have lost the

subacrosomal cone but

plesiomorphically

retain

Derforatorial rod.

A second synapomorphy of the tetrapod spermato-

zoon relative to sarcopterygian fish is the development

of an abrupt shoulder-like transition from the anterior

tapered

portion of the nucleus (constituting the nuclear

rostrum) within the acrosome complex to the long cylin-

drical portion of the nucleus. The nuclear shoulders are

illustrated for Ascaphus (Fig.4), Sphenodon

(Fig. 11),

Crocodylus

johnstoni (Fig. 12) and the squamate

Carlia

rubrigularis (Fig. 15A). The shoulders

and associated

narrowing of the proiimal end of the nucleus as the

nuclear

rostrum, within the acrosome

cornplex,

presum-

ably were an adaptation allowing the newly evolving

acrosome cone to be housed between the nucleus and

the acrosome vesicle.

: Spermatozoa

of the Lissamphibia

The extant Amphibia comprise the subclass

Lissamphibia.

These consist of the Urodela (newts,

VnnreenArs Spenv

PHvI-ocnrv | 307

nuclear

Figve 2. Protopterus

anne ct e ns. Spematozoon,

showing retronucl€ar

body. (From Jamieson

1991, after Boi€son

1963.)

2 flagelh

salamanders and sirenians), the Anura (frogs and toads)

and the Gymnophiona (caecilians). It is argued

(Jamieson

1995a; Lee, Jamieson

1993) that internal fer-

tilization is basic (plesiomorphic) in the Lissamphibia.

It occurs in the great majority of urodeles and in all

gymnophionans,

but in only primitive frogs.

A survey

of the ultrastructure of the sperrnatozoa

the three orders of Lissamphibia permits delineation of

a generalized

lissamphibian spermatozoon (Jamieson

1995a; Fig. 5). Many features of this spematozoon are

deduced to be plesiomorphies

carried over from the pre-

sumed tetrapod

ancestral spermatozoon

as they are also

found in basal amniotes or even in Dipnoi and

Actinistia.

Spermatozoal Symplesiomorphi

es in the

Lissamphibia

Features of the ascaphid

{Fig. 4t

ancestral lissamphibian sDematozoonand hypothetical

(Fig. 5) that are

Abbrcviations used in figures: a, axoneme; af, axial fiber; an, annulus;

av, acrosome vesicle;

b, ba$; cc, cytoplasmic canal:

cd, cltoplasmic

drcplet; cy, cytoplasm; d3, density Cuxta-axonemal

fiber?) at 3: db, dense, inrermitochondrial

body; dc, distal cenbiole; el, electo; lucent

space; ec, endonuclear

canal;

| flagellum; fs, fibrous sheath;

h, head; hmt, helical microtubules;

jl juxta-axonemal

fiberi jf3, juxta-axonemal

fiber at 3: jf8, juxta-axonemal fiber at 8; lc, longitudinal column; lf, lateral fiber; m, mitochondrion; mp, midpiece; mts. sheath of micro-

tubules: n, nucleus; nf, basal nuclear fossa; nk, neck; ni, infolding of nucleus into neckpjece

(retronucliar bod-y):

nr, nuclear rostrum; nri,

nuclear

ridge; p, perforatorium; pa, paraxonemal

rod; pc, proximal centriole; pf, peripheral

fiber; r, retronuclear

body; si, subacrosomal

cone;

sdb, small dense body; stc, striated

(segmented)

column; su, subacrosomal

naterial; u, undulating

membrane.

Fl,gwe l. NeoceratoduJloruter, the Australian lungfish. Ultastructurc of tle spermatozoon.

A: Scanning electron

micrograph (SEM) of the

head and midpiece. B: SEM of the base of the nucleus and midpiece and the flagellum with its lateral fins or undulating membranes.

C, D: Longitudinal sections (LS) of the perforatoria, showing their extension anterior to the nucleus and, in D, in an endonuclear canal.

E: Transverse

sections

(TS) of the pedomtoria within the acrosome.

F: TS of perforatoria

entering

and within tlte nucleus.

G: LS of midpiece,

showing mitochondrial collar. H: TS of two perflratoria, each in a separate endonuclear canal. I: TS of two perforatoria posteriorly emirgent

from the nucleus. J: TS of midpiece tluough the distal centriole. K: TS of far anterior region of axonemi, within thi cytoplaimic canal.

L: Sam€ further distally, showing beginning of lateral and juxta-axonemal fibers on each side. M: Same shortly behindihe midpiece.

Nl Fufiher distally, showing more slender

undulating membranes

still with lateral and greatly reduced

juxta-axonemal

fibers. O: Still furtler

distally, the slender undulating membmnes

now lacking the juxta-axonemal fibe$. P: LS of basal nuilear fossa. e: Axoneme far distally,

shortly before the endpiece,

with greatly

reduced

undulating

membranes.

R: TS ofendpiece. (Original.)

308 / CnePren29

Figure 3- Generalized or hypothetical ancestral tetlapod

spermatozoon.

lissamphibian

symplesiomorphies

are.

as,follow-s'

An ant-erior

acrosomal

vesicle

forms

a hollow cone

i.^r."pl"rv gi"" svmpt

esi

omorphv'

it | ::"^tJ

t::-:

"on" oI ru-bu.rotomal

material:

lhe subacrosomal

cone

;;;;*t the tapered

end of the nucleus

(lis-

r^-pitiUl"" symplesiomorphy

and tetrapod

synapomor-

rigure

4 A:caphus

rrr?i

.Diagrams

of sp€rmar:i:f,tXT#i#::

:**i:Jl

l;H:l$*:i;::i:J#il13""

a,,qq3

nhv I which is

elongate

t lissamphibian

symplesiomorphy

#i'#;;l;;y;;isvnuPomo'PhY

j:ln:i::l.T il:

:il4$:n',T$;"',",t,1Ji,;;;;"is"(lissamphibian

.;;;;;i;;".phv and tetrapod

svnapomorphv)

posterior

to which its form is cylindrical. Axially, within

the acrosome and therefore extending anterior to the

nucleus

(lissamphibian

symplesiomorphy and sar-

copterygian synapomorphy),

there is a rod, the putative

perforatorium, which deeply penetrates

the nucleus

within an endonuclear canal

(lissamphibian,

tetrapod

and sarcopterygian symplesiomorphy). The basal sar-

copterygian feature of a prenuclear, and endonuclear,

axial rod (perforatorium), is present

not only in

Ascaphus but also in other primitive frogs, Discoglossus

and the bombinids Bombina and Alytes.It is also

present

in gymnophionids

though there the perforatorium

lodges

posteriorly in a much shorter endonuclear

canal.

The base of the nucleus

is indented as a basal nucle-

ar fossa (possibly a tetrapod symplesiomorphy as also

seen

in Neoceratodus, Fig. 1P) which, unlike the

dipnoan, contains the proximal centriole. Behind this is

the distal centriole, which forms the basal bodv of the

axoneme.

Spermatozoal Synapomorphies in the

Lissamphibia

Synapomorphies

of the lissarnphibian

sperm relative

to an ancestral

tetrapod may now be considered.

The presence

of an undulating membrane

within the

flagellar complex has long been thought distinctive of

the Lissamphibia

but it is proposed

here,

as in Jamieson

(1995a),

that it is the loss of the undulating membrane

adjacent to doublet 8 which is distinctive and synapo-

morphic of the Lissamphibia, with retention of that at

doublet 3. Evidence for the former existence

of an undu-

lating membrane at doublet 8, as in Actinistia and

Dipnoi, is the persistence

of the

juxta-axonemal

fiber at

doublet 8, in the absence

of a membrane

on that side, in

urodeles

(Figs.

6N, O) and exceptionally

in Anura.

Concomitant with the loss of one undulatins mem-

brane has been the developmenr

of a condition ii which

the axoneme undulates

around the remainine lateral

llber which

has. therefore. long

been termed

ihe axial

fiber. This condition is particularly well demonstrated

by the Urodela, particularly in the non-cryptobranchs

(Figs.6N, O) as these

have a stiffened

axial fiber.

An axial fiber with undulating membrane is present

caecilians

(Fig. 6I), as basically in anurans

(Figs.

5, 6P).

Sperriratozoal

Synapomorphies in th€ Urodela

Ultrastructural aspects of urodele spermatozoon

are

illustrated for the salamandri d, Taricha granulosa (Figs.

6D, E, O) and the plethodontids

Stereochilus

marginatus

(Fie. 6C) and.

Eurycea

quadrldlgitara (Figs. 6M, N).

The structure

of urodele sperm is uniform relative to

the great diversity in anurans though the ground plan

has many similarities with that of anuran and caecilian

sperm. The more striking, synapomorphic,

urodele

VERTEBRATE

SPERM

PHYLocENv / 309

Figure 5. Diagram-

matic representation of

the hypothetical ple-

siomorphic lissam-

phibian spermatozoon.

(From Jamieson 1995.)

Xr@

flJ-6

310 / Cu,qPrBn

sperm

reatures

will

now

*i,H:"1;;iJ*ar

rissam-

*lT'ili:;i:T"*;;:}:"."J'ffJ.:'iJfiti*"ff.1:X

;t''+*'"":'i;::H;'if

::"{i:Hilil-pi"tr'oaontia'

"'r:;i*.'-*lU";r::i,li:';"::':?tlTi:,:lJi:

o,i,ii

q',;"a,iiisit,,ta..^g

i?,Jil1ii:i',"#:iii:

H',,HT:ll::i*lru,Ttffll,n**:ru:i"

#::,1':"tiiltrhi*ilihffi

ffi

Hx*fl

"\:"x'i'Tirfu"',*:t"l,Tf

riLt"i:::J'i:ht*lru:

[:Hr

3l:'f

in.Jtt'ffi"':r;:;;i.'r",

t#:i:: .,| Hil,'

;; i; ;;v s'oma'idae

u"k"own

:i"HtJi""J;1T#,i,':"::Tiil#;trrlru'J;r;

3i'::":''i}rilil'qt:':i:$jll]itr.t"ilfr,Jj'i1]

#'1,'"*-till*ru*l,*i*+ir**+iH'h'l:i-*tn*$',tnt1i:rff

:;.t.ffi

,;:

nal barb

to the acrosome.or.

M.c,eso,

8ae

uas,.,,",,i1iiliTllfiitl

:ilf*:Tliilli;l,jtT.i:IJil.T-Jfr

i:;,I[l:X

fl?,t""rTT#;ilili".tl$;;;;dobra,tchus

striat,rs

.."';;:.i', lj*.*;#i"i"" neck

saramandrids

fu*"*1";ixxi;i:,ii,l'"-$i$1i:lr::i:ili:- :::.'';,:,

r*:tii'#;'#i,xru'.",run:'.;*'"*

-,r,i""'.'"'.i:,!lTtf*

ir;H'[:1"#'ii"fi

:qir*,*g;;*

q+tg*lills*:p*U*

ii'*.i;,lt:;:i{ffil;"#,:i,,i"":'"liJ:ffi:','',T:

lft'&ffiiJt considered

primir

ii::{;i#i!{:illgl';in.';ii:l;*igf,f

i"TlJ"ifr

:T;;:"}l.,"';11,"'-*it*lTnff

"lT:isfJi:

id;il;i;;#'',l","*o*,Ttt:iilt"'l:i:1:{l[rii:,'filli'{.il;]

j.fi]jit:,ii"##j1;;iff

n",ffirTl+'l'ii"':.'i:iiiL'J:Tlf#ihT"'i:l

**r**',**,r,ixi:i.rlri;:"i::#il:1*,;i:i:;

"t"ttiiittY; rtuge

(as

in the

salanandrid

rttrichn

g.ran- the

tail'

l;:t,t# ;lilJ""T"*i'"''-*ilJ':ii:'.';:1i'si:':;

r,**,6 r"y*r,.sr:r1:;::Tj:T'*c'ure

in Lissamphibia

Hfff.:1"Jr,i:i"#:i,',",r'?i.','"":"i"Jili

ili.fu ;i$i:$iiyrffitjJ.'+i.:"fiiffi*il:jii?

#l[L::]'"---*t;l:;*;J*i:;"1*:$fi

'::i*.'';".;;*'..1'"6ff

i1l*r*ir::'fffi

l11

an autapomolphy

of urodeles nu

- e"hina

the'nucreus,

..:.t[,:r"ilfr]:;:J.:?:

:;"i"';,.srti;;*1,:1",:,Tl:ti:#*{ql"1ti.'-:{i95|,,lil

;'"1T"..JTill'J;t:,1:i::i:;;i#,=i."".

uoov

fil,r,i-r,"iili**+tli*t

'i*

l*{;ili,t

*$ll'tiilgfl

;il+lt?h;;$,'g;;

i i nI

?r!:r

;$, ;;

sr$li'ffi

il;il"h

;. tr'"

a"",

ro.,i

iltl;,n*n:;*':

L:X;.1.]i*i'.iiqllifrli'J:#iliii:r:*qq

ll*m**#J,'fi:J:ilT::iii1ili::"{':.ti ia;,',,"

rrogc,,,to,ana

atb,,,,,,*,,

ii.it*"""i""

^a,r'i'r

il;i;il;;;i"p*

(Pi.h"*r

il;;,

ieis;

Pi.'''"'ut-"t

ot +fJf;i

:t,f$ffi

iiff*filn"t *t,"',.lj:""..,,11:

iiil"*l;;:*.fi:"33i"1',1?i.'lli'ii'"F:ft1#P! Hifr*hlxliri"ffi'#'i'i:,;,:i:;:''i;,..,,,,.r

rs.rnuc,eussh.w.

I';:tf

;1.:'';':'lJT'T',',ffiThT',:Jff

:ff#f-*lq+Ul+$;W:ffi.l1{;p;"ffi

,;rx;#

'"","Jh"

"q!v,,eyt"'-t::,::*l"xT:h,::ffi,:flH

,'}:t+Jt;!T*ifi:$*l{*llm*"r1il:Ji,i""'"'

nlarsinatus,

Fig. 6C) show

VERTEBRATE

Spr,nu Pnyt-ocnNv | 3ll

0.25pml

:,,.1

_- 5.:J-,

Pj'

312 I Cudprqp.29

At least some

urodele sperm

have a dense

ringJike

structure

questionably identified with an annulus

(Picheral 1919).

The innel wall of a cytoplasmic collar.

equivalent

to that

in many

fish sperm

(Jamieson

1991),

is occupied

by tuo lalge denre

structures:

the

ring.

forming a lining to much of the canal. and opposite

the

ring, and in Amphitnta contiguous

with it (Barker,

Biesele 1967),

the axial fiber. It is uncertain

whether

the

strong development

of the ring i.s

a general urodele

synapomorphy

or is restricted

to higher

urodeles.

In urodele

sperm,

the axial fiber, at doublet 3, is

connected

to the axoneme

by the undulating membrane

but, typically (Taricha gronulosa, Fig 60). there is no

intervening

juxta-axonemal fiber. This may represent

synapomorphic

loss from a basic

lissamphibian

condi-

tion with juxta-axonemal

fibers at 3 and 8. However,

the

condition

in cryptobranchs

requires

investigation.

The plethodontoid

arnbystomatoids

(for example.

Plethodon albagula, Fig. 7 and Ettrycea quaclridigitala.

Fig. 6N) are exceptional

in having a density on the

adaxonemal

end of the undulating membrane but it is

questionable, because

of its connection

to dense

bodies

near the major fiber rather than to the fiber' (Figs. 6N,

7). that

this is homologous

with the anuran

juxta-axone-

mal

fiber

(Jamieson

1995a).

The axial fiber has the presumed

plesiomorphic sub-

circular ctoss section

in the cryptobranch

H,,-noblus

neb-

a/osas

(Picheral

1979),

as in the Gymnophiona

(Fig.

6I)

and Anura (Fig. 6P). though in cryptobranchs

differing

frorn the typical condition in the latter two orders in also

having the smaller fiber on the opposite

side of the

axoneme

at doublet 8.

In the higher urodeles,

as an internal synapomorphy,

the axial fiber acquires,

for much of its length,

Y-shaped

or trifoliate cross section (Toricha granulosa,

Fig. 60). This sectional

profile corresponds

with that

used

human tool5

itnd engineering

structures

to confer

strength

and rigidity and is deduced to stiffen the fiber

against

bending.

The trifoliate cross section.

and the

accornpaniment

of a fiber on the opposite side of the

axoneme,

has been demonstrated

ultrastructurally

salamandrids

(Baker

19661

Furieri 1960,

1962: Picheral

1961

, 1972.

1979; Taricha

granulttsrt. Fig. 6M); in

amphiumids

(Baker 1962);

in ambystomatids

(Russell

al. 1981)

and the plethodontids Eurycea sp.. Ettr,-cea

bislineata

bislineata,

E. qLtadridigittttct, E. u,ilderae.

Gyrinopltiltts

porphyritictts, Plethodon

albagttla,

Stereochilus

marginatus. and Typhlomolge

rathbttni

(all in this study).

Presence

of an additional

juxta-axonemal

fiber in

many anurans

at doublet 3, connected

with axial fiber by

a thin lamina within the undulating

membtane,

may

well be plesiomorphic,

and attributable

to the

generalized lissamphibian

spetm.

In some ambystomatid

salamanders

(Antb\stoma

and,

Rhyctcosiredon)

a short tail membrane.

absent

in the

ambystomatid

Rhyacotrilon

and unknown

in other

lissamphrbians,

has

been observed

by light microscopy

on the opposite

side

of the major fibel from the undulat-

ing membrane

at the posterior end of, and extending

longitudinally for a fiaction of the length of, the mem-

brane (Brandon

et al. 1974, Martan, Wortham 1972;

Wortham

et al. 1982). It does not appear

to represent

pairing of the undulating membrane,

on opposite sides

of the axoneme,

which is hele hypothesized

as an

ancestral,

possibly

ple-lissamphibian stage.

Sperm of the sirens,

exemplified

by Siren

intenn.edia,

S. lacertina and PseudobranthLts

striatus

axanthus, are biflagellate and have two interconnected

undulating

membranes

(Austin,

Baker 1964) Such

sec-

ondar-1

duplication

of the lronetne

also

occurs

in some

Anuta.

The condition

ol the milochondria

in Neoceratodtrs

forsteri in which the mitochondria are located

in a short

midpiece and in a collar-like posterior extension

of this

alound the base of the axoneme

(Figs. 1G, K, L) may

represent

the plesiomorphic condition for lissarnphibia.

In the cryptobranch urodeles Cryplobranchtts

alle-

ganiensis bishopi (Baker 1963)

and Hynobius nelnlostts

(Picheral

1979),

the mitochondria

seem to be located

a protoplasmic bead around

the nucleus,

even

in the

mature

sperm.

This possibly represents

a reduction

from

the plesiomorphic condition.

The cryptobranch

sperm

tail lacks mitochondria and thele is no rernnant of the

ring (putative

annulus) around

the major fiber.

In salamandrids

(Fig. 60). ambystomatids,

plethod-

ontids (Figs. 6N. 7) and amphiurnids,

small, ovoid mito-

chondrir

ale

presenl in cltopll'm around

a long anterior

region

of the

lr.ial

tmajorr

fiber

rahere.

in cross

section

of the sperm. they form an arc, the whole constitutrng

the intermediate

piece

or midpiece

(Baker

1966;

Fawcett

1970: Grass6

1986:

Jamieson

et al. 1993;

Picheral 1967, 1979; Wortham et al. 1917). This appears

to be a derived

condition.

Thi\ llpe of midpiece

hrs

been said

not lo occur

anurans

but, clearly

homoplasically,

in Li.mnodynasles

peronii an incomplete

ring of mitochondria

surrounds

the major fiber'(Fig. 8) rnuch as

in salamandrids

(Lee et

al. lqg2):

a \omewhat simillr

arrangemenl

is seen

Bombinu

rariegata

(Folliot 1979: Furieri 1975b;

Pugin-

Rios 1980) and Neobatrachtts

pelobatoides

(Lee,

Jamieson

1992)

in which the mitochondria

lie on each

side of the axial fiber.

A major difference of salamandrids,

ambystomatids

and

plethodontids from anurans is the great

length of the

mitochondrial region. occupying a considerable

propor-

tion of the length of the tail. This appears

to be an apo-

morphic

elongation

as the midpiece

is short

in dipnoans

(Figs.

18, G) as it is in basal actinopterygians

such as

Ac io e

ns e r (J

amieson

199 1

Ju a-axonemal

fibre at doublet 8 Density doublfully homologous

wilh anuran

juxla-axonemal

fibre

al doublet 3

Axial fiber

(majorfibl

VERTEBRATE Spnnrr,r

Psvlocsxv I 313

Figure 7. Comparison ol the paraxonemal

rod

in Ascaphus with the accessory axonemal libers

and undulating membrane

of r urodele and a

hylid fiog, as seen in transverse sections of the

flagel1um. Only about

one

half of the length of

the undulating

membrane is shown in the latter

two species. In the urodele, the major fiber is

much larger than illdicated relative to the minor

t'iber. Presence of a fiber on each side of the

axoneme, at doublets 3 and 8, in urodeles is

probably the plesionorphic lissamphibian (and

sarcopterygian)

condition- The juxta axonemal

fiber at 3 is usually absent in urodeles and the

density in this position in Plethodon may not be

homologous

witb this; the axial tiber is hypertro-

phied. The two fibers on each side of the

axonen]e are reduced in Alura to one tiber, at 3,

though both may persisl as an occasional varia-

tion in the same testis- The anuran fiber at 3 is

subdivided by the undulating membrane into a

juxta-axonemal

(minor) fiber and an axjal

(major) fiber. In the hylid, litol/4 graciletlta, the

minor fiber is particularly well developed,

being

unusually large relative to the major fiber. It is

deduced that ),n Ascaphus, as a paedomorphic

condition. lhe undulatjng

membrane has been

greatly reduced in length.

and broadened, but that

homologs of the juxta-axonemal and axial libers

remain well developed.

(From Jamieson et al.

1993.)

Smalldense

Undulaling membrane

mitochondria

Pulative homologue

of axialfibre

mid{ilament'

The Spermatozoa of Gymnophiona

Caecilian spermatozoa have

been examined ultra-

structulally in Typhlonectes

naran.r

(Typhlonectidae)

(van

der Horst 1991; van der Horst et al. 19911

present

study) and by light microscopy fbr other caecilians

(Seshachar'

1939, 1940,

1942, 1943, 1945; Wake 1994).

The spermatozoon has all the components

of the hypo-

thetical lissamphibian

sperm which was delineated

above (Fig. 5), but there are differences

relative to this

in the perforatorium

and endonuclear

canal, the mid-

piece

and the undulating

membrane. At least the first

two are apomorphic.

Van

der Horst et al.

(1991)

considered

the curved tip

of the acrosome in Tlphlonectes

nat(ms to be compara-

ble with the hooked

shape characteristic

of urodeles.

However, in urodeles

the tip is strongly reflexed and the

supposed similarity with gymnophionans

cannot be con-

sidered a synapomorphy. Furthermore,

the tips of the

acrosomes in the oviparous lch.thltophis glutinosLts.

Uraeotyphlus

narayani, Siphonops annulatus and

Gegenopltis carnosus are spatulate,

not hooked

(Seshachar

1940.

1945).

Synapomorphies

of Gymnophionan Sperm

The base of the acrosome of the gymnophionan

Ttpl onectes

t?atans

conforms with the tip of the nucle-

us, which is indented medianly for a short distance as an

anterior nuclear fossa, into which the base of the perfo-

ratorium slightly protrudes.

The fossa is here regarded

as a short endonuclear

canal. The greater part of the

width of the tip of the nucleus forms a straight sided

bowl-like indentation closely titting the posterior end of

the acrosome and the short endonuclear

canal Denetrates

lhe cenler

of this indentation

{Jamieson. unpublished

data; van der Horst et al. 1991). A similar "acrosome

seat," where the lance-shaped

plug given off at the base

of the acrosome fits snugly into a deep

pit in the anterior

end of the nucleus.

has been described for Uraetyphlus

narayani and lchthtophis (Seshachar

1945)

and is

here considered

a significant

synapomorphy and

autapomorphy

of the Gymnophiona,

although the

perforatorium

and endonuclear canal are themselves

symplesiomorphies.

The centrioles

and the anterior part of the

gymnophionan sperm tail are surrounded

by 35-40

spherical mitochondria, which have an extensive array

of delicate

cristae

(van der Horst et al. 1991).

314 / CHapren29

This cylindrical arrangement

of mitochondria

around

the centriolar-axonemal

axis, in the absence

of a collar'

is possibly aponrorphic

relative to that of finned

sarcopterygians

and ancestral

tetrapods.

The major part of the tail consists

of an axial fiber

and a 9+2 axoneme

enclosed

by a plasma membrane'

plesiomorphic

features.

However' the plasma

membranes

on the two surfaces

of the undulating mem-

brane

are not closely apposed

but are separated

by a

considerable

amount of cytoplasm

(T\plllonectes

natans,

Fig.6I). It is uncertain

whether

the wide cyto-

plasmic band.

resulting in a thick undulating

membrane'

is considered

as an apomorphic

condition. It is presum-

ably so as.

in theil widest

region,

the undulating

mem-

branes

of Dipnoi are slender

(Fig. lO), and

thickening

appears

to have evolved sevetal

times in anurans

(ihough unknown

in urodeles).

in C1'clorantt

(Fig. 6J). Nevertheless,

the Tvphl oneoe

s characteristics

1Fig.

6I; are

not dissimilal

to those

01'some

regions of

the Neocerrttodus

undulating membranes

(Fig. 1O) and

could conceivably

be plesiomor-phic.

The unusual

and

presumably

apomorphic

structure

of the undulatjng

membrane

in Ascaphus makes

detelmination of the

plesion'rorphic

condition fol Anura difficult.

So far as

ultrastructural

study

of T\phlonectes

nalons indicttes, the only fiber remaining in the

gymnophionan

undulating memblane is the non-trifbli-

ate axial fiber (Fig. 6I). Juxta-axonetlal

fibers are

absent.

This appears

to be an apomorphy,

although.

again,

there

is considerable

resemblance

in this to the

more posterior regions of the undulating membranes

Neocerdtodus,

The Spermatozoa

of Anura

Space

limitations permit only a brief survey of anu-

run sp"r-. A list of the large literature on spermatozoal

ultrastructure

is available

in Lee and

Jamieson

(1992).

The sperm

of the basal

lamilies of the Anura

(Ascaphidae, Discoglossidae,

Bombinidae)

have

the fea-

tures

described

above

for the hypothetical

lissamphibian

sperm,

with certain

exceptions

which are considered

apomorphic.

Apomorphies

of anuran

sperm

will here

Figure S. Highly diagmmmatic

representation

of tbe ullrastructure

of a generalized

linnodynastine spermatozoon-

The conical

pedbrtt(l

riun, consisting

of bunclles

ol longitudjnal

fibrils. is a butbnoid

autapomo.phy.

Sperm of the tamily Myobatrachidae

differ lionr those

of oiher butbnoids i11

a probable symplesiomorphy' locatior ol the

mitochondrja

bordering

the axial fiber and not' as in eubufonoids.

in a

collar ar'ound

the anterior axoneme.

Exlension of lhe axial fiber into

the nuclear

fbssa characte zing myobatrachids

is also

known. though

lcss prominent, in one bulbr1id.

The unusually

discrete

perjcentriolar

material

may be diagnostic

of Dtyobatrlchids

while the periaxial

sheatb

around

the

prcxinal region of the axial tibcr may be distinctive

o[ limnodymstines.

(From Lee. Jamjeson

1992.)

briefly considered

within the context of the various

basal

and more derived

families.

The Anura ale

definable

spermatologically

on a sin-

gle, negative

autar.pomorphy,

which is the loss

of a longi-

tudinal

fiber adjacent

to axonemal

doublets

7.8 and

(juxta-axonemal fiber at 8), a fiber which exists

urodeles.

The fiber is, however, very rarely retained in

anurans,

as in Bufo morinus

(Swan

et al. 1980), and

transitory

fiber is present

in this position in the sper-

matid of Dr.icoSlossus

pictus

(Pugin-Rios

19801 Sandoz

1974a). This autapomorphy

might. however,

be better

described

as retention of the juxta-axonemal fiber at 3

only, as

gymnophionans

lack both

juxta-axonemal

fibers.

It thus appears that anurans

are plesiomorphic rela-

tive to gymnophionans, which have no juxta-axonemal

fibers (if, indeed, this is secondary), and to urodeles,

which have a well developed

fiber at 8, in retaining the

fiber at doublet 3 while being apomorphic in nonnally

losing the fiber at 8.

Major Tiends in Spermatozoa

of Anura

Ascaphidoe. The sole living membel of the

Ascaphidae, Ascaphus truei, is the only anuran with mor-

phological adaptation for internal fertilization.

This takes the form of a prominent taillike intromittent

organ. In all other

cases of intemal fertilization in anurans

(for example,

Nettophrynoides), there are no intromittent

organs

nor other morphological specializations

for

internal

fertilization, and this occurs simply by cloacal

apposition during prolonged amplexus

(Townsend

et al.

1981).

The Ascaphidae are here retained,

with the

Discoglossidae and Bombinidae,

in the Discoglossoidea,

although a separate

superfamilial status appears

justified.

The sperm of Ascaphus truei has

previously

been

very briefly described fiom light microscopy observa

tion (Metter 1964). Its ultrastructure

has

been described

by James

(1970)

and

Jamieson

et al.

(1993;

Fig. 4).

That ultrastructure accords with that of the basic

lissamphibian sperm,

notably in retaining the basic sar-

copterygian

synapomorphies of long perforatorial

rod

which extends anterior to the nucleus but also

penetrates

a long endonuclear

canal, the tetrapod synapornolphy of

nuclear shoulders,

and the lissamphibian autapomorphy

of retention of only one undulating membrane.

However,

the undulating membrane is of a form, consti-

tuting what has been termed a paraxonemal

rod

(Jamieson

et al. 1993), which differs tiom that shared by

the Urodela, Gymnophiona

and Anura and must be con-

sidered

autapomorphic. Putative apomorphies of the

Ascaphus sperm are as follows.

The paraxonemal

.rod in Ascaphus

sperm is consid-

ered equivalent to the major and minor fibers (axial fiber

and

juxta-axonemal

fiber at 3) in the sperm

tail of other

anurans

and these structures appeal to be jointly homol-

ogous with a coarse

(peripheral)

fiber of the amniote

axoneme

(Jamieson

et al. 1993). The undulating mem-

brane is merely a short, broad bridge between the inner

and outer

regions of the paraxonemal

rod (Fig.4).

The rod carries a longitudinal

groove,

along this bridge,

which houses the mitochondria. This leduction of the

Ventnenars SPERM

PHYLocENv / 315

undulating membrane

cannot be ascribed simply to

occurrence

of internal fertilization

in Ascaphrs and is

tentatively

interpreted

(Jamieson

et al. 1993)

as sperma-

tozoal

paedomorphism,

an extensive undulating

membrane

persists in internally fertilizing urodeles and

in caecilians.

It somewhat

resembles the developing

undulating

membrane complex in the urodele spermatid

(Barker,

Biesele

1967). In Fig. 7, a cross section of the

axoneme of a urodele, Plethotktn ulbngula, and of an

anvan, Litoria gracilenta.

both of which have a well

developed

undulating membrane, is compared

with that

of Ascaphus,

with its reduced membrane.Restriction

mitochondria to a groove of the paraxonemal

rod (Figs.

4, 7) also appeals to be an apomorphy

of Ascaphus.

A fine centlal

"rnid-filament"

(James

1970) which lies

between

the central singlets

though displaced slightly

towards doublet number 1 (Figs.

4, 7) may be an

ascaphid apomorphy,

though of unknown

function.

Discoglossidae-Aponorphies.

The spermatozoon

of the

painted

frog, Discoglossnr

plcras

(Favard

1955a,

b; Furjeri 1975a; Pugin-Rios 1980;

Sandoz

1970a, b,

19'13, 197 4a, b). is the longest known in the Amphibia,

presumably

apomorphically,

measuring

2,300 pm to

2,500

pm. The nucleus,700-800

pm long (Sandoz

1974b),

is penetrated

for almost its whole length by a

narrow

endonuclear canal: the perfbratorium

traverses

the canal

in the late spermatid

(Sandoz

1970b) but is

restricted

to the prenuclear

subacrosomal space

in the

mature spermatozoon

(Pugin-Rios

1980).

The il.rplantation fossa

is almost

entirely filled by

the anterior end of the axial fiber (Pugin-Rios 1980), a

putatively apomorphic characteristic which also occurs,

independently. in the Myobatrachidae. The short undu-

lating membrane

has a thick internal lamina connecting

the axial fiber to the juxta-axonemal fiber at 3 (Pugin-

Rios 1980). As in many anurans,

the mitochondria

around the base of the nucleus have disappeared by

maturity

(Pugin-Rios

1980).

Bontbinidae. The spermatozoa of Alytes

obstetriccols, Bombino variegata and Bombina bombina

(=Borubinatr.tr

lgnerr.s; Retzins 1906), described by

Furieri

(1975b),

Pugin-Rios

(1980)

and Retzius

(1906),

retain the plesiomorphic

condition of penetration

of the

nucleus by an endonuclear

canal containing the perfora-

torial rod. The undulating membrane is also plesiomor-

phically slender.

A synapomorphy

of these three bombinids is apical

truncation of the perforatorium (and

of the acrosome).

Bornbina bombina

(Retzius

1906) and B. t,ariegata

(Furieri

1975b; Pugin-Rios 1980) are apomorphic in

insertion of the axoneme at the anterior end of the

nucleus.

Ptpoidea-Pipidae. In the Pipidae, only the sperm of

Xenoptrs luevis has been examined ultrastructurally

(Bernardini

et al. 1986. 1988, 1990: Furieri 1972;

James

316 / CHePten29

Figure 9. Highly diagrammatic representation ol a generalized

pelodryadid (Australian hylid)-bufonid spermatozoon The conjcal

perlbratorium is diagnostic of the Bufonoidea. The mitochondrial

collar distinguishes

eubufonoids,

inclndlrg, inter alia, bufonids and

hylids, from the Myobatrachidae (From Lee, Jamieson

1993

)

1970;

Pugin-Rios

1980;

Reed

et al. 19721van

der

Horst

1979; Yoshizaki 1987).

The spermatozoon

is highly

apomorphic.

' A nuclear

fossa

is absent.

Mitochondria surround

the posterior

region of the nucleus.

There is a long sim-

o1e

flasellum. The undulating membrane,

axial and

juxta-aionemal fibers have

been

lost.

The Spermatozoa

of the Anura-

Neobatrachia

Whereas

so-called

Archaeobatrachia,

constituted

the previous families, clearly comprise a paraphyletic

assanblage,

the suborder

Neobatrachia

appears

on mor-

phological grounds

to be a monophyletic group'

ihir ttut bein supported

in parsimony analysis

somatic

morphology (Hillis 1991).

The Neobatrachia

contain the diverse and numerous

groups of what may

be termed the modern frogs. For brevity the families

investigated

for sperm

ultrastructure

will be treated

together

under

their respective

superfamilies.

Bufonoidea-Myobatrachidae, Leptodactylidae,

Bufonidae, HYlidae, Pelodryadidae

Most of the advanced

New World frogs and some

frogs in the Old World belong to superfamily

Bufonoidea. There appear to be no somatlc apomor-

phies

corroborating

the monophyly of the bufonoid neo-

batrachians-indeed, phylogenies

based

on general

morphology have suggested

that the group ls para-

phyletic. However,

spermatozoal

evidence

suggests

that

ihe bufonoids are monophyletic, providing a single but

convincing

synapomorPhY.

Bufonoids are united by development

of a conical

oerforatorium (Lee, Jamieson 1992, 1993''

Limnodynastes

peronii,Fig.68; also Figs.

8,9) and

loss

of the axial, rodlike perforatorium lt consists

of sepa-

rate sheaves

or fibers and occupies

the location of the

subacrosomal

cone of plesiomorphic lissamphibian

and

amniote spermatozoa.

It has been argued (Jamieson.et

al. 1993)

that the conical layer of subacrosomal

material

between the acrosome

and the nucleus in urodelils and

ascaohids

(Jamieson

et al. lg93l

and

in many

amnioles

including the Chelonia (Healy, Jamieson 1992' 1994''

Jamieson,

Healy 1992) does not appear

to be homolo-

gous

with the conical

perforatorium

of bufonoids

because

the two structures

differ ultrastructurally and

because

a conical perforatorium does

not exist in anuran

lineases which are more basal than the bufonoids'

How&er. with loss of the central perforatorial rod, it

remains

possible that the original subacrosomal

cone

became

modified as

the bufonoid conical perforatorium'

Within the

bufonoids.

sperm

ulmasructure

strongly

supports the separation of the myobatrachids

(Auitralasian "leptodactylids") from the hylid-bufonid-

ffi

il:iilln,,H1f.lilfl.''',"

New World leptodactylid assemblage

(termed

the true or

eubufonoids) as sister

groups (Lee, Jamieson

1992).

This phylogenetic

and taxonomic arrangement has been

repeatedly

proposed

in the past.

A generalized

myobatrachid

(lymnodynastine)

sper-

matozoon is represented diagrammatically in Figure 8.

Myobatrachidae-Synapomorphy

The myobatrachids do not have the mitochondrial

collar which is distinctive of eubufonoids, but are united

by their own synapomorphy, the extension of the axial

fiber up the centriolar fossa (Lee, Jamieson 1992; Figs.

6L, 8; and Limnodynastes convexiusculus, Fig. 6L),

though this characteristic

is approached in the bufonid

Nectophrynoides and is present

homoplasically

Discoglossus. Penetration in limnodynastines may be a

distinctive apomorphic

reversal from non-penetration in

lineages intervening between them and the origin of

discoglossids. The myobatrachids

appear to be the sister

group of the eubufonoids.

Eubufonoidae-Synapomorphy

The families Bufonidae, Leptodactylidae

(s. strict.),

and Hylidae (including Australian hylids, the

Pelodryadidae) are united, and separated from myoba-

trachids, by a single synapomorphy: a thick collarlike

cytoplasmic sheath

(Fig. 9, and the hylid Zitoria rheoco-

/a, Fig. 6K), that emanates from the centriolar region, is

separated from the flagellum by a cytoplasmic canal,

and contains the mitochondria. This synapomorphy led

to erection of the superfamily Eubufonoidea

by Lee and

Jamieson

(1993). A collar is a widespread characteristic

of spermatozoa

in fish groups, including Dipnoi, and

may have been lost in lower anurans to be regained

the eubufonoids.

Eubufonoids have the full comolement of conical

acrosome

tthough

with a conical rath;r rhan rodlike

per-

foratorium, as in myobatrachids),

elongate nucleus,

and

undulating membrane with axial fiber, juxta-axonemal

fiber at 3, and rarely (as

in Bufo marinus) at 8.

Some hylids show reduction of the undulating mem-

brane. This forms a thick lamina with no discrete

juxta-

axonemal

fibers and ends laterally with the major fiber

in the Australian Cyclorana alboguttata (Fig. 6J), form-

ing a secondary

paraxonemal

rod. Nevertheless

this

species exemplifies the bufonoid conical perforatorium

and the eubufonoid mitochondrial collar. The form of

the undulating

membrane confirmed (Meyer et al. 1991)

that it should

be placed

in Cyclorana and not in Litoria,

in which it is commonly included.

Litoria has a well developed,

thin undulating mem-

brane with a hypermorphosed

juxta-axonemal fiber

(Lee, Jamieson 1993;

Figs 6P,

9; and Litoria eucnemis,

Fig. 6P). In Hyla meridonalis there is further reduction

and the axial fiber directly parallels the axoneme

VnnresnAre SpBnM PHvr-ocsNv | 317

(Pugin-Rios 1980), as is also seen in Hyla japonica

(Kwon,

Lee 1995).

Spermatozoa of the Microhylidae have not previous-

1y been investigated ultrastructurally. That of

Cophixalus

inornatus (Fig. 6F) is apomorphic

in having

lost the undulating membrane and associated fibers, giv-

ing a secondarily simple flagellum (BGM Jamieson,

DM Scheltinga and KR McDonald, unpublished

results). This correlates with tenestrial reproduction in

microhylids. However, loss of the undulating membrane

also occurs in the lentic Ranidae.

Ranoidea

Ranoid families examined for sperm ultrastructure

are the Ranidae, in Mo (1985),

Poirier and Spink

(1971), Serra and Vincente (1960),

Yoshizaki (i987)

and Zirkin (1971);

and Rhacophoridae

(Mainoya

1981;

Wilson et al. 1991). A full listing is provided in Lee and

Jamieson

(1992). The sperm of both families are highly

apomorphic.

The sperm of ranids are much modified and simpli-

fied, most notably in losing the undulating membrane

and axial fiber. The acrosome is caplike, sometimes

asymmetrically, and apomorphically lacks a rodlike or

conical perforatorium. The mitochondria form a

manchette

surrounding the base of the nucleus and

a considerable region of the axoneme.

The sperm of the Japanese rhacophorid species

Buergeriu

buergeri, which lays eggs in streams, has a

long head and thin tail, details of which are unknown

(Fukuyama

et al. 1993). In contrast, the foam-nesting

spectes

Chiromantis x.erampelina (Mainoya 198 1

;

Wilson et a1. 1991; DM Scheltinga, BGM Jamieson and

AN Hodgson, unpublished results;

Figs. 6G, H),

Rhacophorus

arboreus

(Mizuhira et al. 1986) and

R. schlegelii

(Mizuhira et al. 1986;

Oka 1980)

have what

appear

to be the most modified known amphibian

sperm.

Chiromantis spermatozoa have the form of a coun-

terclockwise corkscrew, viewed from the anterior end.

The coils involve the acrosome. nucleus and midoiece-

which consists

of mitochondria

around

the base of the

nucleus. A pair of free paraliel flagella comprise the

tailpiece.

A crystalline matrix composed of many

microtubules

(Fig. 6H) surrounds these.

The coiled head of the sperm is interpreted as an

adaptation

to the special microfertilization environment.

The coil unwinds in different aqueous

media and this

lengthens the sperm seven to eight times. The sperm

also exhibits a "star-spin" movement,

comparable

with

the hyperactivation

of mammalian sperm. The two tails

of the sperm seem to enhance

this movement,

probably

facilitating movement in the gelatinous

foam and pene-

tration of the outer layers

of the egg (Mainoya 1981;

Wilson

et aL l99l t.

318 / CHaprnn29

The Spermatozoa

of the Amniota

From detailed comparative

and cladistic considera-

tions of the anatomy

of amniote

sperm,

Jamieson

(1995a)

recognized

the fbllowing characteristics

of a

hypothetical

plesiomorphic

amniote

spermatozoon

(Fig. 10). This model is not overly speculative,

as it is

virtually identical with that of the lowest extant

amniotes,

the Chelonia and Sphenodontida.

Relative to

the tetrapod ground plan, deduced from common fea-

lures

of the

amniote

and lissamphibian

sperm. amniotes

are seen

to have few basal synapomorphies.

Amniote Spermatozoal

Symplesiomorphies

Plesiomorphic features of basal amniotes, retained

from their tetrapod ancestry,

and still seen

in Chelonia

and Sphenodontida

and to varying extents in other

amniotes,

are as

follows. The generalized

plesiomorphic

amniote spermatozoon

(Fig. 10) is elongated and fili-

form, with a hollow anterior conical acrosome

vesicle

ovedying a simple subacrosomal

cone. The base of the

acrosome

invests the tapered anterior tip (rostrum) of

the nucleus

and rests

on pronounced nuclear

"shoul-

ders."

The subacrosomal

space

within the acrosome

con-

tains two or three axial rods (putative perforatoria) or,

less likeiy, only one rod. These penetrate the nucleus

deeply,

almost

to its base,

in endonuclear

canals

The nucleus is plesiomorphically elongated and cylin-

drical in amniotes from Chelonia thtotgh Sphenodon,

crocodiles,

squamates,

birds, monotremes

and, in theri-

an mammals, the pangolin alone (Leung, Cummins

1988), as in lissamphibirns. At the base of the nucleus

there is a compact fossa (implantation fossa).

Associated

with this are two triplet centrioles.

The distal

centriole forms the basal

body of the flagellar axoneme.

Whether the presence of an annulus

is plesiomolphic or

apomorphic is debatable. The terminal portion of the

9+2 axoneme

forms a short

endpiece

distinguished

from

the principal piece

by the absence of the fibrous sheath.

Amniote Spermatozoal

Synapomorphies

The Chelonia and Sphenodontida

are considered

the

most basal extant amniotes

and have virtually identical

spermatozoa

(Jamieson 1995a;

Jamieson,

Healy 1992).

The characteristics

of these

include features considered

synapomorphies

of the Amniota which are simultane-

ously symplesiomorphies.

for these two orders and for

the remaining amniotes. The amniote synapomorphies

are

listed below.

The distal centriole is extremely elongated and

extends

the entire length of the long midpiece (the latter

defined by its mitochondria) in turtles, the tuatara

(Fig. 11), crocodiles, and ratites, an apparent

basal

synapomorphy of amniotes. These elongate centrioles

differ from.most metazoan

basal bodies in being

penehated by two central singlets from the axoneme.

Thus,

in spermatids

of the ratite

Rhea, the distal centriole

elongates

and,

late in spermiogenesis,

becomes

penetrat-

ed by a central pair of tubules from the developing

axoneme

(Phillips, Asa 1989). The shorter,

though still

elongated,

distal centriole in the rooster and the some-

what shorter

centriole

in guinea

fowl, at 0.6 pm, and

Geopelict

striata, at 0.5pm (Jamieson 1995a); the short

centriole

in squamates;

and the vestigial

centriole

monotremes

possibly represent

secondary

reduction in

length

of the distal centriole

(Healy,

Jamieson

1992)'

cul-

minating in almost

total reduction

in therian

mammals.

In turtles, tuatara

(Healy, Jamieson 1992, 1994;

Jamieson

1995a;

Jamieson,

Healy 7992;

Ftg. 11),

Caiman crocodylus and Crocodyhts

johnstoni (Jamieson

1995a; Jamieson

et al. 1997;

Fig. 12), the mitochondria

have concentric cristae, known elsewhere

in amniotes

only in the sperm of some marsupials, notably the

Woolly opossum,

Caluromys

philander (Fawcett 1970;

Phillips 1970) and the Virginia opossum,

Didelphis vir-

giniana (Temple-Smith et al. 1980) and aiso in the

macropod.

Lagorchestes

hirsutus (Fig. 16). The marsupi-

al condition may be homoplasic

but it is here considered

possible that it is evidence

of an ancient synapsid,

and

therefore,

mammalian link, with the lower, anapsid

amniotes.

The cristae also

tend to a circular arrangement

in monotremes

(Bedford, Rifkin 1979; Carrick, Hughes

1982).

The mitochondrial

cristae in the three "reptilian"

taxa usually surround a large central dense

body. In all

other

amniotes

studied, the cristae

have a "conventional"

appearance,

being linear or curved, as in Lissamphibia,

but never

concentric,

and do not surround

a dense

body.

In spermatids

of Sphenodon

(Healy, Jamieson

1992;

Jamieson,

Healy 1992), the cristae have the linear

appearance

usual for metazoan sperm

and

the concentric

alTangement

is a late development.

Phylogenetic

"rever-

sion" of concentric

cristae

to the linear condition seen

other

amhiotes

would need only suppression

of this fina1

transformation

(Jamieson,

Healy 1992).

A dense ring, the annulus,

at the posterior end of the

midpiece

is a feature of many metazoan

sperm.

It is

clearly plesiomorphic for amniotes,

occurring in all

classes

(Jamieson,

Healy 1992)

but its absence

Dipnoi possibly indicates apomorphic re-acquisition in

tetrapods.

It is well developed

in Chelonia, Sphenodon

(Healy,

Jamieson

1992, 1994t Jamieson

1995a; Fig. lt),

Caiman crocodylrs (Saita

et al. 1987), the American

alligator (Phillips, Asa 1993) and Crocodylus

johnstoni

(Jamieson

et al. 1991|.F\9.

12).

A dense

fibrous sheath

(Fig. l0) must, clearly, have

developed

as an annulated structure in the earliest

amliotes, as

it is present in all amniote

classes.

With the

exception

o[ squamates

(Fig.

l5C).

it commences

imme-

diately behind the midpiece, as in turtles, Sphenodon

Healy, Jamieson

1992; Jamieson,

Healy 1992; Fig. 11),

Cuittrtm crocodylus

and, Crocodylus.johnstoni

(Jamieson

1995a:

Jamieson et al. 19971' Fig. 12), ratites,

non-

passerines

(Fig. 138)

and in mammals

(Jamieson

1995a;

Fi-es. 16, 17F, M-Q).

Nine longitudinal dense fibers (coarse

fibers) periph-

eral to the nine axonemal doublets,

or to the distal centri-

ol!' also where this is elongated

as in Entydura,

Slrhenodon and Crocodylus, are a fundamental feature of

.rmniote sperm

(Fig. 10),

being found

in all classes

Jrmieson 1995a;

Jamieson,

Healy 1992,

Jarnieson,

Scheltinga 1993). As nine peripheral fibers are seen in

..rnpreys and. Pantodon (references

in Jamieson 1991)

rut also in heterobranch

and cephalopod

molluscs

(Healy

i988. 1990),

it might

be considered

that nine is the basic

..rrcopterygian,

rather than merely amniote, number and

:r..ir

amphibians

have lost all but those represented

by the

:lbers

at doublets

3 and 8. However,

there is no evidence

::r

ertant Lissamphibia

for such

a reduction

and the pres-

cnce of only two lateral elements

in dipnoans and

I.,irirrteria suggests

that nine fibers were an amniote

.r napomorphy,

albeit homoplasic

with the other, non-

.:mniote taxa. The dense fibers are small in tufiles, the

::.ratara

(Fig. ll), Caiman crocodylus and Crocodylus

',hrstoni

(Fig. l2), squamates

(Fig. 15B), birds

Fi-es. l3B, F, N, 14) and monotremes;

this is reviewed

:1 Jamieson

(1995a).

It is possible

that a further basal

amniote

apomorphy

:. enlargement

and lateral displacement

of two fibers, at

ioublets 3 and 8, and that a1l fibers in the centriolar

:i'-gion

intruded into the inter-triplet radii, as in "lower"

-rmniotes

(Chelonians,

Sphenodon

and crocodiles),

-.'en

in Jamieson

(1995a).

In the principal piece,

two longitudinal keel-like out-

'*lrd projections (longitudinal colurnns),

at doublets

: lnd 8, may be present,

each aligned with an inward

:rojection of the fibrous sheath, as shown for the mam-

:,:lalian

spermatozoon

by Fawcett

(1975); this is also

.hr.rnn

in the elephant shrew

Macroscelides,

in this

.:udr' (Fig. 17O). At least the two inward projections

are

rresent

in Sphenodon

(Fig. 11)

and

may be a basal

.r napomorphy

for arnniotes.

Loss or transformation

of the retronuclear

bodv, pre-

rcnr

in dipnoans

and

(as

the

neck

structure.l

in urtdeles

:rrpears

to have occurred

as an amniote apomorphy.

li is never

retained

in discrete

form seen in Dipnoi.

li. putalive

homolog in urodeles

but questionable

r.'mology

of the striated

columns

of mammalian

snerm

I-rgs. l7G.

H. Lt has

already

been alluded

ro.

Chelonia and Sphenodontida

The spermatozoa

of the Chelonia and of Sphenodon

Fig. ll) are indistinguishable

from the amniote ground

llan.

VsnrssnArl

Sprnv Psyr-ocsNy

I 319

Crocodilia

The ground plan for the Crocodilia, as exemplified

by Crocodylus

johnstoni (Jamieson

1995a;

Jamieson et

al. 1997', Fig. 12), is very similar to that of the Chelonia

and

Sphenodon. All th-ree have

two or three

endonuclear

canals and, though requiring further confirmation for

crocodiles,

concentric

cristae with intramitochondrial

bodies. In Crocodylus

.iohnstorri,

the mitochondria are

subspheroidal

to slightly elongate

and possess

few sep-

tate to (more externally) concentric cristae (Jamieson

1995a;

Jamieson et al. l99l); a central dense body

reported

for caiman crocodilus

(Saita

et al. 1987) is

questionably present.

The spermatozoon

of C.

johnstonl is apomorphic

relative to those of Chelonia and Sphenodon in

reduction of concentric mitochondrial cristae. It is less

similar to that of ratites than is that of Caiman

crocodilus, differing from ratites in having more than

one

perforatorium.

Synapomorphies

of Crocodilian Spermatozoa

In Caiman crocodylus

(Saita et al. 1987) and

Crocodylus

johnslrni, the investment

of the two central

singlets of the axoneme or of the distal centriole in a

thick dense

sheath

(Fig. 12) differs from the density,

resembling

a fiber, associated

with the singlets

Chelonia.

Sphenodort

{Healy. Jamieson

ll992'1 and

(homoplasically?)

snakes

(Jamieson,

Koehler 1995;

oliver et al. 1996).

Further Apomorphies

in Caiman

Restriction of the endonuclear

canal to the anterior

region of the nucleus

indicated by Saita et al. (1987) is

cleady apomorphic

but requires

confirmation.

The Spermatozoa

of Aves

The conical acrosome,

fibrous sheath, elongated

centriole

and

nine dense

[ibers

of ratites

are svmole-

siomorphies

nol

proving

avian monophyl5.

Funhirmore.

monophyly of ratites cannot be considered proven, as

features

considered

to unify them -conical

acrosome!

fibrous sheath and elongated centriole (Baccetti et al.

199 l)-are all symplesiomorphies.

Spermatozoal

Symplesiomorphies

of Birds

In birds, a conical acrosome

vesicle penetrated

almosl

to its

tip by a subacrosomal

space whiih contains

a rodlike perforatorium has been demonstrated ultra-

strxcturally in the non-passerines:

turkey, Meleagris gal-

lopavo; rooster, Gallus domesticus

(Fig. 13A); guinea

fowI, Numida meleagris (Thurston et al. 1987);

mallard

duck, Anas plotjrhynchos (Humphreys 1972); the quail

Coturnbc

coturnix (Humphreys

1972);

and

parrots

320 |CgtpreR29

Acfosomev€sich-#

Losi

in monotremes

and

only

in other

amniotes

Perfordorium

prenuclear

squamdes

Elonuale

nudgIs

ln basal

members

amniote

clSsses

Esal nuclefilossa

Triple

in ratiies.

Fufinellike

skinks

Densa

body l*erd to tantrio

Sfherodal caiman

and snakEs

sirided columns in

mammals?

Sevs.dl

mitodlondria

sserm

Dttss seclion

Mitorfi

ondrid cridae concentdc

As in Cheloni4 .'p4el?oda\'?,

crocodiles

(and WoollY

oFossum).

'Conventional'

in other amniotes

s dense

peripheral axonemd

fibl€s

All amniotes

excepting

tinafiou

mid- and

Frincipal

piece

0r. in rhe4

princiFal

piEce

(Jarnieson

et al. 1995;

Figs. 13J,

14). This

has also

been

demonstrated

in the ratites (palaeognaths) tinamou,

Eudromia elegans (Asa et al. 1986); ostrich, Stuthio

camelus (Baccetti et al. \991:' Soley 1993, 1994)" and,

emt:ul.,

Dromaius novaehollandiae

(Baccetti et al. 1991).

Figure 10. Diagrammatic lepresentation of the hypothetical

plesiomorphic amniote spernatozoon.

Rffiimd c€Jtti ole

Distd c€fitriole

eHtefl

ding

thl!ughout midPiece

As rn

Chel0nia

sf,4ena?ot,

Cr0,:odilia

anrl ratitEs.

Lost

ih mammals

D€nse

imrfrnitodronddd h0dY

As in

ChElonia"

SphefEdi?r,

and

Cracodilia.

Translormed

into intermitoch0ndrial

structures

ir'r squamates.

Lost in

birds and

monotremes

Annulus

ln all

amniotps

but reduced

0r sbsent

some

squamates

and some

hirds

and

reduced in monotremes

]lo glFsUen shealh

Present

only in

tinamou

Fibrous

shedh ol atmnenE

Annulde,

excepting

n0n-Fassennes

in which it is

amorphous

or lost.

This is a basic sarcopterygian

synapomolphy

and there-

fore avian symplesiomorPhY.

The endonuclear

canal

extends

almost

to the base of

the nucleus as in Chelonia and Sphenodon

(Hea\y,

Jamieson

1992, 1994;

Jamieson,

HeaIy 199! Fig. 11) in

HasnHmet l

SimFle

f,lbaf,msomd

cone

Paracrystalline

itl

Lost in

ratites

T1r[?

€ndonuclefl

canals

2 or

a in Chelonia

and

CroaDdY[is.

5p4efi0dfl1.

1 in other

amnioles

lost

in mrnotremes

and

squamate

Endonud€ffr

cands deep

in Chelonia

G'ofirqY/as,Sphenodon,

an d rhea.

lvost

of length

0f nucleus

in tinam0u.

Noi

exiending into

midPjece

{does

only in squamdes)

Figve 11.

Sphenodon punctatus.

Semrdlagraramatic

representation

of spermatozoal

ultrastructur€

as seen

in longitudinal

iection and

lransverse

sections

by transmission

electron

microscoDv.

rFrom

Heal).

Jamieson

Iqa2.)

putatively

more primitive ratites,

such as

tinamou (Asa et

al. 1986),

where it is probably plesiomorphic

relative to

the

shorter

condirion

in other

non-pasierines

and

advanced

ratites.

the

perforatorium

being

wholly

pre-nuclear

in the emu (Baccetti et al. 1991).

However.

there is a possibility that deep penetration in ratites is

secondary

(Jamieson,

Healy 1992).

Nine dense peripheral

fibers, a basic amniote

synapomorphy,

and avian symplesiomorphy,

have been

observed

in turkey,

rooster

and, though

requiring

VERTEBRATE

SpEnu Pnvr-ocrxv / 321

confirmation, in guinea

fowl (Thurston,

Hess 1987);

mallard duck (Humphreys

1972);

panots

(Jamieson

al. 1995); doves

such as

Geopelia striata

(Jamieson

1995a; Figs.

138, F, H); passerines

stch as Grallina

cyanoleuca

(Fig.

13N); and in the anteriormost

region of

the

principal.piece

of ratite

spermatozoa

(Asa

et al.

1986;

Baccetti et

al. 1991; Soley

1993, 1994).

They are

present

in suboscine

and the more apomorphic

oscine

passerines

(Fig.

13N),

being larger in the latter.

Spermatozoal Synapomorphies of Birds

Like the sperm

of ratites

and other birds,

parrot

sperm differ from those

of "reptiles"

in reduction

of the

subacrosomal

material

(subacrosomal

cone, excluding

any

perforatorium)

to a negligible

amount

(Jamieson

al. 1995t Figs. 13J,

14). Although

avian

sperm

present

several apomorphies,

this

may be the

only basal synapo-

morphy

for bird sperm. It is possible,

though,

that a

fur-

ther basal

synapomorphy

has been

the loss

of concentric

cnstae,

Secondary Spermatozoal Synapomorphies

of Birds

Most

of the

synapomorphies

of bird sperrn appear

have

been

derived within

the class

rather than

basally.

The endonuclear

canal is limited to the anterior

1 to

2 pm of the

nucleus in rooster

(Fig. 13A),

guinea

fowl,

turkey

and

panot (Figs.

13J, 14) and

the anterior

third

the nucleus

in the

ostrich

(Baccetti

et al. 1gg1).

Restriction

of the endonuclear

canal to the anterior

region

of the nucleus

in non-passerines

and

passerines

may be a synapomorphy

of these, homoplasic

with

crocodiles

and

derived ratites

(emu,

ostrich).

In the non-passednes

rooster

(Figs.

138,

C) and

guinea

fowl, the fibrous

sheath

has tansformed

into an

amorphous

sheath

(Thurston,

Hess

1987). Less

certainlv

derived

is adhesion

of all nine dense

fibers

to rheir

axonemal

doublets

(Figs.

13F,

H, N, 14), a feature also

seen in monotremes (Jamieson

1995a).

If birds

are, as is commonly

held,

the sister

group

crocodiles,

their

linear

cristae

(Figs.

13B,

G, H, L, M)

would

indicate

loss

ol the concentric

cristae

of a com-

mon ancestor

with crocodiles.

However,

the sister group

relationship

with crocodiles

was not supported

in a

cladistic

analysis

of sperm

ultrastructure

(Jamieson,

Healy

1992t.

Alrhough

a bird-mammal

sister

group

rela-

tionship

shown

in the latter analysis

is doubtfui,

it is

remarkable

that this relationship

was

supported

results

from

a molecular

study

(Hedges

et al. 1990).

Not all non-passerines

possess

a conical

acrosome,

and the

modifications

which

occur

appear

to be sec-

ondary,

internal

apomorphies

within the Aves.

A small,

approximately

spherical

acrosome

has

been described

for the white-naped

crane,

Grus vipio (phillips et al.

1987),

for Jacana

jacana (Saita

et al. 1993)

a;d most

.".."",,""f/@il

\@d

""",.,

_ .-<7ll

,.rr.n

o-t-ffl

{st

Algi

322 I CnlrPrBB29

Figure 12. Cto.odylLts

.iohtlstc'nr'

Diagram of spermatozoal ultrastruc-

ture as seen

in longitudinal section

and

transverse

sections

by transmission

elecfion microscopy. (From Jamieson

al. 1997.)

g deGe p€r?hs_al

rbres

eno

riec"-[!!J

:-*li[i o

Pri

ncipal

sit mtochondna

Charadriiformes (Fawcett et al. 1971), and for the

woodpecker

Melanerpes

carolinus (Henley et aI 1978)

ThesJ latter avian taxa are considered

to be advanced

non-passerines,

on the basis of DNA hybridization

studies

(Sibley

et al. 1988,

1990).

In the columbiforms,

such

as Geopelia

striala

(Fig. 13D) and Ocyphaps lophotes (Fig. 13K), even a

perToratorium

is absenl

although.

at leart

in Ceopelia

'\lrlala,

some

longitudinalll

orientaled

subacrosomal

material

present cnd

lacunae

are

present in the

nucle-

us which may represent

a vestigial

endonuclear

canal

(Jamieson 1995a).

Both of these columbiforms appear

plesiomorphic

in retaining a nuclear

rostrum (Figs' 13D,

k), ttroogtt

it i. possible

that its presence

is a teversal'

In narrots

(Jamieson

et al. 1995;

Fig. 14) and doves'

str.r,ch

is Ocyphaps lophotes and Geopelia striata' the

fibrous sheath

is lost (Fig. l3I). These

taxa would there-

fore orovide valuable

controls

in experimental

investiga-

tionJof the lunction of the fibrous sheath.

A sheath

of putative

glycogen

external

to the fibrous

sheath

is known only in the tinamou (Phillips' Asa

1989) and cannot be ascribed

to the plesiomorphic

amniote

sperm.

An annulus

is basic to bird sperm,

being seen

ratites,

rooster

(Fig. l3B), guinea fowl, and columbi-

forms (Aia, Phillips 1987),

but is apomorphically

absent

in oarrots

(Jamieson

et al. 1995;

Fig. 14).

Dense

fibers are described

"tiny" for the rhea,

are

absent

from the tinamou (Asa et a1.

1986), and are

greatly reduced

in columbiforms (Jamieson 1995a)'

Very imall dense

fibers are present only in the distal

region of the midpiece in the rooster

(Fig. 138) and

mallard; dense fibers in turtle dove

sperm disappear

before maturation is complete

(Asa,

Phillips 1987),

rhough they

persist

through a short

region of the mid-

piece

in Geopelia striata

(Jamieson

1995a; Fig. 13F).

Oscine Synapomorphies

Oscine spermatozoa have

an extremely

large acroso-

mal complex in contrast to non-passerine

and suboscine

\permatozoa

in which the acrosome

is shofi relative

to the

nucleus,

as in reptiles

(Jamieson,

Healy

1992;

Jamieson,

Scheltinga 1993, 1994).

In passerines

the acrosome

\esicle

becomes an elongate

single-keeled

helix, with no

elident subacrosomal

cone, like that

of finches

(Kondo

al. 19881 Koehler, 1995).

A helix of densely

packed

microtubules

invests the

aroneme in passerines,

seen in Grallina cyanoleuca

Fie.

13N).

Spermatozoa

of the Squamata

Squamate sperm have

all of the basic amniote

:\ napomorphies

described

above as their plesiomor-

phies.

Their synapomorphies

strongly support squamate

monophyly.

Synapomorphies

of Squamate

Sperm

The perforatorial rod in the Squamata is wholly

prenuclear.

It sits on the tip of a well developed

subacro-

somal cone and might be termed pre-subacrosomal

tCarlia rubigularis, Fig. 15A). Endonuclear

canals

are

absent

(Jamieson

1995b;

Jamieson

et al. 1996;

Oliver et

aL 1996). This anterior restriction, and reduction to a

single perforatorium, are clearly apomorphic

relative to

the basal amniotes

Chelonia

and Sohenodontida.

Presence

ol a well developed

epinuclear

electron

lucent region is a squamate autapomorphy (Jamieson

1995a, b; Jamieson

et al. 1996;

Oiiver et a|. 1996;

Fig. 15A). The intermitochondrial

rings or dense

bodies

tctenotus inornatus, Fig. l5B; C. rawlinsoni, Fig. 15C)

of squamate sperm are regarded as derivations of the

intramitochondrial

dense

bodies (Carcupino

et al. 1989;

Healy, Jamieson

1992;

Jamieson, Healy 1992)

seen in

basal amniotes (chelonians

and sphenodontids).

Origin

of intermitochondrial material from mitochondria has

been confirmed ontogenetically in the sperm of some

squamates

(Oliver et al. 1996). Extramitochondrial

dense bodies are almost limited to squamates

but are

seen,

poorly developed, in the doves Geopelia striata

(Jamieson

1995a; Figs. l3E, G) and Ocyphaps

lophotes

(Fig. 13L) in which, although

appearing

homoplasic,

they may well indicate persistence

of a genetic

basis laid

down in early amniotes.

In squamates,

alone in the Amniota, the fibrous

sheath

extends anteriorly into the midpiece (Clenotus

rawlinsoni, Fig. 15C), a striking squamate

autapomor-

VsnrnsnArB SpERM PHyLocEN\ | 323

phy (Healy, Jamieson 1992; Jamieson,

Healy 19921

Jamieson, Scheltinga 1993).

In the Squamata, the subacrosomal

cone

has a

paracrystalline

substructure

(Butler, Gabri 1984;

Carcupino et al. 1989; Furieri 1970),

recently confirmed

for Sphenomorphus and Eugongylus group skinks

(Carlia rubrigularis, Fig. 15A), the gekkonid

Heteronotia binoei, and in snakes

(Oliver et al. 1996).

It constitutes a basal synapomorphy of the Squamata

(Jamieson.

Healy |

992;.

Shortening of the distal centriole (Fig. 15C) from

the elongated

basal amniote

Character is a basal

apomor-

phy of squamates

(Jamieson

1995a).

In squamates,

there

are nine peripheral

fibers in the midpiece (Fig. 15B),

but

at the level of the annulus

the oniy well developed,

though small, peripheral fibers are the double fibers at

doublets 3 and 8; by the beginning

of the principal piece

all nine dense fibers are already vestigial (Ctenotus

ornatus, Fig. l5B) or absent

(Jamieson

1995a;

Jamieson,

Scheltinga

1 993).

Secondary Spermatozoal Apomorphies

in Squamates

The many internal apomorphic changes

within the

Squamata

are discussed

in some detail by Jamieson

(1995b).

Consideration

of these, and the various

sub-

groups, is beyond the scope of this chapter. Only the

striking apomoryhies

of snake sperm

will be discussed.

Snake spem are characterized,

apomorphically, by

multilaminar membranes

in place of the normal plasma

membrane

of the midpiece and axoneme

(Jamieson,

Koehler 1995;

Oliver et al. 7996); this can be compared

to pygopodids (Jamieson

1995b;

Jamieson

et al.. 1996).

Snake sperm are unique in the Squamata

in the

immense elongation of the midpiece (Jamieson 1995a,

b). Snake sperm show reduction of the epinuclear

elec-

tron-lucent

region and reduction

or loss of the pedorato-

rial base

plate.

and

grealer

development

of exrracellular

tubules

than is known in any other squamate

(Jamieson

1995a.

b).

Spermatozoa

of Mammalia

A recent paper on the spem of Tarsius bancanus

found spermatozoal

ultrastructure

to be of limited value,

in isolation, in reconstructing

the phylogeny of primates

(Robson

et al. 1997).

However,

other works have

shown

the utility of spermatozoal

ultrastructure for this pur-

pose, notably for marsupials (Harding et al. 1987;

Temple-Smith

1987), rodents

(Breed 1997),

elephant

shrews

(Woodall

er al. 1995)

and megachiropGrans

(Rouse,

Robson

1986).

Particularly pertinent to establishing

the plesiomor-

phic sperm

type for the Mammalia, and at the same

time

mammalian synapomorphies,

are accounts

of the

324 I Cueprpp.29

..d

&:.

Figure 13. Spermatozoa of Aves. TEM sections.

A-C: Gall&.t

ilomesticus,

rooster. A: LS of acrosome,

pedoratodum and endonu-

clear canal.

B: LS of midpiece

at junction with principal piece.

Note amorphous

fibrcus sheath distal to annulus. C: TS of principal

piece and, below it, an endpiece. D-l: Geopelia striata. peacefiil

dove. D: Acrcsome, showing absence

of perfbratodum. E: Anterior

midpiece, showing dense bodies resembling

squamate intermitochon-

drial bodies. F-H: TS of midpiece, showing progressive

distalw ds

reduction of peripheral fibers (in the order F, H, G) and, in H, a

dense body. I: TS of principal piece or endpiece; a fibrous sheath

absent from all sections.

J: PLatlcercus elegazs,

crimson rosella.

acrosome, perforatorium and endonuclear canal- K-Ml. Oclphaps

Iophotes, crested

pigeon. K: Showing a nuclear rostrum in the

absence

of a pefbratorium. L & M: TS of midpiece, in L, showing a

dense body. N'. Grallina cyanoleuca,

magpie lark. Helical sheath

micrctubules

around the midpiece of the oscine sperm.

(Original.)

remarkably reptilian sperm of monotremes (Bedford,

Rifkin 1979; Canick, Hughes

1982) and of pangolins,

in the order Pholidota

(Ballowitz

1907;

Leung 1987).

Symplesiomorphies of Mammalian Sperm

The sperm of monotremes

(Bedford, Rifkin 1979;

Carrick,

Hughes 1982; Jamieson,

Healy 1992)

are

remarkably primitive in retaining the elongated

conical

structure of the acrosome and the subacrosomal cone.

The acrosome retains

its plesiomorphic conical form in

the pangolin,

Manis pentadactyla

(LKP Leung 1987,

and

personal

communication).

The acrosome in M. pen-

tadactyla (1.7

-2.2 ;tm long) sits caplike over the rostral

fifth of the nucleus, with no substantial

subacrosomal

material, and does not appear

to have the well defined

equatorial segment

which distinguishes

most other

Eutheria, which further differ in having flattened sperm

heads, from the pangolin

sperm

(Leung 1987).

Thus, monotremes

appear to be the only mammals

which retain the subacrosomal

cone. The nucleus is

pointed,

elongated

and circular in cross section

monotremes (Bedford, Rifkin 1979; Carrick, Hughes

1982)

and in pangolins

(Ballowitz 1907; Leung 1987)

but this condition is unknown in marsupials.

Synapomorphies

of Mammalian Sperm

The attempt may be made to distinguish basal

synapomorphies

of mammalian sperm from secondary

synapomorphies

of the three chief constituent groups,

Prototheria (monotremes),

Metatheria (marsupials)

and

Eutheria

(here

termed allantoplacental mammals).

Monotreme sperm are in many respects

"reptilian."

Synapomorphies

which they show and which can rea-

sonably be attributed to an ancestral mammal are as

follows.

In mammals, a rodlike perforatorium and endo-

nuclear canal do not occur (Jamieson

1995a), a loss

which is homoplasic with that in some

non-ratite birds.

In the spiral midpiece of mammals, the number of

&"^l

[],

acrosome

ves|cle

perforatorium

enoonuctear

canal

distalcentriole

mitochondrion

1 prn

of midpiece

principalpiece

Frgure 14. Melopsiftacus

undu-

lalus, the budgedgar. Diagram of

spermatozoal ultrastructure as

seen in longitudinal section and

transverse sections by transmis-

sion electuon microscopy. (From

Jamieson et al. 1995.)

enoonuctear canat

mitochondrion

posterior

end

VsnrnsnAre SpenM

PHvr-ocsl,rv

I 325

Synapomorphies of Extant Monotreme Sperm

The synapornorphies noted for the mammalian

ground plan are those attributable to monotremes.

As the first mammals were

presumably

egg-laying

monotremes, and the sister

group

of Metatheria

and

Eutheria

(Rowe

1988), it is questionable

that there are

any features of the sperm of extant monotremes which

may be considered apomorphic departures from the

mammalian

ground plan.

Synapomorphies of Therian Sperm

No mammalian sperm above the Monotremata is

known to possess

a subacrosomal

cone

(Macroscelides,

Figs. 17A-D). In reptiles, the

peripheral

fibers

at 3 and 8

are detached from their corresponding doublets,

while

the

other seven fibers are attached to their doublets.

In birds and monotremes,

all of the

peripheral

fibers are

attached to the corresponding doublets

(Fawcett,

Phillips 1970). It has

been said that in marsupial and

eutherian sperm, the

peripheral

fibers are detached from

their doublets

with the exception of fibers

3 and

(Jamieson

1995a). in contrast to what is seen in birds

and monotremes.

However, in the present

study, such

differential

attachment of fibers 3 and 8 could not be

confirmed from a survey of micrographs

from the litera-

ture on marsupial or eutherian sperm

(Macroscelides,

Figs. 17I-K, O). On the

other hand, attachment of these

two fibers to the longitudinal

columns of the fibrous

sheath

(Fawcett,

Phillips 1970) is demonstrable,

but this

association is also

seen in squamates

(Jamieson

1995b).

The fibers at 3 and 8 may be lost in at least

some

therians,

as in elephant

shrews

(Fig. 17O).

Synapomorphies of Marsupial Sperm

The sperm of the rufous hare-wallaby,

Lagorchestes

hirsutus, is shown

in Figure 16,

to exemplify

a marsupi-

al spermatozoon.

In marsupials,

the acrosome

covers, in

varying

degrees, only

one surface of the

"anteriorly"

flat-

tened

(antero-posteriody

compressed)

nucleus.

This con-

trasts with the condition

in monotreme

and eutherian

spem, including

the

pangolin,

in which the acrosome

forms

a cap

which surrounds

at least the

proximal

region

of the nucleus

and

projects

for varying

distances

to its tip (Harding

et al. 1979:

Macroscelides,

Figs. 17A,

B). The

marsupial

condition

is partly a correlate

distal-proximal

flattening

of the

nucleus.

In marsupials,

as exemplified

by a member

of the

primitive

family Didelphidae

and more

derived

taxa

(Fig.

16), the nucleus

is compressed

during

spermiogene-

sis in a plane

perpendicular

to the flagellum (phillips

1970).

The result

is that

the implanration

fossa

is situated

at the middle

of the resultant

long axis

of the nucleus,

giving a T-junction.

This condition

is tvpical

marsupials.

Only in the

koala,

Phascolarrtos

cinereus

gyres varies from 55 ro 300 (Fawcett 1975;Fig. l7F)

but is not specified for monotremes (Bedford, Ritkin

1979:

Carrick,

Hughes 1982).

Great reduction

of the

distal centriole is a mammalian apornorphy.

Though

normally forming the basal

body of flagella, it is at most

a vestlge

rn mature mammalian sperm

(Baccetti.

Afzelius

1976; Fawcett

1975).

perforatorium

coarse fibre

326 I Cnlrlep.29

Figure

J6.

A marsupi-

al spernatozoon

exern-

plified by La

rc he

ste s

,l i r,r'/

r& r. rufous hare

wallaby.

LS, showlDg

lnarsupial feature ol

compression of lhe

nucleus

in a plane at

right angles

to axoneme

and subsequent

rotatton

of the nucleus so that il

lies along the midpiece.

(Frorn S Johnston,

L Smith, F Carrick

and BGM Jamieson,

unpublished

data.)

Figure 15. Sper-rnatozot

of Squtmata TEM sections A: Callid

rubrigularis.LS tirough acrosome

and anteior nucleus.

Anow indi-

cates

nucleaf shouldefi. B: Ctenotus

itlonrdtus. TS through an inter-

mirochonclrial

dense body. showing that these bodies

form large

rings iD sphenomorph

skinks. C: C rarLlirsoli. LS through posterior

nucleus

and complete

midpiece, showing series

of intermilochondri-

al lings ancl ijblous sheaLh

p(]netrating

midpiece (Ftom

DM Scheltinga

and

BGM Jamieson,

unpublished

data

)

(Phascolarctidae)

(Har-ding

et al. 1979) and, from light

microscopy

observatious,

the wombat (Hughes

1965)

does ihe flagellum implant at the short axis of the

nucleus,

the long axis of the nucleus being continuous

with the long axis of the flagellum. Harding et al. (1919)

appear

to regard

the koala sperm

type as close

to that of

the ancestral

marsupial.

having considered

the altemative

possibility that the "eutherian" form of the nucleus

is a

highly specialized

product of the long phylogenetic sepa-

ration of vombatoids

fiom other marsupials.

The present

writer inclines to the view that one "limb" of the nucleus

may have

been

lost

in the

koala

sperm,

giving a

Figure l7 (dght). The eutherian

spermatozoon

exenplified by thc Macroscelidae.

clephant shrews. All arc Mactosrclides proboscideus,

except F, whjch is petrotlro,tlus tetra(t0ad\lus.

A-Bt LS of acrosome

and nucleus.

C-D: proximal and more distal TS oI acrosome E: TS of

nucleus,

which is envelope.l

by the acrosome

for its entire length-

Fr LS of e[tire midpiece.

showing cytoplasmic

droplet and helical anange-

nent of mitochondlia.

with adjacenr

nuclcus ind librous sheath

of principal piece. G-H: LS of basc

of nucleus

and anteriol region of midpiece

in planes approximately

at right angles.

I-J: TS of midpiece,

showing axoneme

with periphelal fibers. K: same' through

cytoplasmic

droplet L:

LS of base

of nucleus

and antenor

regron

of midpiece,

showing pariof cytoplasmic

droplet.

M: LS ofjunction of midpiece

and principal piece,

showing annulus

ar the commencenenr

of fhe tibious sheath.

N: iS of fibrous sheath lurther distally. Ol TS ol principal piece where the periph-

eral tibers at 3 and ll have each

been

replaced by an inward keel fiom the longiludinal column oi the fibrous sheath P: TS of principal pieces ol

which some,

nore distal. have

lost the peripher;l fibers.

Q: Ts of two endpieces.

with disrupted

arangemeit ot' microtubules (odginal )

VsnrrsnAre SpERM

PHyLocENy

/ 327

.cF,\ |

- l.Opm

I '.o

328 I CgrprsB29

secondary

appearance

of implantation

of the axoneme at

one end of the nucleus.

The fact that, as in other marsu-

pial sperm,

the acrosome

does

not cap the free tip of the

nucleus

but lies along

the abaxonemal

face of the

nucleus

seems to support

this view.

Although developmentally

the axoneme forms a

T-junction with the nucleus in didelphid sperm,

the head

rotates so that its long axis comes

to lie along that of the

axoneme.

This rotation is reported not only for didel-

phids but also for phalangerids,

petaurids,

dasyurids,

and peramelids

by Harding et al. (1979) and is seen

the macropod

Zagorchestes

hirsutus

(FiE. \6).

Synapomorphies

of Eutherian Sperm

If the elongated,

cylindrical nucleus and caplike

acrosome

of the pholidotan Manis are accepted

as ple-

siomorphic

features of therian and eutherian

sperm, flat-

tening of the nucleus

(here

in the plane of the flagellum)

(Figs.

17A, E), with the acrosome

(Figs. l7C, D), and

the many variations

of acrosomal structure

in the

Eutheria must be accepted as internal, secondary

euthe-

rian apomorphies.

What features, if any, of the basal

eutherian

sperm were apomorphic

relative to the ances-

tral therians

is uncertain.

It cannot be determined

whether ancestral marsupials

had cylindrical sperm

nuclei, but if antero-posterior

compression

were a basal

synapomorphy

for marsupials,

it would have

to be con-

cluded that eutherian sperm, as exemplified by Manis,

are

plesiomorphic relative

to those of marsupials.

This is certainly the situation with regard to extant

pholidotan

sperm

relative

to those of marsupials.

If the spermatozoal

plesiomorphy of eutherians

were

extrapolated

to their phylogenetic

position relative

marsupials,

reseatchers

would be presented with the

profound irnplication that eutherians with pholidotelike

sperm preceded

origin of the Marsupalia. This would

further imply that yolk sac

placentation and premature

pafiurition in rnarsupials

were secondary

conditions or

that the allantoic placenta of eutherians

has originated

more than once. This heuristic outcome

of spermatozoal

studies

deserves further consideration.

However,

the

existence

of a sperm-type

with an elongated,

cylindrical

nucleus and capping acrosome

in ancestral

marsupials

remains

a possibility. The fact that an autosomal

marsu-

pial gene appears to have become an XJinked gene in

eutherians

(Fitzgerald

et al. 1993)

argues for the derived

status

of the Eutheria.

Acknowledgements

This work was made possible by a grant from the

Australian

Research

Council. David Scheltinga

thanked

for excellent

technical and

other assistance.

Author Update

Since

this chapter

was written, a well developed

fiber at 8 has been demonstrated

for the primitive fiog

Leiopelma Inchstetteri. Therefore, absence

of a fiber at

8 cannot be considered a synapomorphy

of the Anura

(DM Scheltinga,

BGM Jamieson,

K Eggars and DM

Green.

unoublished

data).

References

Afzelius BA, Munay A. The acrosomal reaction oi spernatozoa

dur-

ing fertilization or treatment

with egg water. Exp Cell Res 1957;

12:325-33'l

Asa C, Phillips DM, Siover J. Ultrastructure

ol spermatozoa

of the

crested tinamou.

J Uhasffuct Res 19861

94:170-175.

Asa CS, Phillips DM. Ulffasfiucture of avian spematozoat a shoft

review. In: (Mohd H, ed) New Horizons in Sperm Cell Research.

Tokyo.A{ew York: Japan Sci Soc Press, Gordon and Breach Sci

Publ, 1987,

pp365-373.

Austin CR, Baker CL. Spematozoon of Pseudobranchus striatus

a.rantlras.

J Reprod FerrlI 1964t'7.1,23-125.

Baccetti B, Al2elius BA. The Biology of the Sperm Cell. Karger,

Basel, 1976,

pp254.

Baccetti B, Burrini AG, Falchetti E. Spermatozoa

and relationships

in Palaeognath

birds. Biol Cell 1991; 71:209-216.

Baker CL. Spermatozoa of Amphiumae: spermateleosis,

helical

motility and reversibility. J Tennessee Acad Sci 1962; 37r23-39.

Baker CL. Spermatozoa

and spermateleosis

in Cryptobrancltus ard

Necluru.r. J

Tennessee Acad Sci 1963;

38:l-11.

Baker CL. SpennaLozoa

and spermateleosis

in the Salamandridae

with electron microscopy ol Dienict|ltts. J Tennessee

Acad Sci

1966;41:2-25.

Ballowitz E. Die form und sffuktur der schuppelltierspermien.

Zeit

Wiss Zool 1907: 86:619-625.

Barker KR, Biesele JJ. Spemateleosis of a salafiandet Amphiuma

tridactylum Cuvier: a correlated light and electron microscope

study. Cellule

1967; 67:90-l

18.

Bedford JM, Rifkin JM. An evolutionary view of the male leproduc-

tive ffact and sperm maturation in a monotreme mammal-the

Echidna lac,irlglosra

s aclieatus. Am I Ati'at 19'79, 156:207

'230.

Benardini G, Stipani GR, Melone G. The uhasffucttte of Xenopus

spematozoon. J

Ultrastruct Res

1986;

94t188-194.

Bernardini G, Andrietti F, Camatini M, Cosson MP. xenopus sper-

matozoon: correlation

between shape

and motility. Gam Res 1988;

20:165-175.

Bernardini G, Podini P, Maci R, Camatini M. Spermiogenesis

Xenopus Laevis:

from late spermatids

to spennatozoa.

Mol Reprod

Dev 1990; 26:347-355.

Boisson C. La spermiogenCse

de Protopterus anne(/etr (Dipneuste)

du Sdn6gal

6tudi6e au microscope optique et quelques

d6tails au

microscope

Electronique. Ann Fac Sci Uriv Dakar 1963; t0:43-72.

Boisson C, Mattei C, Mattei X. TroisiCme note sur la spermiogendse

de Protopterus owrectens (Dipneoste) du S6n6gal.

Institut Fond

Afr Noire Bull 56r A (Sci

Nat) 1967:29:1097

1121.

Brandon RA, Martan J, Wortham JWE, Englert DC. The influerice of

interspecific hybridization on the morphology of the spematozoa

of Ambystoma

(Caudata,

Ambystomatidae).

J Reprod Fertil 1974;

4l t2'7 5-284.

Breed WG. Evolution of the spematozoon in Australasian rodents.

Austral J Zool 199'1; 45:459-478.

Butler RD, Gabri MS. Structure and development of the sperm head

in the llzard Podarcis

(= l,<tcerta) tdarlcd. J Ultrastruct Res 1984;

88:261-274.

Carcupino

M, Corso G,

Pala M. Spermiogenesis i\ Chalcides ocella-

tus tiligugu (Gmelin) (Squamata,

Scincidae): an electon micro-

scope study. Boll Zool 1989, 56:119-124.

Carrick FN, Hughes RL. Aspects of the stuucture

and development

of monotreme

spermatozoa and their relevance to the evolution of

mammalian

sperm

morphology.

Cell Tissue Res 1982; 222:121-

141.

CheII GN, Clark WHJ. An acrosome reaction in sperm from the

white sturgeon Acipenser transmontanus.

J Exp Zool 1984;

232:129-139 .

Detlaf TA, Ginzburg AS. Acrosome reaction in sturgeons and the

role of calcium ions in the union of gametes.

Dokl Akad Nauk

SSSR lq6l; | 53:1461

1464.

Favard P. Mise en dvidence d'une sdcrdtion acrosomioue avant la

fdcondation chez les spermatozoides de Disgogloss,s piclas

Otth.

et de Ratxa tetuporaria L. C R Acad

Sci (Paris)

1955a;

240:2563-

2565.

Favard P. Spermatogendse de Discoglossus pictus

Otth.: 6tude

cytologique-maturation du spermatozoide.

Ann Sci Nat Zool

1955b; ll:369-394.

Fawcett DW. A comparative view of sperm ultrastructure. Biol

Reprod 1970; 2 (Suppl):90-127.

Fawcett DW. The mammalian spermatozoon.

Dev Biol 1975:

44:394-436.

Fawcett DW, Phillips DM. Recent observations

on the ultrastructure

and development of mammalian spermatozoa. In: (Baccetti

B, ed)

Comparative Spermatology. Rome: Accademia Nazionale dei

Lincei, 1970,

ppl3-28.

Fawcett DW, Anderson WA, Phillips DM. Morphogenetic factors

influencing the shape of the sperm head. Dey

BioI 1971,261220-

251.

Fitzgerald J, Wilcox SA, Graves JAM, Dahl HHM. A Eutherian X-

linled gene,

PDHAI, is autosomal

in marsupials: a model for the

evolution of a second, testis-specific

variant in Euthedan mam,

mals. Genomics 1993l' 18:636-642.

Folliot R. Ultrastructural study of spermiogenesis

of the anuran

amphibian Bombina variegata. In: (Fawcett

DW, Bedford

JM,

eds) The Spermatozoon. B

altimore-Munich: Urban and

Schwarzenberg, 19'7 9, pp333

^339.

Fukuyama K, Miyazaki K, Kusano

T. Spematozoa and

breeding

systems in Japanese anuran

species with special reference

to the

spiral shape of sperm in foam-nesting

rhacophorid

species.

2nd

World Congress of Herpetology. Adelaide,

South Australia. 1993-

6: Abshacts:92-93.

Furieri P. Prime osservazioni al microscoDio

elettronico sullo sDer-

matozoo di lrhurus cristatus camyex rLaurenti.l:

studio al miiro-

scopio elettronico. Boll Soc Ital Biol Sperim 1960; 36:1006-1009.

Furieri P. Osservazioni sullo spermatozoo

di Tritutus cristatus

carnifex (Laventi); studio al microscopio

elettronico. Monit Zool

Ital 1962. 68:90-1O2.

Furied P. Sperm morphology

of some reptiles:

Squamata and

Chelonia. In: (Baccetti

B, ed)

Comparative Spermatology. Rome:

Accademia

Nazionale dei Lincei, 1970,

1 15-

131.

Furieri P. La morfologia

degli spermi di alcuni

anfibi anud. Boll

Zool l9'121391618.

Furied P. La mofologia comparata

degli spermi di Discoglossus

VsRrpsnAre SpBma PnvlocBNv / 329

pictus Otth., Bombina variegata (L.) e Alytes obstetricans

fl-aurenti). Boll Zool l9'15a:. 42t458-459.

Furieri P. The peculiar morphology of the spematozoon of Bombina

yarie

gata (L.). Morlir Zool Ital I 975b; 9:

I 85-20

I .

Grass6 PP. La spematogendse. In: (Grass6

PP, Delsol M, eds) Traite

de Zoologie: Anatomie, Syst6matique, Biologie. Vol. l4:

Batraciens. Apareil uro-genital (suite) +mbryogdndse, dthologie,

odgine, evolution, systematique. Fascicule lB Pads: Masson,

1986,

ppl-20.

Harding HR, Carrick FN, Shorey CD. Special

features

of sperm

stucture and function in marsupials. In: (Fawcett

DW, Bedford

JM, eds) The Spermatozoon.

Baltimore-Munich:

Urban

and

Schwarzenberg,

197

9, pp289

-303.

Harding HR, Aplin K, Shorey CD: Parsimony analysis of marsupial

sperm structure: a preliminary report. In: (Mohri H, ed) New

Horizons in Sperm Cell Research. Tokyo/New York: Japan Sci

Soc

Prcss,

Gordon

and Breach

Sci

Publ, 1987,

pp375-385.

Healy

JM. Sperm

morphology

and

its systematic importance in the

Gastopoda. Malacol Rev 1988; 4 (Suppl):251-266.

Healy JM. Euspermatozoa and

paraspermatozoa

in the trochoid

gas-

topod Zalipais laseroni (Trochoidea:

Skeneidae).

Mar Biol 1990;

lo5t49'7

-5O'1

Healy

JM, Jamieson

BGM.

Ultrastructure of the spermatozoon of the

t\rat[a (Sphenodon

punctatus)

and its relevance to the relation-

ships of the Sphenodontida. Phil Trans R Soc Lond Biol Sci B

1992. 335:193-205 .

Healy JM, Jamieson BGM. The ulhastucture of spermatogenesis

and epididymal spermatozoa of the tuatara Sphenodon

punctatus

(Sphenodontida,

Amniota): Phil Trans R Soc Lond B Biol Sci B

1994: 344:18'7

-199

Hedges SB, Moberg KD, Maxson LR. Tetrapod

phylogeny

infened

from 18S and 28S ribosomal RNA sequences and a review of the

evidence for amniote relationships. Mol Biol Evol 19901 7:607-

633.

Henley

C, Feduccia A, Costello DP. Oscine spermatozoa: a light and

electron-microscopy

study. Condor 1978; 80:41-48.

Hess

RA, Thurston RJ,

Gist

DH. Ultrastuctwe of the turtle sperma-

lozoon. Anar Rec l99l:. 229:47

3-481.

Hillis DM. The

phylogeny

of amphibians: current knowledge and the

role of cytogenetics. In: (Green

DM, Sessions S, eds) Amphibian

Cytogenetics and Evolution. New York: Academic Press, 1991,

pp7-31.

Hillis DM, Dixon MT, Ammerman LK. The relationships

of the

coelaaafih

Latimetia chalumnae: evidence from sequences

of ver-

tebrate 28S ribosomal

RNA genes.

Environ Biol Fishes 1991;

321119-130.

Hughes RL. Compamtive morphology of spermatozoa from five

marsupial

families. Austral J Zool 1965; 13t533-543.

Humphreys PN. Brief observations

on the semen and spermatozoa of

certain

passedne

and non-passerine

birds. J Reprod Fefill 19'121

29:327

-336.

James WS. The Ultrastructure

of Anuran Spermatids and

Spermatozoa. Ph.D. thesis 1970,

University ofTennessee, USA.

Jamieson BGM. Fish Evolution and Systematicst Evidence from

Spermatozoa. Carnbridge University Press,

Cambridge

(UK),

1991.

Jamieson BGM. Evolution of tetrapod

spermatozoa with particular

rcference to amniotes. M6m Mus Natl Hist Nat 1995a], 166:.343-

358.

Jamieson BGM. The ultrastucturc

of spematozoa of the

Squamata

(Reptilia)

with phylogenetic

considerations. M6m Mus Nad Hist

Nat 1 995b: 166:359

-383.

Jamieson BGM, Healy JM. The

phylogenetic

position

of rhe tuatara,

Sphenodon

(Sphenodontida,

Amniota), as indicated

by cladistic

analysis of the ultrastructure

of spermatozoa. Phil Trans

R Soc

330 / Cs.qPren

Lond Biol Sci B 1992:335:20'7

-219.

Jaoieson BGM, Koehler L. The ultrastruc re of the spernatozoon

of the Northem Water Snake,

Nertttlia sipedon (Colubridae'

Sementes).

Can J Zool 19951'7211648

1652'

Jamie.son

BGM. Scheltinga

DM The ultrastructurc of spermatozoa

of Nangura

rpinora (Siincidae'

Reptilia) Mern Qld Mus 1993;

34:169 119.

Jamieson

BCM, Scheltinga

DM The ultrastructure

ol spermatozoa

of the Australian skinks, Ctenolr.r tdeni'i(ttlts' Carlia pectordlis

ald.

Tiliqua scincoide,s

scincoitles

(Scincidae.

Reptilia) Mem Qld

Mus

1994;37:181-193

Jamieson

BGM, Lee MSY, Long K Ultrastructule of the spermato-

zoon of the intemally l'ertilizing ftog ,4sc.lpirrs

t'aei (Ascaphidae:

Anura: Amphibia) wiih phylogenetic considerations'

Hemetologica

1993;

49:52-65

Jamieson

BGM, Koehler L, Todcl

BJ Spermatozoal

ultrastructure

tlnee species

of parrots

(Aves' Psittacifbrmes)

and its phylogenetic

implic;dons.

Anat

Rec 1995; 241:46i 468'

Jamiison BGM, Oliver SC, Scheltinga

DM The ultrastructure

ol the

sDermatozoa

ol \qurmalx l scincidce.

Cekkonidae

9nd

Pygopodidue

rRepriiiat. Actc

Zool

{CopenhirPeni

lqq6:

77r85

100.

Jamieson

BGM, Scheltinga

DM, Tucker AD. The ultrastructure

sDermatozoa

of the Australian freshwater crocodile' Crocod]/'l'l

i;lnstoni Ktefft, 1873

(Crocodylidae' Reptilia) J Submicrosc

Cvtol Pathol 199'7:. 29:265-274.

lesperren

{. Fine

slructure of the (permaloToon

ol lhe Auslralian

Lunglr\h

N.,o,

. rdlrd

]ir\tc

ri 'Krelltr.

I I lrra'lrucl

Re5

lqTl:

37:178-

185.

Koehler LD. Diversity of avian spermatozoa

ultrastructure

with

emphasis

on the members

of the order Passeriformes M€m Mus

Natl Hist Nat 1995:-

166:43'7'444.

Kondo T. Hasegawa K, Uchida T. Formalion of the microtubule

bunclle

and helical shaping

of the spermatid

in the common fil1ch'

ktnchura striata var. clomestica-

J Ultrastruct Res 1988i 981158

168.

Kwon AS. Lee YH. Comparative

spermatology

of anurans

with spe-

cial references

to phylogeny. In: (Jamieson BGM. Ausio J. Justine

J-L, eds)

Advances

in Spermatozoal

Phylogeny

and

Taxonomy'

vol. 166.

Paris:

M6m Mus Natn His Nat' 1995,

pp32l-332

Lee MSY, Jamieson

BGM. The ultrastructure

of the spematozoa of

three species

of myobatrachid

frogs (Anura. Amphibia) with phy

logeneiic

considerations.

Acta Zool (Stockholm) 1992:. 73:213-

222.

Lee MSY, Jamieson

BGM The ultrastructure

of the spermatozoa

bufonid and hylid frogs (Anura. Amphibia): lmplications for phy-

logeny

and

fertilization

biology Zool Scr 1993;22:309-323

feuig if. Fish Spermatology: Ultrastructure'

phylogeny and cryo-

oreiervrtion.

Honours

thesi..

I087.

Uni\er.lt) of Queen'land

LJr-rng

LKP. Cr-rmmins

JM Morphologl

of immarure

'pennrto/oa

rhe{hinese Pangolin (Manus penta(ktctlld: Pholidota) Proc Aust

Soc Reprod

Biol (Newcastle,

Australia)'

20th Ann Conf' 19881

p94.

Mainoya JR. Observations

on the ultrastructure

of spermatids

in the

Elrjsid;

ol Chiromantis xetampelina (Anura: Rhacophoridae) Afric

J Ecol

1981;

19:365-368.

Martan J, Worlham E A tail membmne

on the spennatozoa

of some

ambystomatid

salamanders.

Anat Rec 19'72:.

l-12:460'

Matos"E,

Azevedo C. Ulffastructural

sludy of the spermatogenesis

Lepidosiren pttradoxa (Pisces' Dipnoi) in Amazon region Rev

Bras Ci6n Mortbl 1989t 6:6'7

-'7

Mattei X. Spermiogendse

compar6e

des

poissons ln: (Baccetti

ed) Comparative

Sperrnatology New York: Academic Press'

1970,

pp57-69.

Mattei X. The flagellar apparatus

of spelnratozoa

in fish: ultrastruc-

ture and

evolution.

Biol Cell 1988;63:l5l 158

Mattei X. Spennatozoon

ultrasttucture and its systematic implica-

tions

in flshes.

Can

J Zool l99l ; 69:3038-3055

Mattei X, Siau Y, Seret B. 6tude ultrastructurale

du spermatozold€

du coelacanthe:

Lati eritt clltllutlna€ J Ultrastruct

Res 1988;

101:243-251.

McGregor JH. The spermatogenesis

ol Atrphitttfia J Morphol 1899;

| 5:57-

104.

Metter DE. On breeding and sperm retention in Ascaphus Copeta

1964l-

1964:1 l0-1 | | .

Meyer A, Dolven SI. Molecules. Iossils' and the origin of tetrapods

Mol

Evol

1992;35:102

l13.

Mever A. Wilson \C Origin ol lelrapodc

inlerred

from their

milo-

cLondrial

DNA cllilialion

lo lungti'h

J Mol Evol la90:31:J5q

364.

Meyer E, Jamieson

BGM, Scheltinga

DM. Sperm ulfrastructure

six Australian hylid frogs from two genera

(lilori4 and

Clclor(rno)t phylogenetic

implications.

J Submicrosc

Cytol Pathol

199'l

: 29:443-451

Mizuhira V, Futaesaku

Y, Ono M, Ueno M' Yokotujjta J' Oka T'

The fine structure

of the spermatozoa

ol two species

Rhacophorus

(arboreus, sc hleBelii) I Phase-contnst

microscope'

scanning

elecffon microscope. and cytochemical observations

the

head

piece. J Ultrastruct

Res

1986:

96:41-53

Mo H. Ul;astructural studies

on the spennatozoa

of the frog RdDu

nigromactrlata and the toad Bulb btlfo osiaticus Zool Res 1985;

6:381-390.

Oka T. Ultrastructural

observatioos

on the sperm

ln a lrog'

Rhlcophorus st:hlegelii.

Japan

J Herpetd 1980;

8:137

Oliver SC, Jarnieson

BGM, Scheltinga

DM. The ultraslructure

sDermalozoa

of squamata:

II. Aganridae' Vamnidae, Colubridae'

dlapidae,

and Boidae (Reptilia) Helpetologica 1996:

52:216-241

Phillips DM. Ultrastructure

of spermatozoa

ol the woolly opossum

Caiuron,-s

philtnu:ler.

J Ultrastruct

Res

1970; 33:381

397

Phillips DM. Asa CS. Development

ot'

spermatozoa

in the Rhea'

Anat

Rec 1989:

223:276-282.

Phillips DM, Asa CS. Strategies

for fbrmation of the midpiece ln:

(Baccetti B, ecl) Comparative Sperm^tology 20 Years After" Vol'

75. Serono

Symposia

Fublications New York: Raven

Press.

1993'

pp997- 1000.

Phiilips DM, Asa CS, Stover J Ultrastructure

of spermatozoa

of the

white-naped

crane. J Submicrosc

Cytol 1987; l9:489-49'1

Picheral

i. Structure

et organisalion

du spermatozoide

Pleurotleles

ndillil Michah (Amphibien UrodEle) Arcb Biol'

(Lidge)

1967:

78i 193-221

Picher;l B. Les 6l6meits cyloplasmiques au couts de la spermlo-

genese

du triton Pleurodeles traillii Michah IlI L'dvolution des

iormations caudales.

Z Zellforsch 1972; I 3 I :399-416'

Picheral B. Structural. comparative.

and functional aspects

of sper-

matozoa

in urodeles.

In: (Fawcett

DW, Bedtbtd JM. eds) The

Spermatozoon.

Baltimore: Urban and Schwarzenberg,

1979'

pp267

-281

Picheral

B. Folliot R. Maillet PL Sur la structure

du noyau du spet

matozoide de Pletrrttrleles

l'al ii Michah (Amphibien Uroddle)'

C R S6anc

Acad Sci t966iD 262.15'19'1582'

Poirier GR, Spink GC The ultrastructure

of testicular spermatozoa

t\ o species

R,Ir?rl.

J Ultrr\lrucl

Res 197l:

t6:455-465

Pugin.Rio.

E. Erude

comparltire

\ur lr 'tructule

'permatozoide

ies Amphibiens Anourls. Comportement des gamEles lors de la

ldcondation.

Thdse 1980,

L'Universitd de Rennes.

France

Purkerson

ML. Jarvis JUM, Luse SA' Dempsey

EW X ray analysis

coupled

with scanning

and transmission

electron

microscopic

obsirvations of spermatozoa

of the African lulgflsh' Ptotopter s

apthnPi'

trs. J

Zool

lqTl: l7l:l-ll

Reed Si, Stanley

HP. Fine structure

of spermatogenesis

in the Soxth

Atiican Clawid '[o^d, Xe]1oPus

ldevi,t Daudin J Ultrastruct Res

l9'7

2, 4l:21'7

'295

Retzius

G. Die spermien der amphibien.

Biol Utrtersuch, N F 1906;

13:49-'10.

Robson SK, Rouse Gw, Pettigrew JD. Sperm ultrastructure of

Tarsius bancanus

(Tarsiidae,

Primates)r implications for pdmate

phylogeny

and

the use of spenn in systematics. Acta Zool 19971

'18:269-2'18.

Rouse GW,

Robson SK. An ultrastuctural study of megachircpteran

(Mammalia:

Chiroptera) spermatozoa:

implications for chiropteran

phylogeny. J Submicrosc Cytol 1986;

18:13'7-152.

Rowe T. Definition, diagnosis, and origin of Mammalia.

J Vert

Paleontol

1988:

8t241-264.

Russell LD, Brandon

RA, Zalisko EJ, Martan J. Spematophores of

the salamander AmDysloma

texanum. Tissue Cell 1981: 13:609-

621.

Saita

A, Longo OM, Tripepe S. Osservazioni

comparative sulla

spermiogenesi.

IIL Aspetti ultrastrutturali

della spermiogenesi di

Jacana

jacana (Caradriformes).

Accad Naz Lincei. 1983;'14:417-

430.

Saita

A, Comazzi M, Perrotta E. Electron

microscope study of

spermiogenesis

rn Caiman crocotlylus L. Boll Zool 198'7;'4:3O7-

318.

Sandoz D. Etude cytochimique des

polysaccharides

au cours de la

spermatogenbse

d'un amphibien anoure: le discoglosse

Discoglossus

pictus

(Otth.).

Microscopie l9'7 0a:, 9:243-262.

Sandoz

D. Etude ultrastructurale

et cytochimique de la formation de

l'acrosome

du discoglosse

(Amphibjen

Anoure).

In: (Baccetti

ed) Comparative

Spermatology. Rome:

Accademia Nazionale dei

Lincei,

1970b,

pp93-113.

Sandoz

D. Participation du r6ticulum endoplasmique a l'6laboration

de l'anneau dans les spermatides du discoglosse

(Amphibien

,{noure). J Microscopie 1973; 17:185-198.

Sandoz

D. Development of the neck region and the ring dudng

spermiogenesis of Discoglossus

piclr,r (Anuran

Amphibia). In:

(Afzelius

BA, ed) The Functional

Anatomy

of the Spermatozoon.

Oxford: Pergamon,

197 4a, pp231

-241

Sandoz D. Modifications in the nuclear e[velope during spermiogen-

esis of Di.rcoglo,rsrs

piclus (Anuran

Amphibia). J Submicrosc

Cytol

l9'1 4b

| 6:399

-4

19.

Serra

JA, Vicente MJ. New stuctures of spermatozoa of Rana in

relation to lipid localization. Revist Portug Zool Biol Gen 1960;

):223-242.

Seshachar

BR. The spermatogenesl,s of Uraeotyphlus naraya i

Seshachar. Cellule

1939

| 48:63-7

Seshachar

BR. The apodan sperm.

Cunent Sc:t

1940t 10:464 465.

Seshachar

BR. Stages ifl the spermatogenesis of Siphonops

annula-

rrr Mikan. and Dermophis

gregorii Blgt. (Amphibia:

Apoda).

Proc Indian Acad Sci 1942i 15:266-2'77.

Seshachar BR. The spermatogenesis of lchthyophis

glutinoszs

Linn.

Part III. Spermateleosis. Proc Natl Institut Sci, India 19431

9:271-

286.

Seshachar BR. Spermateleosis in Uraeotyphlus narayani Seshachat

and Gegenophis camosus Beddor\e

(Apoda).

Proc Natl Institut

Sci. India 1945: l1:336-340.

Sibley CG, Ahlquist JE. Phylogeny

and Classificatior of Birds:

A Study in Molecular Evolution.

Yaie University Press, New

Haven. Conn. 1990.

Sibley CG, A} quist

JE,

Monroe

BL. A classification

of the living

birds of the world based

DNA-DNA hybddizarion

studies. Auk

1988i 105:409-423.

Soley JT. Ultrastructure of ost ch (Struthio

camelus) spetmatozoa:

l. Transmission

electron microscopy,

Onderstepoort J Vet Res

1993:60:119-130.

Soley JT. Centdole development

and lbrmation of the flagellum dur-

ing spermiogenesis in the ostrich

(Struthio

c1melus). J Anat

1994:,

185:301-313.

VBnrsgnAre

Sprnru PHvr-ocel{v / 331

Swan MA. Linck RW.

Ito S.

Fawcett DW. Stucture and

function of

the undulating membrane in spermatozoan

propulsion in the toad

Bufo marinus. J Cell Biol 1980i

85:866-880.

Temple-Smith P. Sperm structure

and marsupial

phylogeny.

In:

(Archer

M, ed)

Possums and opossums: studies

in evolution.

Sydney: Surrey Beatty & Sons

and the Royal Society of New

South Wales,

1987,

pp171-193.

Temple-Smith PD, Bedford JM. Sperm

maturation and the formation

of sperm

pairs

in the epididymis of the opossum,

Didelphis rir-

giniana.

I Exp Btol 19

80:

214: 16 1- 17 1 .

Thurston zu, Hess

RA. Ultrastructure

of spermatozoa from domesti-

cated birds: comparative study of turkey, chicken and

guinea

fowl.

Scan

Microsc 1987; l:1829-1838.

Townsend DS, Stewart

MM, PoughFH, Brussard PF.

Intemal fefiili-

sation

in an

oviparous frog. Science 1981;212:469-4'10.

Tuzet O, Millot J.

La spermatogenase de ktimeria chalumnae Smith

(Crossopt4rygien

coelacanthidl).

Annal Sci Nat Zool 1959; l:61-

69.

van der Horst G. Spematozoon

structure of three anuran

(Amphibia)

species. Proc Electron

Microsc

Soc