ArticlePDF Available

A Baby Sea-Serpent No More: Reinterpreting Hagelund’s Juvenile “Cadborosaur” Report


Abstract and Figures

Eyewitness reports and cultural representations have been interpreted by some researchers to suggest the existence of a large, long-bodied marine vertebrate in the northeast Pacifi c. Dubbed “Caddy” or “Cadborosaurus” (after Cadboro Bay, British Columbia), it was formally named and described as Cadborosaurus willsi by Bousfi eld and LeBlond in 1995. Among the supposedly most informative accounts is the alleged 1968 capture of a juvenile by William Hagelund, detailed in his 1987 book Whalers No More. Reportedly morphologically similar to adult “Cadborosaurs,” the specimen was comparatively tiny, and apparently precocial. Bousfield and LeBlond argue that this strongly supports their contention that “Caddy” is reptilian (juvenile reptiles are typically precocial, recalling “miniature adults” in both behavior and morphology). Anomalous traits suggest some degree of misrecollection in Hagelund’s account, furthermore a quantitative analysis of the similarity of 14 candidate identities with the specimen indicates that it most strongly resembles the bay pipefi sh (Syngnathus leptorhynchus)—far more so than a cryptid or reptile. While this detracts from the plausibility of the cryptid, the reidentification of this particular specimen does not discount the data as a whole nor does it suggest that all “Caddy” reports are necessarily of known fish. We contend that the “reptilian hypothesis” does, however, need to be seriously re-examined in light of the removal of a strong piece of evidence.
Content may be subject to copyright.
A Baby Sea-Serpent No More:
Reinterpreting Hagelund’s Juvenile “Cadborosaur” Report
School of Biological Sciences, Royal Holloway, University of London,
Egham, Surrey, TW20 0EX, UK
School of Earth and Environmental Sciences,
University of Portsmouth, Portsmouth, PO1 3QL, UK
Submitted 10/11/2010; Accepted 4/10/2011
Abstract—Eyewitness reports and cultural representations have been inter-
preted by some researchers to suggest the existence of a large, long-bodied ma-
rine vertebrate in the northeast Paci c. Dubbed “Caddy” or “Cadborosaurus”
(after Cadboro Bay, British Columbia), it was formally named and described
as Cadborosaurus willsi by Bous eld and LeBlond in 1995. Among the sup-
posedly most informative accounts is the alleged 1968 capture of a juvenile by
William Hagelund, detailed in his 1987 book Whalers No More. Reportedly
morphologically similar to adult “Cadborosaurs,” the specimen was com-
paratively tiny, and apparently precocial. Bous eld and LeBlond argue that
this strongly supports their contention that “Caddy” is reptilian (juvenile
reptiles are typically precocial, recalling “miniature adults” in both behavior
and morphology). Anomalous traits suggest some degree of misrecollection
in Hagelund’s account, furthermore a quantitative analysis of the similarity
of 14 candidate identities with the specimen indicates that it most strongly
resembles the bay pipe sh (Syngnathus leptorhynchus)—far more so than a
cryptid or reptile. While this detracts from the plausibility of the cryptid, the
re-identi cation of this particular specimen does not discount the data as a
whole nor does it suggest that all “Caddy” reports are necessarily of known
sh. We contend that the “reptilian hypothesis” does, however, need to be
seriously re-examined in light of the removal of a strong piece of evidence.
Keywords: bay pipe sh—Syngnathus leptorhynchus—sturgeon poacher—
Podothecus accipenserinus—Cryptozoology—Cadborosaurus
willsi—reptile—William Hagelund
Journal of Scienti c Exploration, Vol. 25, No. 3, pp. 497–514, 2011 0892-3310/11
498 M. A. Woodley, D. Naish, C. A. McCormick
It is not accurate to presuppose that technically quali ed scientists are
uninterested in examining cryptozoological evidence. Several workers have
assessed such data and reported their ndings in the peer-reviewed technical
literature (e.g., Scott and Rines, 1975, Raynal & Sylvestre, 1991, Paxton et
al., 2005, Paxton & Holland, 2005, Paxton, 2009). Critical discussions of
large undescribed species are not necessarily outlandish, as numerous marine
vertebrate megafaunal species (>45 kg) have been discovered and described
in recent years (e.g., Megamouth shark Megachasma pelagios Taylor et al.,
1983, Bandolero beaked whale Mesoplodon peruvianus Reyes et al., 1991,
West African skate Bathyraja hesperafricana Stehmann, 1995, Indonesian
coelacanth Latimeria menadoensis Pouyaud et al., 1999), and statistical work
suggests that several such species remain to be documented (Paxton, 1998,
2001, Raynal, 2001, Solow & Smith, 2005, Woodley et al., 2008). Large marine
vertebrates can be surprisingly cryptic due to rarity, habitat, and/or avoidance
of vessels (Heyning, 1989); Megachasma pelagios was not recorded from
the Atlantic until 1995 (Amorim et al., 2000); the family Ziphiidae (“beaked
whales”) includes numerous poorly known species, including Mesoplodon
traversii, which is known from three partial skulls and has no live records (van
Helden et al., 2002).
Between the early 1990s and the present (see also LeBlond & Sibert,
1973), Edward L. Bous eld and Paul H. LeBlond have collected and analyzed
aquatic cryptid reports from the northeast Paci c Ocean (Bous eld & LeBlond,
1992, 1995, LeBlond & Bous eld, 1995, LeBlond, 2001). These authors have
speculated on the existence of a large, long-bodied marine vertebrate species
in the northeast Paci c Ocean based on their interpretations of eyewitness
reports and local cultural representations. Purportedly, the anecdotal evidence
generally describes an animal with a length of 5 –15 meters, a serpentine body,
undulatory locomotion, sheep/horse/giraffe/camel-like head, a neck 1–4 meters
in length, anterior ippers, posterior ippers either absent or fused with the
body, top of the tail “toothed” or spiny, and a tail which is horizontally split
or uke-like; less frequently reported characteristics include whiskers, large
eyes, “colored” eyes, ears, and/or horns; the color is variously described as
orange, green, brown, grey, black, and “gun metal” blue; the skin has been
described as smooth, although sometimes fur/fuzz/hair on the neck or body is
reported; the back has been variably described as serrated and smooth (LeBlond
& Bous eld 1995). This cryptid was informally dubbed “Cadborosaurus” (after
Cadboro Bay, British Columbia) by journalist Archie H. Wills in 1933 and was
formally, albeit controversially, described as Cadborosaurus willsi by Bous eld
and LeBlond in 1995.
A Baby Sea-Serpent No More 499
A number of speculative inferences and suggestions led Bous eld and
LeBlond (1995) to promote a hypothesis of the putative taxon’s phylogenetic
af nities. The long, slender body reported for C. willsi was argued by Bous eld
and LeBlond (1995) to be most compatible with a poikilothermic physiology,
and hence with a reptilian identity (Bous eld & LeBlond, 1995:9). It was also
suggested that the existence of small, precocial “Cadborosaurus” juveniles
evidenced reptilian identity since the juveniles of mammals generally undergo
an extended period of parental care in which they are incapable of surviving
by themselves. By comparing the inferred morphology of an enigmatic carcass
from Naden Harbour, British Columbia, with that of various extinct and
extant vertebrates, they concluded that the closest similarities were with the
extinct reptiles of the clade Sauropterygia, and speci cally with plesiosaurs.
Sauropterygians are otherwise known only from the Mesozoic, and there is no
evidence from the fossil record that they survived beyond the end of the Late
Cretaceous (Lucas & Reynolds, 1993). Bous eld and LeBlond’s hypothesis
has been criticized on the grounds that eyewitness evidence used to support
the reality of C. willsi is more compatible with a mammalian identity than a
reptilian one, as the putative animal inhabits cold water, exhibits maximum
exibility in its vertical plane, and is sometimes described as possessing hair
(Saggese, 2006, Woodley, 2008, Woodley et al., 2008). Now, further analysis
has demonstrated that one of the best-described juvenile “cadborosaurs” cannot
be considered reptilian and most likely represents a known, albeit unfamiliar,
species of sh.
Captain William Hagelund’s Account
In August 1968 while at Pirate’s Cove, De Courcy Island, British Columbia,
former whaler Captain William Hagelund gave the following description of an
animal he could not identify; the account is taken from Hagelund (1987:177–
With my two sons and their grandfather aboard our centre cockpit sloop, we
spotted a small surface disturbance in the calm anchorage where we had dropped
the hook for the night. Lowering the dinghy, my youngest son Gerry and I rowed
out to investigate. We found a small, eel-like, sea creature swimming along with
its head held completely out of the water, the undulation of its long, slender body
causing portions of its spine to break the surface. My rst thought that it was
a sea snake was quickly discarded when, on drawing closer, I noticed the dark
limpid eyes, large in proportion to the slender head, which had given it a seal-like
appearance when viewed from the front. When it turned away, a long, slightly
hooked snout could be discerned.
As the evening’s darkness made observation dif cult, and the swiftness
of the creature’s progress warned that he could quickly disappear, I decided to
attempt a capture and bring it aboard the sloop for closer examination. Reaching
500 M. A. Woodley, D. Naish, C. A. McCormick
out with a small dip net as Gerry swung the stern of our dinghy into the path of
the small vee of wavelets that were the only indication of the creature’s position,
I was pleased to nd him twisting angrily in the net when I lifted it up.
Under the bright lights aboard the sloop, we examined our catch and found
he was approximately sixteen inches long, and just over an inch in diameter. His
lower jaw had a set of sharp tiny teeth and his back was protected by plate-like
scales, while his undersides were covered in a soft yellow fuzz. A pair of small,
ipper-like feet protruded from his shoulder area, and a spade-shaped tail proved
to be two tiny ipper-like ns that overlapped each other.
I felt the biological people at Departure Bay would be interested in this nd,
but without a radiophone to contact them, the next best thing was to sail up there
in the morning. Agreeing on this, we lled a large plastic bucket with seawater
and dumped our creature into it. We retired early, for I intended to leave at rst
light, but sleep would not come to me. Instead, I lay awake, acutely aware of the
little creature trapped in our bucket. In the stillness of the anchorage I could hear
the splashes made by his tail, and the scratching of his little teeth and ippers as
he attempted to grasp the smooth surface of the bucket. Such exertion, I began to
realize, could cause him to perish before morning.
My uneasiness grew until I nally climbed back on deck and shone my
ashlight down into the bucket. He stopped swimming immediately, and faced
the light as though it were an enemy, his mouth opened slightly, the lips drawn
back exposing his teeth, and the tufts of whiskers standing stif y out from each
side of his snout, while his large eyes re ected the glare of my ashlight. I felt
a strong compassion for that little face staring up at me, so bravely awaiting its
Just as strongly came the feeling that, if he was as rare a creature as my
limited knowledge led me to believe, then the miracle of his being in Pirate’s
Cove at all should not be undone by my impulsive capture. He should be allowed
to go free, to survive, if possible, and to ful ll his purpose. If he were successful,
we could possibly see more of his kind, not less.
If he perished in my hands, he would only be a forgotten curiosity. I
lowered the bucket over the side and watched him swim quickly away into the
darkness, then returning to my bunk for a peaceful rest, my mind untroubled by
the encounter.
Hagelund included an illustration of his creature (reproduced in Figure 1)
with the following notes: hooked upper jaw, 3 inches (brackets include head),
large eyes, plate scales, black on top and brown on sides, ragged ends (arrow
pointing to tail), yellow tail, approximately 1 to 1½ inches in diameter, yellow
fuzz (pointing to belly), tiny teeth on both jaws, 16 inches (brackets including
head, body, and tail).
Hagelund’s account includes 24 reported traits: an eel-like appearance (= sea
snake–like; long, slender body), head held out of the water while swimming,
A Baby Sea-Serpent No More 501
undulatory movement, dark eyes, limpid eyes, large eyes, seal-like face,
slender head, slightly hooked snout, long snout, length of 16 inches (40 cm),
diameter of 1 to 1.5 inches (2.5–3.8 cm), tiny teeth in both jaws, plate-like
scales on the back, undersides with a soft yellow fuzz, ipper-like feet near
the shoulder, spade-shaped tail, tail composed of two overlapping ipper-like
ns, lips, whiskers, coloration of black on top and brown on the sides, yellow
tail, and a head length of 3 inches (7.6 cm). It is likely the ragged ends of the
tail mentioned in the illustration’s text represent damage rather than an actual
trait, which raises the possibility that “overlapping ipper-like ns” are also the
result of damage.
Our objective was to use these and other traits in comparing Hagelund’s
specimen with other plausible candidate species. In order for some of the traits
to be usable in comparison with known animals and proposed cryptids, some
needed to be more strictly de ned or modi ed. Body diameter and head length
were not used directly, but were rather modi ed into proportions relative to
the total length (i.e. including the tail appendage), giving a depth:TL of 1:16
to 1:10.7, and a head:TL of 1:5.33. “Eel-like” was taken to mean an elongate
and cylindrical body, and is not necessarily redundant in conjunction with the
aforementioned proportions, as eel-like animals may not be proportionally
similar and animals with similar head:TL proportions may not be eel-like.
For the given traits of “eel-like,” “large eyes,” “slightly hooked snout,” “long
snout,” and “slender head,” the illustrated morphology was used to determine
the threshold of these traits, e.g., if the eyes are at least the same size in
proportion they were considered “large.” While the account suggests that only
the back had plate-like scales, the illustration shows them covering most of
the animal’s side; due to this apparent contradiction, the trait was simpli ed
to the presence of plate-like scales. Since several of the candidates lack teeth
entirely, it was decided that the simple presence or absence of teeth would be
a more worthwhile trait rather than attempting to determine the threshold for
“tiny” teeth; specifying teeth in both jaws is unnecessary in this analysis as
none of the candidates, and few animals in general, have teeth in only one jaw.
The trait of “undulatory” movement unfortunately does not specify a plane, so
was interpreted as eel- or snake-like locomotion occurring either horizontally
or laterally. It is not clear if “slender head” refers to head depth or width, so the
illustrated depth was used as a guide but given low weight (see below).
For comparisons with sh candidates, the traits of dorsal n(s), pelvic
ns, and anal n were added to determine which candidates are the closest
t. The description of “ ipper-like feet” near the shoulder is confusing since
Hagelund’s illustration appears to show n rays typical for actinopterygian sh;
accordingly, this trait is regarded as synonymous with pectoral appendages of
any type.
502 M. A. Woodley, D. Naish, C. A. McCormick
Hagelund’s illustration does not clarify how dark the eyes are, so this
trait was assumed to indicate that the eyes were somewhat darker than the
surrounding body. It is hard to objectively de ne how eyes could be described
as “limpid,” although this trait was retained since it still has comparative value
with one aberrant report.
Due to the uncertainty of eyewitness observations, multiple character states
were used in our analysis. While a certain trait may be present (e.g., poacher
pelvic ns), there may be a plausible reason why it could be overlooked (e.g.,
the n could be folded and/or is small and transparent enough to be easily
overlooked). Additionally, a trait may not be present (e.g., tail composed of two
separate ippers in cutlass shes) but could be explained by the misinterpretation
of another trait (e.g., a strongly forked tail). This does not mean that overlooking
or inventing a trait is impossible; it is merely less likely than the suggestion that
these may be awed descriptions of real character states.
It is apparent that not every trait should carry the same comparative weight
due to both occasional nebulous descriptions and the potential for eyewitness
confusion: We therefore argue that traits should be sorted into different
categories based on their importance. Traits regarded as having high importance
are clearly described and dif cult to misinterpret: They involved such obvious
morphological traits as plate-like scales, eel-like appearance, teeth, head
held out of the water while swimming, black dorsal coloration, brown lateral
coloration, and yellow tail.
In total 29 morphological characters were used. A simple coding system was
developed to determine whether a particular trait was present or not. When a trait is
unambiguosly present, it is coded with P (for present), when a trait is occasionally
present due to individual variability, it is coded with an O (for occasionally
present). The use of A indicates the absence of a particular trait, whereas a
question mark indicates that it is not known whether or not this trait or something
that could be mistaken for it is present. As there is an element of subjectivity to
the inference of a particular trait, the addition of * next to a trait indicates that the
presence of the trait may be open to eyewitness interpretation. Each character was
ranked as either similar (s), in that it co-occurs (is either simultaneously present
or absent) in both the Hagelund specimen and candidate, somewhat similar (ss),
in that it or something that could be mistaken for it could co-occur, or dissimilar
(d) indicating no co-occurrence. Co-occurrent (similar) traits were awarded a
whole point. An occasionally present (somewhat similar) trait was given half a
point in all cases as the trait must be either present or absent at least some of the
time in the candidate. Stand-alone question marks (indicating unknown presence
or absence) were awarded zero points, and a co-occurrent score followed by *
(suggesting uncertainty) incurred a “subjectivity penalty” of a quarter of a point.
This gives rise to a simple formula with which the overall similarity with respect
to the Hagelund specimen could be ascertained for a given candidate:
A Baby Sea-Serpent No More 503
Similar scores (1 point) + Somewhat similar scores (0.5 point) − Subjectivity penalty (0.25 point)
= Candidate’s overall similarity score.
It was decided not to include a separate penalty for dissimilarity, as this
would have further complicated the measure. For example, calculating separate
dissimilarity scores and then subtracting them from the similarity score resulted
in negative scores in the case of some candidate identities. Furthermore,
dissimilarity scores were deemed to be less informative in instances where
there were large numbers of “unknown” character states which couldn’t be
honestly described as dissimilar (for example in the case of the Finn John
description there were 21 “uncertain” character states vs. 8 character states that
were unambiguously present, i.e. were “similar” with respect to the Hagelund
specimen and no unambiguously disimilar characters—in this instance
similarity scores are simply more informative). Therefore, while not perfect, the
similarity measure developed here should serve as an adequate metric allowing
for candidate identities to be easily compared.
To determine the possible identi cation of the Hagelund specimen,
the reported characteristics were compared with those of the cryptid C.
willsi proposed by LeBlond and Bous eld and with those of an unusual
“caddy” report from one Finn John reported by Hagelund. Crocodilians and
elasmosaurids were also included in the analysis following LeBlond and
Bous eld’s proposal of reptilian af nities for C. willsi. The traits of ear and/
or horn presence, long neck (quanti ed as more than twice the length of head),
and presence of a “toothed” or spiny tail were also included in our comparative
analysis. A generic pinniped was also included as a candidate owing to
Hagelund’s description of his specimen possessing a “seal-like face.” One of
us (McCormick) noted a strong similarity between the Hagelund specimen and
pipe sh; the hypothesis that the specimen represents a known sh rather than
a reptilian cryptid is signi cant, so a number of sh from the northeast Paci c
(all morphologically reminiscent of the specimen) were incorporated as well:
bay pipe sh (Syngnathidae; Syngnathus leptorhynchus), poachers (Agonidae;
Podothecus accipenserinus, Pallasina barbata, Sarritor frenatus), tube-snout
(Aulorhynchidae; Aulorhynchus avidus), green sturgeon (Acipenseridae;
Acipenser medirostris), and cutlass shes (Trichiuridae; Aphanopus arigato,
Lepidopus tchi). A decapod candidate (Pandalus platyceros) was also included
due to Staude and Lambert’s (1995) suggestion that a representative of this
order might have been the culprit behind Hagelund’s specimen.
For a listing of the traits ascribed to “caddy” by LeBlond and Bous eld
(1995), refer to the Introduction above. A review of the reports utilized by
LeBlond and Bous eld (1995) shows that no one individual account possesses
504 M. A. Woodley, D. Naish, C. A. McCormick
the full complement of traits present in the Hagelund description, furthermore
contradictions are frequent (as discussed above), and singular traits such
as “ ns all over the body,” “turtle-like,” “cat-like head,” “like a huge diver
wearing a helmet,” et cetera, are not uncommon. Clearly, the establishment
of C. willsi by LeBlond and Bous eld was subjective, and conclusions drawn
from comparisons with the set of traits have to be limited.
Table 1 presents an analysis of the similarity of 14 candidate identities
to Hagelund’s specimen with respect to 29 different characters. Based on this
analysis it is evident that con ating LeBlond and Bous eld’s C. willsi with the
Hagelund specimen is problematic for a number of reasons: most strikingly, the
plate-like scales of the latter. While the traits of a serrated crest and toothed/
spiny tail are reportedly present in C. willsi, the former does not necessarily
imply armor (it could be a soft structure or even hair), and the latter appears
to have occurred exclusively in the enigmatic Naden Harbour carcass. Also
conspicuously absent in the Hagelund specimen is a long neck. While the
Hagelund specimen and C. willsi share the traits of large eyes and whiskers,
the rarity of the traits in “caddy” reports, coupled with their fairly generic
nature, suggests they have little diagnostic value. Overall, C. willsi exhibited a
relatively low similarity to the Hagelund specimen (8.25 points out of a possible
29). The Finn John “caddy” report shares a number of speci c traits with
Hagelund’s encounter such as a spade-shaped tail, large limpid eyes, whiskers,
fur, and plates on the back, although its overall similarity was low (8 points out
of a possible 29); this unusual report was in fact recalled by Hagelund, and its
dissimilarity to other reports aside from traits speci c to Hagelund’s specimen
is no doubt signi cant (see below).
Interestingly, the Hagelund specimen shares more traits in common with the
generalized crocodilians (plate-like scales, sometimes coloration) (scoring 13
points out of 29); however, signi cantly with respect to LeBlond and Bous eld’s
thesis, it seems to share no more traits in common with the elasmosaurids than
it does with C. willsi (8 points out of 29). A generic pinniped was included for
comparison owing to Hagelund’s reference to his specimen possessing a “seal-
like face.” There are ve seal species native to British Columbia belonging to
both the Otariidae family (northern fur seal Callorhinus ursinus, Steller sea lion
Eumetopias jubatus, and California sea-lion Zalophus californianus), and the
Phocidae family (northern elephant seal Mirounga angustirostris and harbor
seal Phoca vitulina) (Allen, 1974). While the pinnipeds as a whole scored a
respectable 14.75 out of 29, no single species possesses all of the traits present
in Hagelund’s specimen. The fact that Hagelund’s specimen was small (33cm)
is also grounds to rule out pinnipeds as plausible candidates, as this size falls
far below that of any pinniped species. One point of similarity between these
candidates and the Hagelund specimen is swimming with the head out of the
A Baby Sea-Serpent No More 505
* open to eyewitness interpretation; ? unknown presence; O occasionally present due to variability; P present; A absent; s co-occurs as present or absent in the Hagelund specimen;
ss somewhat similar in that it or something that could be mistaken for it sometimes co-occurs; d dissimilar indicating no co -occurrence
Traits Hagelund C. willsi Finn Thalatto- Elasmosaur- Pinnipedia S. lepto- P. accipen- Pa. bar- Sa. Fren
Au. Flav- A. arigato L. fitchi Ac. Medi- Pan. Platy-
John suchia spp . idae spp. spp. rhynchus serinus bata atus idus rostris ceros
Teeth P ? ? P (s) P (s) P (s) A (d) P (s) P (s) P (s) P (s) P (s) P (s) A (d) A (d)
Lips P ? ? A (d) A (d) P (s) P (s) P (s) P (s) P (s) P (s) P (s) P (s) P (s) A (d)
Head held out of water P P (s) ? P (s) P (s) P (s) A (d) A (d) A (d) A (d) A (d) A (d) A (d) A (d) A (d)
Ears and/or horns A O (ss) ? A (s) A (s) O (ss) A (s) A (s) A (s) A (s) A (s) A (s) A (s) A (s) A (s)
Tail dorsally toothed, spin
A P (d) ? A (s) A (s) A (s) A (s) A (s) A (s) A (s) A (s) A (s) A (s) A (s) A (s)
Black dorsally, brown laterall
P A (d) ? ? ? O (ss) O (ss) P (s) A (d) P (s) A (d) A (d) A (d) A (d) A (d)
Yellow tail P A (d) ? ? ? A (d) A (d) P (s) A (d) A (d) A (d) A (d) A (d) O (ss) A* (d)
Plate-like scales P ? P (s) O (ss) A (d) A (d) P (s) P (s) P (s) P (s) A* (d) A (d) A (d) P (s) A* (d)
Eel-like P P (s) P (s) A (d) A (d) A* (d) P (s) A* (d) P (s) A* (d) P (s) P (s) P (s) A (d) A (d)
Long snou
P P (s) P (s) P (s) A (d) O (ss) P (s) P (s) P (s) P (s) P (s) P (s) P (s) P (s) A* (d)
Large eyes P O (ss) P (s) O (ss) O (ss) P (s) P (s) P (s) P (s) P (s) P (s) P (s) P (s) A (d) P (s)
Long nec
A P (d) ? A (s) P (d) A* (s) A (s) A (s) A (s) A (s) A (s) A (s) A (s) A (s) A (s)
Snout slightly hooke
P A (d) ? A (d) A (d) A (d) A* (d) A (d) A (d) A (d) A (d) A (d) A (d) A (d) A (d)
40 cm approximate total lengt
P A (d) ? P (s) ? A (d) O* (ss) A* (d) A (d) A* (d) A (d) P (s) P (s) P (s) A (d)
Diameter:TL ~ 1:16 to 1:10.7 P A (d) ? O (ss) O (ss) A* (d) P (s) P (s) P (s) A* (d) P (s) P (s) P (s) A* (d) P (s)
Head:TL ~ 1:5.33 P A (d) ? O (ss) A (d) A (d) P (s) P (s) P (s) P (s) P (s) P (s) P (s) P (s) A (d)
Undulatory movemen
P P (s) ? P (s) A (d) O (ss) P (s) A* (d) A* (d) A* (d) A (d) O (ss) O (ss) A* (d) A (d)
Flipper-like feet near shoulde
P P (s) P (s) P (s) P (s) P (s) P (s) P (s) P (s) P (s) P (s) P (s) P (s) P (s) A (d)
Dorsal fin(s) A A (s) ? A (s) A (s) A (s) P* (d) P* (d) P* (d) P* (d) P* (d) P* (d) P* (d) P (d) A (s)
Pelvic fins A O* (ss) ? P (d) P (d) P* (d) A (s) P* (d) P* (d) P* (d) P* (d) P* (d) P* (d) P (d) A (s)
Anal fin A A (s) ? A (s) A (s) A (s) P* (d) P* (d) P* (d) P* (d) P* (d) P* (d) P* (d) P (d) A (s)
Dark eyes P A (d) ? ? ? P (s) P (s) A* (d) A* (d) P (s) P (s) P (s) P (s) P (s) A (s)
Tail composed of separate flippers P A* (d) ? A (d) A (d) P (s) A (d) A (d) A (d) A (d) A* (d) A* (d) A* (d) A* (d) A* (d)
Whiskers P O (ss) P (s) A (d) A (d) P (s) A (d) A* (d) A* (d) A* (d) A (d) A (d) A (d) A* (d) A* (d)
Underbelly w/ soft yellow fuz
P A* (d) ? A (d) A (d) A (d) A* (d) A* (d) A* (d) A* (d) A (d) A (d) A (d) A (d) A* (d)
Slender head P ? ? ? ? ? P (s) ? P (s) ? P (d) ? ? ? ?
Seal-like face P ? ? ? ? P (s) ? ? ? ? ? ? ? ? ?
Limpid eye
P ? P (s) ? ? ? ? ? ? ? ? ? ? ? ?
Spade-shaped tail P ? P (s) ? ? ? O* (ss) ? ? ? ? ? ? ? ?
Total similar (out of 29
29 8.25 8 13 8 14.75 16 13 13 12 12 13.5 13.5 10.5 9
Analysis of the Similarity of 14 Candidate Identities to Hagelund’s Specimen with Respect to 29 Different Characters
506 M. A. Woodley, D. Naish, C. A. McCormick
water, which as air breathers these candidates must do at least some of the time.
This trait would seemingly imply that the Hagelund specimen was also an air-
breather; however, Hagelund never mentioned breathing. Rather than dismiss
the local sh species that super cially resemble the Hagelund specimen on the
basis of surface activity, the possibility of aberrant behavior in those sh must
be considered: After all, this is in no way less likely than the identi cation
of Hagelund’s animal as a cryptid. Another possibility is that Hagelund added
“caddy”-like behavior during his recollection of the encounter, although it
would still require the animal to be at or near the surface.
The bay pipe sh is common to the bays and sloughs of the west coast of
Canada, the US, and Mexico, where it lives among common eelgrass, feeding
on small invertebrates (Eschmeyer & Herald, 1983). Data suggest that this
species does venture into more open water (Hart, 1973), so the location of
Hagelund’s encounter is not necessarily problematic. The presence of plate-like
scales, pectoral ns (= “fore ippers”), large eyes, a long snout, and lips all
strongly agree with traits reported from the Hagelund specimen. In Hagelund’s
illustration (reproduced in Figure 1), the specimen was described as having a
black dorsum, brown anks, and yellow ventral fur; this appears to fall within
the known color variation of S. leptorhynchus which has been described as
varying from green to mottled brown with streaks of black on the dorsal portion,
coupled with a yellowish ventrum (Filisky & Peterson, 1998, Girard, 1858).
The reported head length of 7.62 cm (3) for the Hagelund specimen gives a
head:TL (TL = total length) proportion of 1:5.3, which is similar to the head:SL
(Standard Length = tip of snout to end of caudal peduncle) proportion of 1:6.5
reported for S. leptorhynchus (Girard, 1858). Since the individual measured by
Girard (1858) was 15 cm in length, it might be possible that positive allometric
growth of the pipe sh’s head would yield an even closer proportion for a large
While S. leptorhynchus has not been observed to reach 40 cm in length,
the maximum reported length of 33 cm is within a reasonable margin of
observer error (Eschmeyer & Herald, 1983). The 2.54 cm (1) body depth of
the Hagelund specimen (body depth:TL = 1:16) seems excessive for a pipe sh,
but this could be explained by the possibility that the Hagelunds captured a
pregnant male, as pregnancy in pipe sh causes considerable swelling. The lack
of a dorsal n and anal n in Hagelund’s description could be explained by
their being folded down or being overlooked entirely as they are semi-opaque
structures composed of only 32 and 3 soft rays, respectively (Girard, 1858). The
presence of a bifurcated caudal appendage reportedly composed of two separate
and overlapping ns could be due to a misinterpretation of a folded caudal
n or damage to the n. Hagelund’s description of integumentary structures
is curious; hair-like appendages used for camou age are present in some
A Baby Sea-Serpent No More 507
syngnathids such as Urocampus carinirostris, however these sh are weak
swimmers which lack a dispersive egg and larval phase, implying that the entire
life cycle occurs in localized areas in the western Paci c (Neira et al., 1998,
Chenoweth et al., 2002). While the possibility exists that the Hagelund pipe sh
represents an undescribed, large, “hairy” syngnathid from the northeast Paci c,
we contend that the fur-like growth reported on the specimen’s underside could
be a misinterpretation of the ventral coloration, which has been described as
yellowish with a mesh work pattern of brown lines (Girard, 1858). Alternately,
a covering of algae could create the impression of a fur-like integument.
The most problematic morphological aspect reported by Hagelund is the
presence of “whiskers” and of teeth. Since Hagelund already referred to “fore
ippers,” it seems very unlikely that he would interpret the pectoral ns again
as “whiskers,” and it is also unlikely that the coloration pattern would give the
impression of a structure protruding from the head of the sh. Pipe sh have
very small, toothless mouths quite unlike what Hagelund described. Despite
these problematic details, the pipe sh is still the strongest candidate for the
Hagelund specimen, scoring 16 out of a possible 29. Other local sh species,
however, also deserve consideration:
Poachers (Agonidae spp) are morphologically distinctive scorpaeniforms
with rows of bony armor covering their typically elongated bodies; due to a
super cially similar appearance to pipe shes, they are also somewhat plausible
candidates for the Hagelund specimen. Large eyes, long snouts, small teeth, and
lips are typically present in members of the clade, as are barbel-like structures,
which could be interpreted as “whiskers.” The spines associated with the
armor of some species could potentially be confused for hairs. The often-
large pectoral ns are the primary means of locomotion except for the c-start
escape behavior, which utilizes the caudal n; this appears to be comparable
with the behavior that Hagelund observed in his specimen (Nowroozi et al.,
2009). Problematically, however, poachers are benthic and strongly negatively
buoyant (Nowroozi et al., 2009), which makes the capture of a specimen near
the surface in relatively deep water seem rather improbable. The presence of a
prominent spiny rst dorsal n (sometimes absent), a soft-rayed second dorsal
n, a prominent anal n, and a small, thoracic pelvic n would have to have
gone unnoticed in Hagelund’s specimen if an agonid was involved, as such
traits go unreported (one possibility is that they were folded). Like pipe sh,
the caudal n is rounded and could only be interpreted as bifurcated if it was
Most of the approximately 50 poacher species occur in the North Paci c,
and it is noteworthy that they are not well-known (Jensen, 2005). While at least
17 species are known to occur in the same broad geographical area as Hagelund’s
sighting, most of these species fall considerably under 40 cm in length and
508 M. A. Woodley, D. Naish, C. A. McCormick
some are even fairly stout-bodied (Froese & Pauly, 2009). The super cially
sturgeon-like Podothecus accipenserinus can reach 30.5 cm in length, has
very prominent barbels, and has a coloration which can be roughly similar to
that of the Hagelund specimen (Hart, 1973); this species obtained a similarity
score of 13 out of 29, however, it is not particularly attenuated since the body
depth:total length is approximately 1:7.5 instead of 1:16, which was observed
in the Hagelund specimen. This difference falls far outside a comfortable
margin of error. Other strong candidate species include Pallasina barbata and
Sarritor frenatus; P. barbata has a slender head and similar proportions as the
Hagelund specimen, but is small (max. 17 cm) with a prominent stripe on its
head (Eschmeyer & Herald, 1983), Sa. frenatus is somewhat larger (max. 27
cm) and matches the coloration description as well as the head:total length
proportion, but not the body depth:total length, and the known range does not
include southern British Columbia (although it is close) (Eschmeyer & Herald,
1983). These candidates both obtained similarity scores of 13 and 12 out of
29, respectively. It does not appear that there is any long-snouted poacher,
which matches the size, proportions, and reported coloration of the Hagelund
specimen (Tokranov & Orlov, 2005, Miller & Lea, 1976, Froese & Pauly, 2009).
While Agonidae as a whole may display almost all of the characters reported
in the Hagelund specimen, no one species is more probable as a candidate
identity than the bay pipe sh, and suggesting an unknown species is outside the
applicability of the present data.
The tubesnout (Aulorhynchus avidus) is a marine gasterosteoid
(stickleback relative) with a long snout, large eyes, and body that is both slender
and elongate; notably, the caudal n is forked (Hart, 1973), and sharp teeth are
present in the mouth (Jordan & Gilbert, 1882). 24–27 membrane-free dorsal
spines are located anterior to the soft rays (Hart, 1973). The skin is naked with
the exception of rugose shields near the lateral line and dorsal n (Jordan &
Gilbert, 1882). It could be possible for the shields and/or spines to suggest
more extensive plate-like scales to an eyewitness, but this is not very likely.
A soft dorsal n is located far back on the body and mirrors the anal n (Hart,
1973). The sh is described as pale mottled brown with an olive-brown or
yellow-brown dorsum, white ventrum, and a silver patch bordered by a dark
band near the head and gills (Hart, 1973). This candidate species obtained a
similarity score of 12 out of 29, however the maximum recorded size is 18.8
cm (Bayer, 1980), and coupled with the distinctive coloration, lack of actual
armor, and lack of morphology which can be interpreted as whiskers and fuzz,
the tubesnout can be ruled out as a strong candidate for Hagelund’s specimen.
Cutlass shes regularly exceed 40 cm in length, and have big eyes, teeth,
subtle pelvic ns, and a forked caudal n. Aphanopus arigato (formerly A.
intermedius) and Lepidopus tchi occur in the northeast Paci c Ocean and
A Baby Sea-Serpent No More 509
appear to be the strongest candidates for the Hagelund specimen within the
group; A. arigato is coppery black in color with an iridescent tint, has a body
depth:standard length proportion ranging from 1:12 to 1:16.4 and a head
length:standard length proportion ranging from 1:4.9 to 1:5.5; L. tchi has a
black or brown coloration with a silver abdomen, a body depth:standard length
proportion of 1:9.2 to 1:13.3, and a head length:standard length proportion
of 1:4.2 to 1:5.5 (Nakamura & Parin, 1993). It is presumed both species use
similar locomotion as A. carbo, which involves ostraciiform swimming (i.e.
caudal n only) when stalking prey and anguilliform when in striking distance;
it is notable that the dorsal and anal ns are retracted when in the former and
the median ns are erected in the latter (Bone, 1971). Both candidate species
obtained similarity scores of 13.5 out of 29, however the absence of plate-like
scales, structures which may explain “yellow fuzz” (the anal n occurs only in
the far posterior of the body), and strikingly different coloration compared with
Hagelund’s specimen provides substantative grounds upon which this group
can be ruled out as a strong candidate identity.
All sturgeon species (Acipenseridae) can exceed 40 cm in length and have
ve rows of dorsal, lateral, and ventrolateral scutes, which could complement
Hagelund’s description of plate-like scales. The Green sturgeon (Acipenser
medirostris) is present in the northeast Paci c Ocean, has a nearly homocercal
tail (i.e. consisting of two equal lobes) which could be regarded as separate
ippers,” lips, pectoral ns, barbels (potential “whiskers”), and a similar
head:body length proportion as the Hagelund specimen (Girard, 1858). This
candidate obtained a similarity score of 10.5 out of 29, however it is not eel-like
and it seems unlikely an observer could overlook a dorsal n, pelvic ns, an
anal n, strongly ventral mouth, and prominent lateral stripes, which effectively
rules out this candidate species as a strong contender.
Staude and Lambert (1995) proposed decapods as a candidate for Hagelund’s
specimen, but did not specify a species. Presumably, the authors suggested this
crustacean order due to the possibility of an exoskeleton being interpreted as
plate-like scales, antennae as whiskers, pereopods and pleopods (swimming
and walking legs) as “fuzz,” and uropods as a tail composed of two ippers. The
reported presence of lips, teeth, pectoral appendages, and the overall vertebrate-
like appearance of the drawing are problematic for any decapod candidate. The
Spot prawn (Pandalus platyceros) is the largest shrimp in the region, with an
eyestalk-telson length that may exceed 20 cm, and coloration that is orange in
large individuals with white spots on the rst and fth pleura (Hoffman, 1972).
American lobster (Homarus americanus) have been introduced to the region
(Ray, 2005), and t the size criterion, but are highly unlikely to be found at
the surface and to go unrecognized. While the decapod hypothesis is certainly
novel and thought-provoking, the lack of a viable candidate species, coupled
510 M. A. Woodley, D. Naish, C. A. McCormick
with a low similarity score for P. platyceros (9 out of 29), means that it can be
ruled out as a plausible candidate.
As can be seen through comparison of the illustrations in Figure 1,
Hagelund’s drawing bears more than a super cial resemblance to the bay
pipe sh. Despite the problematic description of whiskers and teeth, given how
well the majority of the details reported by Hagelund seem to complement our
proposed S. leptorhynchus identi cation, is it therefore possible that Hagelund
simply misremembered the details of the specimen in his 1987 description?
One piece of evidence supportive of this is the 18-year gap between Hagelund
making his observations and recounting them in his book. Memories are hardly
infallible and are subject to distortion with the passage of time (Wiseman &
Lamont, 1996, Wright & Loftus, 2008). Hagelund claimed that the description
of C. willsi he heard from Finn John had slipped his mind at Pirate Cove but
perhaps there was a subconscious in uence upon recollection; Finn John’s
account shares traits with Hagelund’s account that are reported in no other
“Caddy” reports (ventral fur, dorsal overlapping plates) and shares several
details with little variation (horse-like head, large eyes, whiskers, slender
body, ippers, spade-shaped tail). Perhaps as Hagelund became increasingly
convinced that what he and his family had caught was a baby sea-serpent, his
memory of the specimen correspondingly adapted by taking on the “appropriate”
The bay pipe sh shares more characteristics in common with Hagelund’s
specimen than any other candidate species. Although certain observed traits
are congruent with similar traits reported in C. willsi, Hagelund’s “sea-serpent”
differs markedly from typical accounts of C. willsi and is comparatively far
less similar. Based on this analysis it is therefore suggested that the theory that
Hagelund (1987) describes a pipe sh, most probably S. leptorhynchus, is the
most probable explanation.
Both cryptozoologists and those skeptical of the eld have reinterpreted
reports of sea-serpents and lake-monsters as reports of either known animals
perhaps engaged in unusual behaviors or as inanimate objects viewed under
unusual conditions. Naish (1997) argued that photographs of an alleged C.
willsi “carcass” taken on the beach at Camp Fircom, British Columbia, in
the 1930s, do not represent an animal carcass but are in fact composed of a
montage of beach debris. Similarly Naish (2001) and Radford (2003) have
both independently suggested that the “lake monster” photographed by Sandra
Mansi in 1977 at Lake Champlain might have been a tree stump propelled to the
surface by gas generated through bacterial decay. Paxton et al. (2005) suggested
that Egede’s alleged account of a “most dreadful monster” witnessed off the
A Baby Sea-Serpent No More 511
coast of Greenland in 1734, which was interpreted by Heuvelmans (1968) as
a sighting of a “super-otter” (a hypothetical primitive mega-archaeocete), may
actually have been of a whale in a state of arousal, owing to morphological
similarities between the description of the terminal end of the monster and
whale penises (Paxton et al., 2005). Many additional examples represent
probable misidenti cation of this kind; even Heuvelmans (1968) ruled out 52
reports in In the Wake of the Sea-Serpents on the grounds that they were likely
misidenti cations of known species or other more mundane objects.
At this stage it is necessary to make two points. Firstly, in suggesting
that Hagelund’s 1968 baby sea-serpent was in all likelihood a bay pipe sh,
we are in no way implying that he deliberately falsi ed the details of his
encounter, although we have reason to suspect that certain details may have
Figure 1. A reproduction of Hagelund’s drawing of his “baby sea-serpent”
(top) along with an illustration of the bay pipe sh S. leptorhynchus
(middle), and C. willsi (bottom) along with scales (40 cm and 3m,
Note that the maximum girth:total length ratio in Hagelund’s illustration is approximately 1:11
while the description gives a signi cantly more attenuated ratio of 1:16. Pipe sh and Hagelund
specimen by Cameron A. McCormick, C. willsi by Darren Naish.
512 M. A. Woodley, D. Naish, C. A. McCormick
been misremembered subsequently. Similarly, there is no reason to believe that
Hagelund (or his family) should have been familiar with pipe sh as they are
far less well-known than their close relatives the sea horses; even experienced
whalers like Hagelund would never normally encounter pipe sh in their line of
work. It is therefore easy to imagine how a bizarre-looking, long, serpentine,
armour-plated sea creature might excite the imaginations of those encountering
it for the rst time.
Additionally, in suggesting a bay pipe sh identity for Hagelund’s animal,
we are not suggesting that all “Caddy” reports are simply cases of mistaken
identity. Since the Hagelund specimen, by far the smallest reported “Caddy,”
was already at the extreme of bay pipe sh size variation, it is highly improbable
that pipe sh misidenti cation resulted in any other reports of the cryptid. The
possibility that cutlass shes and sturgeons may have been responsible for some
“Caddy” sightings cannot be ruled out, however.
It is our contention that Hagelund’s encounter should not be used in support
of the existence of “Caddy,” let alone form the basis of entirely speculative
theories concerning its putative reproductive and life-history characteristics.
Although the elimination of this encounter lessens the overall likelihood for
the existence of “Caddy,” rooting out the probable misidenti cations advances
enquiry, as researchers can focus their efforts on the more robust and ultimately
intriguing data.
We would like to thank Charles Paxton and one anonymous reviewer for comments that
greatly improved the quality of this manuscript.
Allen, J. A. (1974). History of the North American Pinnipeds: A Monograph of the Walruses, Sea-
Lions, Sea-Bears, and Seals of North America (Natural Sciences in America). Stratford,
NH: Ayer Co. Publishers.
Amorim, A. F., Arfelli, C. A., & Castro, J. I. (2000). Description of a juvenile Megamouth Shark,
Megachasma pelagios, caught off Brazil. Environmental Biology of Fishes, 59, 117–123.
Bayer, R. D. (1980). Size and age of the tube-snout (Aulorhynchus avidus) in the Yaquina Estuary,
Oregon. Northwest Science, 54, 306–310.
Bone, Q. (1971). On the Scabbard Fish Aphanopus carbo. Journal of the Marine Biological
Association of the United Kingdom, 51, 219–225.
Bous eld, E. L., & LeBlond, P. H. (1992). Preliminary studies on the biology of a large marine
cryptid in coastal waters of British Columbia. American Zoologist, 32(Abstracts), 2A.
Bous eld, E. L., & LeBlond, P. H. (1995). An account of Cadborosaurus willsi, new genus, new
species, a large aquatic reptile from the Pacic coast of North America. Amphipacica, 1,
A Baby Sea-Serpent No More 513
Chenoweth, S. F., Hughes, J. M., & Connolly, R. C. (2002). Phylogeography of the pipe sh,
Urocampus carinirostris, suggests secondary intergradation of ancient lineages. Marine
Biology, 141, 541–547.
Eschmeyer, W. N., & Herald, E. S. (1983). A Field Guide to the Paci c Coast Fishes. Boston, MA:
Houghton Mif in Company.
Filisky, M. & Peterson, R. T. (1998). Peterson First Guide to Fishes of North America. Boston,
MA: Houghton Mif in Company.
Froese, F., & Pauly, D. (2009). FishBase. http://www.
Girard, C. (1858). Fishes: Part IV. Washington, DC: U.S War Department.
Hagelund, W. A. (1987). Whalers No More. Vancouver: Harbour Publishing.
Hart, J. L. (1973). Paci c Fishes of Canada. Ottawa: Fisheries Research Board of Canada.
Heuvelmans, B. (1968). In the Wake of the Sea-Serpents. New York: Hill and Wang.
Heyning, J. E. (1989). Cuvier’s beaked whale Ziphius cavirostris G. Cuvier, 1823. In S. H. Ridgway
& R. Harrison (Eds.), Handbook of Marine Mammals, Volume 4, River Dolphins and the
Larger Toothed Whales, London: Academic Press.
Hoffman, D. L. (1972). The development of the ovotestis and copulatory organs in a population of
Protandric shrimp Pandalus platyceros Brandt from Lopez Sound, Washington. Biological
Bulletin 142, 251–270.
Jensen, G. C. (2005). A unique feeding method by a teleost sh, the fourhorn poacher Hypsagonus
quadricornis (Agonidae). Biological Bulletin, 209, 165–167.
Jordan, D. S., & Gilbert, C. H. (1882). Synopsis of the Fishes of North America. Washington:
Government Printing Press.
LeBlond, P. H. (2001). Caddy—An update. Crypto (Dracontology Special), 1, 55–59.
LeBlond, P. H., & Bous eld, E. L. (1995). Cadborosaurus, Survivor from the Deep. Victoria,
British Columbia: Horsdal & Schubart.
LeBlond, P. H., & Sibert, J. (1973). Observations of Large Unidentied Marine Animals in British
Columbia and Adjacent Waters. Vancouver: University of British Columbia. Institute of
Oceanography. Manuscript #28.
Lucas, S. G., & Reynolds, R. E. (1993). Putative Paleocene plesiosaurs from Cajon Pass, California,
U.S.A. Cretaceous Research, 14, 107–111.
Miller, D. J., & Lea, R. L. (1976). Guide to the Coastal Marine Fishes of California. California:
University of California, Division of Agriculture.
Naish, D. (1997). Another Caddy carcass? Cryptozoology Review, 2, 26–29.
Naish, D. (2001). Sea serpents, seals and coelacanths: An attempt at a holistic approach to the
identity of large aquatic cryptids. Fortean Studies, 7, 75–94.
Nakamura, I., & Parin, N. V. (1993). Snake Mackerels and Cutlass shes of the World (Families
Gempylidae and Trichiuridae). An Annotated and Illustrated Catalogue of the Snake
Mackerels, Snoeks, Escolars, Gem shes, Sack shes, Domine, Oil sh, Cutlass shes,
Scabbard shes, Hairtails, and Frost shes Known to Date. New York: Food and
Agricultural Organization Fisheries Synopsis Number 125, Volume 15.
Neira, F. J., Miskiewicz, A. G., & Trnski, T. (1998). Larvae of temperate Australian shes.
Melbourne: University of Western Australia Press.
Nowroozi, B. N., Strother, J. A., Horton, J. M., Summers, A. P., & Brainerd, E. L. (2009). Whole-
body lift and ground effect during pectoral n locomotion in the northern spearnose
poacher (Agonopsis vulsa). Zoology, 112, 393–402.
Paxton, C. G. M. (1998). A cumulative species description curve for large open water marine
animals. Journal of the Marine Biological Association of the United Kingdom, 78, 1389–
Paxton, C. G. M. (2001). Predicting pelagic peculiarities: Some thoughts on future discoveries in
the open seas. Crypto (Dracontology Special), 1, 60–65.
Paxton, C. G. M. (2009). The plural of ‘anecdote’ can be ‘data’: Statistical analysis of viewing
514 M. A. Woodley, D. Naish, C. A. McCormick
distances in reports of unidenti ed large marine animals 1758–2000. Journal of Zoology,
279, 381–387.
Paxton, C. G. M., & Holland, R. (2005). Was Steenstrup right? A new interpretation of the 16th
century sea monk of the Øresund. Steenstrupia, 29, 39–47.
Paxton, C. G. M., Knatterud, E., & Hedley, S. L. (2005). Cetaceans, sex, and sea serpents: An
analysis of the Egede accounts of a “most dreadful monster” seen off the coast of Greenland
in 1734. Archives of Natural History, 32, 1–9.
Radford, B. (2003). The measure of a monster: Investigating the champ photo. Skeptical Inquirer,
Ray, G. L. (2005). Invasive Marine and Estuarine Animals of the Paci c Northwest and Alaska.
Vicksburg, MS: U.S. Army Engineer Research and Development Center.
Raynal, M. (2001). Cryptocetology and mathematics: How many cetaceans remain to be discovered?
Crypto (Dracontology Special), 1, 93–112.
Raynal, M., & Sylvestre, J.-P. (1991). Cetaceans with two dorsal ns. Aquatic Mammals, 17, 31–36.
Saggese, P. (2006). Cadborosaurus willsi: Attributive inquiry. Bipedia, 24, 50–68.
Scott, P., & Rines, R. (1975). Naming the Loch Ness monster. Nature, 258, 466–468.
Solow, A. R., & Smith, W. K. (2005). On estimating the number of species from the discovery
record. Proceedings of the Royal Society, Series B: Biological Sciences, 272, 285–287.
Staude, C. P., & Lambert, P. (1995). Editorial . . . an opposing view. Amphipacica, 1(Supplement),
Tokranov, A. M., & Orlov, A. M. (2005). Spatial–bathymetric distribution and size composition
of Aleutian alligator sh Aspidophoroides bartoni (Agonidae) in Paci c waters of the
northern Kuril islands and southeastern Kamchatka. Journal of Ichthyology, 45, 380–384.
Van Helden, A. L., Baker, A. N., Dalebout, M. L., Reyes, J. C., Van Waerebeek, K., & Baker,
C. S. (2002). Resurrection of Mesoplodon traversii (Gray, 1874), senior synonym of M.
bahamodi (Reyes, van Waerebeek, Cardenas, & Yanewz, 1995) (Cetacean: Ziphiidae).
Marine Mammal Science, 18, 609–621.
Wiseman, R., & Lamont, P. (1996). Unraveling the Indian rope-trick. Nature, 383, 212–213.
Woodley, M. A. (2008). In the Wake of Bernard Heuvelmans: An Introduction to the History and
Future of Sea Serpent Classi cation. Bideford, Devon: Centre for Fortean Zoology Press.
Woodley, M. A., Naish, D., & Shanahan, H. P. (2008). How many extant pinniped species remain
to be described? Historical Biology, 20, 225–235.
Wright, D., & Loftus, E. (2008). Memory for events: Eyewitness testimony. In G. Cohen & M. A.
Conway (Eds.), Memory in the Real World, New York: Psychology Press.
... Hagelund initially confused the specimen for a sea snake and referred to it as "eel-like," which implies lateral undulation and a vertical (fi sh-like) tail, contra the claim of our critics. Our analysis (Woodley, Naish, & McCormick 2011) specifi cally noted that the description of teeth is problematic, but we again note the 18-year gap between Hagelund making his observations and recounting them in his book. Recollection of any memory after such a long interval will surely be prone to distortion. ...
... However, refusal to posit such an entity does not violate Occam's razor, because pink elephants are not antecedently plausible. Like the pink elephant, the idea that there exist extant sauropterygians corresponding to the descriptions of Caddy is not even remotely antecedently plausible, as has been repeatedly argued elsewhere (Bauer & Russell 1996, Naish 2001, Staude & Lambert 1995, Woodley, Naish, & Shanahan 2008. Extant sauropterygians aside, the idea that one type of animal is being reported across Caddy encounters also does not appear to be antecedently plausible (we refer again to Figure 1). ...
Full-text available
In their response to our recent article (Woodley et al., 2011), Bousfield and LeBlond (2011) argue on the basis of purported morphological and behavioral differences that the case for Hagelund's juvenile 'Cadborosaurus' being a pipefish is weakened into triviality. We note several major problems with their response and feel that their dismissive tone indicates a biased and unscientific approach to the investigation of this subject. Firstly, note that Bousfield and LeBlond (2011) thought that, by attempting to dismiss our pipefish identification, they had completed their task of critiquing our paper. While their 'critique' is grossly inadequate in any case, note that they made no mention of the assorted additional fish taxa also found by us (Woodley et al., 2011) to be more similar to the Hagelund specimen than the specimen was to their 'Cadborosaurus' construct.
... While the example is contrived, in practice such type reinterpretations are not infrequent (e.g., Ribot et al. 1996;Woodley et al. 2011;Cappellini et al. 2013;Laloy et al. 2013;Witteveen 2015). Reassessments of the taxonomic identity of type specimens are very frequently involved in determining heterotypic synonymy. ...
Full-text available
We utilize an Answer Set Programming (ASP) approach to show that the principles of nomenclature are tractable in computational logic. To this end we design a hypothetical, 20 nomenclatural taxon use case with starting conditions that embody several overarching principles of the International Code of Zoological Nomenclature; including Binomial Nomenclature, Priority, Coordination, Homonymy, Typification, and the structural requirement of Gender Agreement. The use case ending conditions are triggered by the reinterpretation of the diagnostic features of one of 12 type specimens anchoring the corresponding species-level names. Permutations of this child-to-parent reassignment action lead to 36 alternative scenarios, where each scenario requires 1-14 logically contingent nomenclatural emendations. We show that an ASP transition system approach can correctly infer the Code-mandated changes for each scenario, and visually output the ending conditions. The results provide a foundation for further developing logic-based nomenclatural change optimization and compliance verification services, which could be applied in globally coordinated nomenclatural registries. More generally, logic explorations of nomenclatural and taxonomic change scenarios provide a novel means of assessing design biases inherent in the principles of nomenclature, and thus may inform the design of future, big data-compatible identifier systems for systematic products that recognize and mitigate these constraints.
... The largest living representatives of most taxa occur in the oceans. Many of these ocean giants have played considerable roles in lore about sea monsters (Carr et al., 2002; Gatschet, 1899; Lenik, 2010; Papadopoulos & Ruscillo, 2002; Paxton, 2009; Verrill, 1897; Woodley, Naish & McCormick, 2011). Today, these formidable species, such as blue whales and giant squids, continue to attract considerable attention from scientists, media, and the public alike. ...
Full-text available
What are the greatest sizes that the largest marine megafauna obtain? This is a simple question with a difficult and complex answer. Many of the largest-sized species occur in the world’s oceans. For many of these, rarity, remoteness, and quite simply the logistics of measuring these giants has made obtaining accurate size measurements difficult. Inaccurate reports of maximum sizes run rampant through the scientific literature and popular media. Moreover, how intraspecific variation in the body sizes of these animals relates to sex, population structure, the environment, and interac- tions with humans remains underappreciated. Here, we review and analyze body size for 25 ocean giants ranging across the animal kingdom. For each taxon we document body size for the largest known marine species of several clades. We also analyze intraspecific variation and identify the largest known individuals for each species. Where data allows, we analyze spatial and temporal intraspecific size variation. We also provide allometric scaling equations between different size measurements as resources to other researchers. In some cases, the lack of data prevents us from fully examining these topics and instead we specifically highlight these deficiencies and the barriers that exist for data collection. Overall, we found considerable variability in intraspecific size distributions from strongly left- to strongly right-skewed. We provide several allometric equations that allow for estimation of total lengths and weights from more easily obtained measurements. In several cases, we also quantify considerable geographic variation and decreases in size likely attributed to humans.
Full-text available
We utilize an Answer Set Programming (ASP) approach to show that the principles of nomenclature are tractable in computational logic. To this end we design a hypothetical, 20 nomenclatural taxon use case, with starting conditions that embody several overarching principles of the International Code of Zoological Nomenclature, including Binomial Nomenclature, Priority, Coordination, Homonymy, Typification and the structural requirement of Gender Agreement. The use case ending conditions are triggered by the reinterpretation of the diagnostic features of one of 12 type specimens anchoring the corresponding species-level epithets. Permutations of this child-to-parent reassignment action lead to 36 alternative scenarios, where each scenario requires a set of 1-14 logically contingent nomenclatural emendations. We show that an ASP transition system approach can correctly infer the Code-mandated changes for each scenario, and visually output the ending conditions. The results provide a foundation for further developing logic-based nomenclatural change optimization and validation services, which could be applied in global nomenclatural registries. More generally, logic explorations of nomenclatural and taxonomic change scenarios provide a novel means of assessing design biases inherent in the principles of nomenclature, and can therefore inform the design of future, big data-compatible identifier systems that recognize and mitigate these constraints.
This Letter is a reply to the research article “A Baby Sea-Serpent No More: Reinterpreting Hagelund’s Juvenile ‘Cadborosaur’ Report,” in JSE 25:3, Fall 2011 (Woodley, Naish, & McCormick, 2011). Naish and colleagues indulge in the common home-quarterbacking habit of insisting that anything described as different must be an erroneous description of something found in a book that vaguely looks like it. A comparison of Hagelund’s “baby-Caddy” with a pipefish (Figure 1) shows significant differences in the latter.
Technical Report
Full-text available
This report presents the results of a survey beginning at the end of 1969 on "sea-monsters", displayed, discussed and compared with previously published information on the subject. The fundamental question of existence remains unresolved. Heuvelmans' categories have not proved useful for classifying our data and doubt is thrown upon their validity.
Full-text available
Mesoplodon traversii (Gray, 1874) is shown to be a senior synonym of the recently described beaked whale Mesoplodon bahamondi Reyes et al., 1995 on the basis of a phylogenetic analysis of mitochondrial DNA control region sequences. The mandible and teeth of M. traversii, first reported in 1873 by Hector as Dolichodon layardii, are redescribed. The species can be distinguished by features of the calvaria; including the large jugal, broad rostrum, and small distance between premaxillary foramina. The male teeth, which are large and spade-shaped with a strong terminal denticle, are also diagnostic. M. traversii is known only from Pitt Island and White Island, New Zealand and Robinson Crusoe Island, Juan Fernandez Archipelago, Chile.
Full-text available
A re-evaluation of the "most dreadful monster" originally described by the "Apostle of Greenland" Hans Egede in 1741 suggests that the missionary's son Poul probably saw an unfamiliar cetacean. The species seen was likely to have been a humpback whale (Megaptera novaeangliae), a North Atlantic right whale (Eubalaena glacialis) or one of the last remaining Atlantic grey whales (Eschrichtius robustus) either without fl ukes or possibly a male in a state of arousal.
Two elasmosaurid plesiosaur specimens have been recovered from rocks of the San Francisquito Formation(?) in Cajon Pass, California. Some invertebrate fossils and palynomorphs from the San Francisquito Formation(?) here indicate a Paleocene age, but no direct association of Paleocene fossils and the plesiosaur fossils can be demonstrated. Therefore, the Cajon Pass plesiosaurs are most reasonably considered Late Cretaceous.