Content uploaded by Rudolf N. Nikolaevitch Burukovsky
Author content
All content in this area was uploaded by Rudolf N. Nikolaevitch Burukovsky on Oct 14, 2020
Content may be subject to copyright.
Crustaceana 90 (7-10) 865-881
Michael Türkay Memorial Issue
FEEDING ECOLOGY OF THE SHRIMP CRANGON ALLMANNI KINAHAN,
1860 (DECAPODA, CRANGONIDAE) IN THE NORTH AND WHITE SEAS
BY
R. N. BURUKOVSKY1)
Kaliningrad State Technical University, Kaliningrad 236022, Russia
ABSTRACT
The food composition of the shrimp Crangon allmanni from the Helgoland Trench (North
Sea) and Onega Bay (White Sea) is described. The main food items (>60%) include detritus,
representatives of about 30 benthic species, dominated by polychaetes, malacostracans, ophiuroids,
bivalves and ophistobranchs (Cylichna,Diaphana spp.) as well as plant remains. In the North
Sea, C. allmanni is closer to an attacking predator than to a predator-gatherer like in the White
Sea. This difference may be caused by diverging habitat- and community characteristics and
dissimilar size composition of the two studied shrimp populations. Moreover, C. allmanni changes its
foraging mode (grazing, gathering, attacking) during ontogenesis. A comparison of the obtained food
composition data of C. allmanni with literature data on six other species of the same genus showed
all to be benthos feeders, predator-gatherers with elements of detrito- and necrophagia, using grains
of sand as millstones in their gastric mill.
ZUSAMMENFASSUNG
Hier wird die Nahrungszusammensetzung der Garnele Crangon allmanni aus der Helgolän-
der Tiefen Rinne (Nordsee) und der Onega-Bucht (Weißes Meer) beschrieben. Die Nahrungs-
Hauptbestandteile (>60%) sind Detritus und Pflanzenreste, sowie Vertreter von ungefähr 30 benthis-
chen Arten, dominiert von den Tiergruppen Polychaeta, Malacostraca, Ophiuroidea, Bivalvia und
Ophistobranchia (Cylichna,Diaphana spp.). In der Nordsee ist C. allmanni eher ein angreifender
Jäger gegenüber einem Jäger und Sammler im Weißen Meer. Dieser Unterschied könnte durch eine
Divergenz der Habitate und Lebensgemeinschaften, sowie der Größenklassen der beiden Garnelen-
populationen verursacht werden. Weiterhin ändert C. allmanni die Art seiner Nahrungssuche (grasen,
sammeln, attackieren) im Laufe seiner Entwicklung. Ein Vergleich der erhaltenen Nahrungszusam-
mensetzungsdaten von C. allmanni mit Literaturdaten von sechs weiteren Arten derselben Gattung
zeigt, dass diese durchweg Benthosfresser und Jäger/Sammler mit Zügen von Reste- und Aasfressern
sind, welche Sandkörner als “Mühlsteine” in ihren Magenmühlen verwenden.
1)e-mail: burukovsky@klgtu.ru
©Koninklijke Brill NV, Leiden, 2017 DOI 10.1163/15685403-00003568
866 R. N. BURUKOVSKY
INTRODUCTION
The genus Crangon Fabricius, 1798 includes 20 species (De Grave & Fransen,
2011); however, the eastern North Atlantic is inhabited by only two of them:
Crangon crangon (Linnaeus, 1758) and C. allmanni Kinahan, 1860. The first one
is the target species for fishery in the North Sea (Schwinn et al., 2014).
Crangon allmanni is a mobile epibenthic species that regularly occurs in bottom
trawl catches but it does not form dense aggregations. Data on the biology of C.
allmanni are very scarce, and information on its food composition is limited to
three paragraphs in a fundamental paper by Allen (1960). This shrimp is very
sensitive to effects of stressful environmental factors, due to which it can serve
as an environmental indicator (Blahudka & Türkay, 2002). Because of this, C.
allmanni was chosen for a study in the framework of project INTAS 51-5458
“Population and trophic biology of Crangon allmanni” with project coordinator
Dr. M. Türkay (Crustacean Section, Senckenberg Society for Nature Research,
Frankfurt am Main, Germany), the initiator of sampling for this study. The study
of the food composition of C. allmanni in different parts of its range was my task
in this project. The results of this investigation with comparison on other Crangon
species are presented in this paper.
MATERIAL AND METHODS
For this study, four samples from the Helgoland Trench (North Sea) and one
from the Onega Bay (White Sea) were analysed.
The Helgoland Trench samples were collected on 11 August 2004, 8 August
2005, 22 July 2006 and 27 July 2006 at 54°08N7°50
-7°53E at a depth range of
55-57 m. In total 639 stomachs of shrimps as well as their carapace length (from
the end of the rostrum to the posterior edge of the carapace) were analysed. It was
found that 351 stomachs contained food remains and 92 stomachs were full with
food.
The second sampling was performed in July 2006 in the southeastern part
of Onega Bay at 64°03.676-64°37.147N 36°53.107-37°54.998E at a depth of
6.9-31.8 m. In total 169 stomachs were analysed and the carapace length of
the respective males and females was determined. We found that 103 stomachs
contained food remains and 27 stomachs were full.
The shrimp carapace length was measured to the nearest 0.1 mm, using the
micrometer of a binocular microscope (MBS-9), as the shortest distance between
the anterior end of the rostrum to the posterior edge of the carapace in the median
line of its dorsal side.
FEEDING ECOLOGY OF CRANGON ALLMANNI IN THE NORTH AND WHITE SEAS 867
For this feeding study the methodology of Burukovsky (Burukovsky & Trunova,
2007; Burukovsky, 2009) was used. After stomach opening, the degree of its filling
was determined on a 4-point scale: 0, the stomach is empty; 1, the food occupied
less than half the stomach volume; 2, food occupied about half (1/3-2/3) of the
volume of the stomach; 3, the stomach is full.
The food lump from each stomach was studied in a drop of water in a Petri
dish. Identification of the taxonomic status of the food remains usually was made
on the level of class or order (for example, Gastropoda or Bivalvia, Mysidacea,
Euphausiacea or Isopoda), trying to determine the taxonomic affiliation of the
prey as accurately as possible, ideally to the species level. Special attention was
paid to find out to which particular lifestyle, including the specific habitat (pelagic,
benthic, sessile, burrows etc.), the prey belonged to.
The food organism remains were counted and measured with the ruler of the
ocular micrometer of the microscope. Measuring the prey length entirely was
rarely possible because the remains were relatively crushed. Therefore, the parts of
the body that could be measured (primarily skeletal elements, scales, eye lenses,
otoliths or vertebrae of fish, the chaetae of Chaetognatha and Polychaeta, statoliths
of mysids etc.) were used for size determination.
Regardless of stomach fullness, the composition of food items was identified in
all stomachs with food. In full stomachs, the components of the food lump were
visually estimated by volume with an accuracy of 10%. Full stomachs were chosen
for visual estimation to avoid the impact of different degrees of digestion of the
food remains in the stomachs on the obtained result. The food and inedible items
that constituted less than 10% of the total volume were just noted. According to
Burukovsky (2009), the results of that study were calculated using:
(1) The frequency of occurrence (FO, percent of occurrence of given food
component from the total number of examined stomachs with food).
(2) The Froerman Index (mean number of prey items in the full stomach
excepting grains of sand, detritus and plant remains). The Froerman Index
was calculated as the sum of all food group’s FO in percent, divided by
100.
(3) The frequency of prevalence (FP, frequency of occurrence of full stomachs
(in %) where one of the food items consists of 60% and more of the food
lump volume, i.e., prevailed).
(4) The reconstructed averaged (virtual) food lump volume (VFLV, mean share
of each food component in the total food lump volume in %).
VFLV and FP were calculated using only data on full stomachs. The term “food
components” means both alive and non-living remains (grains of sand, spicules
of sponges) that were detected in the stomachs, in contrast to the “food items”,
868 R. N. BURUKOVSKY
i.e., those components that are directly used by shrimps as food. The terminology
for describing the methods of shrimp hunting (feeding behaviour) of Burukovsky
(2009) was used.
These feeding characteristics should be used together because they complement
each other. Separately they give an one-sided image of the diet of the studied
animals. The role of foraminifera in the diet of some shrimp species has been
demonstrated to be misinterpreted (Burukovsky, 2009). Their FO values can reach
up to 60-70%, so they fall into the category of the most common and important
food group. But their share in the VFLV was less than 0.1-0.2%. Hence, studying
the ontogenetic variability of food composition using both parameters can be
especially effective. In some species during ontogenesis the FO of given food
item gradually decreases, but its VFLV increases. Therefore, adult shrimps feed
on a given prey less frequently but in larger quantities (Burukovsky, 2009). Thus,
the use of these two parameters (FO and VFLV) together gives a realistic idea
on the role of given food groups in the diet of a shrimp. This double approach
also provides an effective “instrument” for reconstruction of feeding behaviour
(Burukovsky, 2009).
A stepwise characterization of the food composition was used:
(1) Description of food remains. This is very important, as it allows to estimate
the type (pattern) of feeding. This is the first step towards descriptive results
against a mere listing of food composition.
(2) Description of the frequency of occurrence of food items in all stomachs
with food, regardless of its amount in the stomach.
(3) Description of the volume ratios of food components in full stomachs. It is
also very important, as it allows to reconstruct the virtual food lump.
RESULTS
The North Sea
The carapace length of Crangon allmanni from the North Sea ranged from 3.2 to
14.0 mm. The food components obtained here include sand grains, detritus, plant
remains, non-identified animal remains and remains that taxonomic status could be
identified at least to class or order levels (table I). The sand grains were presented
in nearly all stomachs. Their numbers varied from 3-10 to hundreds of grains with
sizes from 0.05 to 0.7 mm (modal 0.3-0.5 mm), i.e., it was the fraction of medium
sand (Petelin, 1967). The sand FO was 81.8%. In approximately half full stomachs
the share of sand in their total food lump volume was less than 10%, varying in
limits of 60-80% in 14.3% of full stomachs. The sand content in the VFLV was
16.3%.
FEEDING ECOLOGY OF CRANGON ALLMANNI IN THE NORTH AND WHITE SEAS 869
TABLE I
Food composition in stomachs of the shrimp Crangon allmanni Kinahan, 1860 in the North Sea and
the White Sea
Component Frequency of
occurrence
(FO, %)
Share of virtual food
lump volume
(VFLV, %)
Frequency of
prevalence
(FP, %)
North Sea White Sea North Sea White Sea North Sea White Sea
Detritus 72.486.816.331.66.515.8
Polychaeta 29.154.421.35.821.75.2
Mysidacea 15.1–13.7–14.1–
Ophiuroidea 14.5– 4.1– 4.3–
Amphipoda 12.85.95.61.13.3–
Echinoidea 11.4– 0.7–––
Foraminifera 9.116.2– 4.2– –
Bivalvia 8.848.51.412.1– –
Gastropoda 8.02.92.2– 2.2–
Copepoda and
Cladocera
8.022.1–11.0–10.5
Cumacea – 4.4– 6.8– 5.2
Pisces 4.3– 1.7– 2.2–
Euphausiacea 3.7– 0.6–––
Indetermined eggs 3.119.1––––
Paguroidea 2.6– – – 3.3–
Shrimps 2.3– 4.1– 4.3–
Ostracoda 2.014.7– 4.7– –
Gastropoda eggs 1.7– – – – –
Brachyura and
other “Reptantia”
1.4– 0.2– – –
Isopoda 1.4– 0.2– – –
Cnidaria (Hydrozoa) 1.4– 1.0– 1.1–
Chaetognatha 1.41.50.2– – –
Nematoda 1.44.4– – – –
Plant debris 1.41.50.7– 1.1–
Diatomea 1.1– – – – –
Tintinoidea 0.8– – – – –
Acari and
their nymphs
0.6–––––
Insecta 0.6– 1.7– 3.3–
Oligochaeta 0.6– 0.5–––
Brachiopoda 0.3–––––
Holothurioidea 0.3–––––
Priapulida 0.3– 0.4– 0.4–
Tanaidacea – 2.9– 1.6– –
Undetermined
remains
20.510.96.7– 5.4–
Sand-grains 81.879.414.321.08.715.8
Spicules and
other debris
3.11.5– – – –
Total stomachs 351 103 92 27 92 27
Froerman Index 2.20 2.93 – – – –
Frequency of
prevalence (%)
––––73.336.7
870 R. N. BURUKOVSKY
The obtained detritus was a loose mass of grey or greyish-beige, rarely almost
black, colour. It is more or less uniform and does not contain inclusions. The
detritus FO was 72.4%, i.e., it also occurs in the majority of studied stomachs,
at least in trace amounts. The detritus in full stomachs usually does not exceed half
the volume of the food lump, however, in 6.5% of the stomachs, its share exceeded
60%, and two of the stomachs were filled solely by detritus. Its share in the VFLV
was 16.3%.
Plant remains occurred very rarely — only in 5 stomachs from 351 studied
(1.4%), and their share in the VFLV was 0.7%. It was not possible to identify these
remains. In one full stomach their share was 60% of the total food volume. Plant
remains should be considered as an accidental food component.
Unidentified food components were divided into two groups. The first one
included pieces of chitin in different stages of degradation, that were impossible to
identify. Probably, they are crustacean remains that were eaten during the previous
feeding event, or maybe remains of dead, decomposed prey. Their FO was 6.5%
and as a rule they occurred in stomachs that were hardly full, but in three full
stomachs they occupied 10-20% of the volume of food lump. Their share in the
VFLV was very small (0.6%).
The other group of non-identified food components includes tissue fragments
with fibrous or gelatinous consistency, wrinkled and covered with detritus. Their
FO was 14.0%. Probably this were the half-decayed remains of dead animals that
were ingested. Most likely C. allmanni feeds on dead animals, as evidenced by the
occurrence of adults and nymphs of mites and imagines of insects in its stomach
that only could drop to a depth of about 50 m, as they were already dead. One
insect specimen had a length of 2.5 mm. In two stomachs the share of insects in the
volume of the food lump was 60 and 100%, respectively. By preliminary estimation
necrophagy in C. allmanni occurred at least in 20%, and the share of these dead
animals was 8.4% of the VFLV. The last three groups of food components in total
accounted for 39% of the VFLV.
Food items eaten alive can be divided into three groups, depending on their
FO values. The first group includes polychaetes, which are found almost 2.5
times less frequently than detritus (FO 29.1%), being completely dominated by
other food groups. Polychaetes presented at least 4-5 species that were mainly
the representatives of errant life forms of the families Polinoidae and Glyceridae
(Glycera sp.). They occupied 80-100% of the lump food volume in 15 of 92 full
stomachs. Their body lengths were about 3.0-35.0 mm, commonly 3.0-4.0 mm.
These pieces were the fragments of single specimens. Only once more than 10
fragments of polychaetes of 3 or 4 species were found, each having a length of
1-2 mm. The share of polychaetes (21.3%) in the VFLV is almost 1.5 times that of
FEEDING ECOLOGY OF CRANGON ALLMANNI IN THE NORTH AND WHITE SEAS 871
detritus, which, in contrast, occurs 2.5 times more often (table I). In the first place,
polychaetes had a FP value of 21.7%.
Secondary food items include mysids, amphipods and ophiuroids, the largest
fraction (FO 15.1%) of which were mysids. In the stomachs they occurred either
as whole specimens or, more often, as statoliths. Mysids constitute 13.7% of the
VFLV and are the dominating group in full stomachs (FP 14.1%). In 10 stomachs
they occupy 90-100% of the volume. Usually, in the stomachs we found 1-3
specimens of adult mysids with a length of 13-14 mm. Female marsupiums were
found to contain embryos or larvae. Conclusively, polychaetes and mysids were
found to be the main prey of C. allmanni.
Ophiuroidea remains occurred in 14.5% of all studied stomachs. Their body
fragments were easily identified by skeleton elements of arms that most frequently
occurred in stomach contents. Their share in the VFLV was 4.1%, probably due
to dominating small fragments. Whole specimens with a disc diameter of nearly
2 mm were found in only two instances.
The amphipod FO (12.8%) and the value of share in the VFLV (5.6%) were
nearly the same as those for Ophiuroidea. In two cases more than half of the full
stomach volume consisted solely of amphipod remains. Among the amphipods, the
representatives of Gammarida dominated the diet. As a rule, they were represented
by remains of single specimens (maximum of 4) with a length of 0.8-11.0 mm,
mainly 2.0-3.5 mm. Also fragments of Caprellidae occurred relatively often, and
once there were found specimens of Hyperiidae with 5.0-6.0 mm length.
All other food items were insignificant or rare. The insignificant food is
considered to be the components that had a relatively high frequency of occurrence
but VFLV of full stomachs never reached or exceeded 10%. Among others, sea
urchins and foraminifera were found (table I).
Rare food items with low FO can be found in relatively large numbers, occu-
pying more than 60% of the food lump volume. For example, there were bivalves
(Macoma sp. with shell length 1.2 mm) and gastropods. The opisthobranch Di-
aphana sp. and its egg masses were dominating among the gastropods. In one
stomach, up to 5 specimens of Diaphana sp. were found, having shell sizes of 2.0-
3.5 mm. In one stomach, Natica sp. with nearly the same shell size as Diaphana
sp. was detected.
The rare food group also includes various crustaceans: juveniles of hermit crabs,
crabs, juveniles of Munida sp. and the small crustacean family Axiidae with a
length of 3-4 mm. Interesting data were obtained for shrimps. Their FO was
only 2.3%, but the share in the VFLV had a relatively high value (4.1%), being
a clear indication for cannibalism: most of the shrimp remains were represented
exclusively by juveniles of C. allmanni. Shrimps only occurred in 8 stomachs with
872 R. N. BURUKOVSKY
4 full stomachs being occupied up to 90-100% of their food lump volume. The
total lengths of shrimps were 12-13 mm, probably being molting specimens.
A large number of dominant food items (12), high total the FP value (79.4%) and
a relatively low Froerman Index value (2.20) suggested that the feeding behavior
of C. allmanni in the North Sea can be characterized as an attacking predator
in combination with detritophagy and necrophagy, with active feeding on living
animals dominating (about 60% of the VFLV).
The White Sea
At Onega Bay, the carapace length of males studied was 4.2-6.9 mm, while
females reached 3.5-12.3 mm. Sand grains occurred in most of the stomachs
studied with an FO of 79.4%. Their number varied from 2-3 up to hundreds, on
average 10-20. Grain size was 0.05-0.8 mm (modal 0.15-0.3 mm) (Petelin, 1967).
The share of sand in the VFLV was 21%. In 16% of the stomachs studied sand
occupied more than 60% of the food lump volumes.
Detritus in the stomachs of C. allmanni from the Onega Bay (table I) is
represented by two types. The first type of detritus was grey or brownish in
colour, disintegrating into flakes in a drop of water. As a rule, this type occurred
in stomachs with small food remains, but in one instance the share of detritus
was 90% of the food lump volume. The second type of detritus was in the form
of a pulp consisting of very fine pieces of chitin. All transitions of these chitin
pieces were found during the maceration process. In several stomachs, copepod
and cladoceran parts, as well as other skeleton elements, were found among these
pieces of chitin, being a mixture of body parts with inclusions of particles of
detritus. Thirty specimens of whole harpacticoid copepods with length 0.4-0.5
and 0.7-1.5 mm only occurred in two stomachs. The whole cladocerean Podon
leuckartii (Sars, G. O., 1862) was found once. In the remaining cases, cladocerean
residues were identified by the morphology of their mandibles. That is why the
cladoceran FO and the share in the VFLV are underestimated, while the situation
for copepods is the opposite. Therefore, the data on cladocerans and copepods were
combined (table I).
Most probably shrimps in Onega Bay primarily feed on dead copepods and
cladocerans in their places of aggregation at the bottom (for example, in the
hydrologic situations that promote the sedimentation and accumulation of these
dead crustaceans at the bottom). In these places the dead bodies undergo all stages
of transformation to detritus and serve as a food for C. allmanni. Detritus occured
more often than sand (86.8%) and it occupied 31.6% of VFLV. Plant remains were
presented by significantly macerative pieces of brown algae (Phaeophyta) in the
only one stomach.
FEEDING ECOLOGY OF CRANGON ALLMANNI IN THE NORTH AND WHITE SEAS 873
It is very difficult to estimate the role of dead animals in shrimp feeding ecology.
Doubtless, dead (before ingestion) animals (fish larvae with an intact integument,
insects, mites, etc.) were absent in the stomachs from Onega Bay. There, little worn
pieces of chitin and different fibrous fragments with the traces of decomposition
were found. They were classified as unidentified remains; these fragments occurred
only rarely (FO 10.9%) and in very small amounts. In reality, according to our
data the role of necrophagy in this shrimp is probably undervalued. Copepods and
cladocerans are a relatively important food group (sometimes they amount to 80%
of the food lump volume in some stomachs). However, it is impossible to determine
which of them were eaten dead or alive. The mixture of these small crustaceans
with detritus occurred in 22.1% of all observed stomachs and it amounted 11.0%
of the VFLV. By the rank of their FP value (10.5%), copepods and cladocerans are
in 3rd place after sand grains and detritus.
Detritus together with the mixture of copepods and cladocerans at different
stages of decomposition occupied almost half of the VFLV (42.6%). They were
the main food groups for C. allmanni.
Polychaetes and bivalve molluscs with FO values of, respectively, 54.4 and
48.5% occupy the first two places among animals eaten alive. Polychaetes were
represented almost exclusively by errant forms. Only once the chaetae of the
sedentary polychaete family Spionidae were found. In general, relatively small
fragments of bodies and chaetae were found in the stomachs. In one stomach there
was a piece of a worm belonging to the family Polinoidae of 1.4 mm length and
almost the whole specimen of 2.7 mm length with few elytra on the dorsal side.
The VFLV of polychaetes was at 5.8%.
In the Onega Bay stomachs of C. allmanni bivalves were presented by Ma-
coma sp. (shell length 0.7 mm), Musculus sp. (1.25 mm) occurring only once.
More important in shrimp food were fragments of the bivalve family Cardiidae
(Cerastoderma sp.?): 1-3 specimens with a shell size of 0.65-1.25 mm, mainly
0.7-0.8 mm. Although the mollusc FO is a little less than that of polychaetes, its
share in the VFLV is twice as high (12.1%). Gastropods in the stomachs were
very rare (FO 2.9%) and they were represented by single specimens, mostly by the
young opisthobranch Cylichne sp. with a size of 0.9-1.7 mm.
Moreover, unidentified, spherical eggs of a diameter from 0.07-0.25 mm oc-
curred relatively often (FO 19.1%), but they were found solitary and thus do
not play a significant role in the diet of the shrimp species. On the other hand,
foraminifera with the FO 16.2% had VFLV values of up to 4.2%. They were repre-
sented by the agglutinated species Ammobaculites cassis (Parker, 1870) [currently
as: Ammotium cassis (Parker, 1870)]. The size of whole animals varied from 0.7 to
2.5 mm.
874 R. N. BURUKOVSKY
The remaining part of the VFLV consisted of crustaceans (table I). Among
them, ostracods were the most important ones (FO 14.7% and VFLV 4.7%); they
occurred singly or rarely with 2-3 specimens with a shell length of 0.1-0.9 mm. The
FO value of cumaceans was 4.4% and an VFLV value of 6.8% (due to the much
larger body size of 3-4 mm). Based on their body’s degree of preservation it can be
assumed that they were probably eaten alive. Fragments of other malacostracans
were used for body length reconstruction and had roughly similar sizes: amphipods
1-2 mm and tanaidaceans 3 mm.
Thus, the virtual food lump of C. allmanni in the Onega Bay consisted of
three main components: (1) the detritus together with inclusions of copepods and
cladocerans (VFLV 42.6%), (2) crustaceans (14.2%) and (3) bivalves (12.1%).
Polychaetes (VFLV 5.8%) and foraminiferans (4.2%) were a secondary food and
they had an appreciable role in the shrimp diet.
In the studied sample the total FP value was the relatively low (table I). Primarily
it includes the detritus together with inclusions of copepods and cladocerans.
Judging by the value of the Froerman Index (2.93), C. allmanni combines the
hunting types of grazing and gathering, but rather, it is a predator-gatherer. A large
share of detritophagous brings this shrimp to a predator-opportunist by type of
hunting.
DISCUSSION
Comparison of food composition in Crangon allmanni at different
geographic areas
In this study the food composition of Crangon allmanni in the Northumberland
area of the western part of the North Sea (Allen, 1960) is compared with the food
composition at the Helgoland Trench (North Sea) and at Onega Bay (White Sea).
In the Northumberland area, shrimps fed on live crustaceans (FO 63.3%) and
polychaetes (51.6%). Molluscs, foraminifers and Ophiuroidea occurred signifi-
cantly less, as well as scales of the whiting (a fish, Merlangius merlangus (L.,
1758)). The last was found only in stomachs of the shrimps that were caught at
shallow depths. In food lumps silt was always present (personal remark: most
likely it was detritus) together with sand grains. Among eaten polychaetes, Neph-
thys sp. had dominated (FO about 90%) and Glycera sp. occurred at about in
10% of stomachs. Crustaceans were represented by small cumaceans, amphipods,
copepods, and juveniles of C. allmanni. Among the identified molluscs, the bi-
valves Dosinia lupinus (L., 1758), Venus striatula Da Costa, 1778 [currently as:
Chamelea striatula (Da Costa, 1778)] and the opisthobranch gastropod Cylichna
cylindrica (Bruguière, 1792) [currently as: Cylichna cylindracea (Pennant, 1777)]
FEEDING ECOLOGY OF CRANGON ALLMANNI IN THE NORTH AND WHITE SEAS 875
were the most common ones. The food composition of shrimps that were caught
at inshore and offshore stations and in males and females insignificantly differed
(Allen, 1960).
These data have both a marked difference and a definite similarity in comparison
with our results. There are important details that indicate a similarity in the diet of
the shrimp at different parts of its distributional range: in all three studied areas, C.
allmanni prefer to eat opisthobranchs among the gastropods — Cylichne cylindrica
in Northumberland waters (Allen, 1960), Diaphana sp. in the Helgoland Trench
and Cylichne sp. in the Onega Bay.
Unfortunately, it is impossible to carry out a more complete comparison of
Allen’s (1960) data with ours because her description of food composition was
presented in a much more generalized form. Comparison of the food composition
of C. allmanni in the Helgoland Trench and Onega Bay finds both general
similarities and many particular differences.
In the Helgoland Trench area (excluding diatoms and tintinnidean ciliats, which
were probably a transit food: Nigmatullin & Toporova, 1982) food remains belong
to 29 taxa and in Onega Bay 15 species were found in shrimp stomachs. In
the last area, 7 taxa of malacostracans and all echinoderms and fish found in
the Helgoland trench were absent, but here cladocerans were found. There were
significant differences in the FO, VFLV and FP values for different food groups
(table I). This is reflected in the values of the Froerman Index and total values
of the FP. The Froerman Index in shrimp from the Helgoland Trench and from
Onega Bay was 2.20 and 2.93, respectively, while the total FP were 73.3 and
36.7%, respectively. C. allmanni inhabiting the North Sea is closer to an attacking
predator, while in the White Sea it can be considered as a predator-gatherer. This
difference, probably, may be explained by differences in habitat and community
characteristics and the size composition of studied shrimps in these parts of the
range.
The C. allmanni habitat depth geographically varies between the different parts
of its range. In the western part of the North Sea it occurs at depths from 20-160 m,
mainly 40-100 m (Walker, 1892; Allen, 1960), but it was also caught together with
C. crangon at depths less than 15 m (Allen, 1960). In the Helgoland Trench, it was
found at depths of 30-50 m on a relatively hard bottom, mainly on shell limestone
(Blahudka & Türkay, 2002). In the Denmark Strait and Faeroe plateau slopes its
maximum depth of occurrence is 900 m (Spiridonov et al., 2008). In the Barents
Sea this species was caught at depths of 85-104 m, in the northern part of the
Throat of the White Sea it was found at depths of 35.6-48.3 m in June 2004, and
in Onega Bay soft bottoms at depths of 6.9-31.8 m in early August 2006.
In Onega Bay, shrimps live at depths less than the optimum specified by Allen
(1960) and in the Helgoland Trench it occurs almost within the most common
876 R. N. BURUKOVSKY
Fig. 1. Size composition of the studied shrimp, Crangon allmanni Kinahan, 1860 from (1) the
Helgoland Trench area (North Sea) and (2) Onega Bay (White Sea).
depths for this species. Salinity at these two locations is different: it is 24-30h
in Onega Bay and 32-34hin the Helgoland Trench, respectively (Spiridonov et
al., 2008). Despite such differences in habitat conditions, the range of the shrimp
sizes and their bimodal (5.5 and 9.5 mm) structure in both areas is almost identical
(fig. 1). However, in Onega Bay about 60% of the studied shrimps had a size of
less than 8 mm, and in the Helgoland Trench, in contrast, about 60% of shrimps
had a size of more than 8 mm (fig. 1).
Crangon crangon and C. allmanni are close sister species. The composition
of food and types of hunting of C. crangon in the North and White Sea (Bu-
rukovsky & Trunova, 2007; Burukowsky, 2009) changes during ontogenesis. Ju-
venile shrimps mainly feed on copepods (Harpacticoida) and behave as grazing
predators. Sub-adult middle-sized shrimps become predator-gatherers and adult
mature specimens with carapace lengths of more than 9 mm change their hunting
method to the one of an attacking predator. There are also very tentative data on
the same ontogenetic shifts in feeding behaviour of C. allmanni (cf. Burukovsky,
unpublished data). Probably these ontogenetic distinctions explain the difference
in diet composition of the shrimp C. allmanni in the Helgoland Trench area of the
North Sea and Onega Bay (White Sea).
Comparative characteristics of food composition in some shrimps
of the genus Crangon
The genus Crangon actually comprises 20 species (De Grave & Fransen,
2011). All species of this genus are the inhabitants of the shelf, mostly its
upper part, often from the water’s edge. The most common deep-water species,
Crangon dalli Rathbun, 1902 from the Far East, inhabits depths from 3 to
FEEDING ECOLOGY OF CRANGON ALLMANNI IN THE NORTH AND WHITE SEAS 877
630 m (Hayashi & Kim, 1999), as does C. allmanni (see above). All species
are morphologically very similar. They show the characteristics of a buried
shrimp lifestyle: body dorsoventrally flattened and rostrum in form of small ledge.
Interspecific morphological differences are mainly the presence or absence of
single or paired keels on the posterior segments of abdomen; the rounded, flattened
or sulcated dorsal side of these segments, and other relatively poorly expressed
traits (Hayashi & Kim, 1999).
These facts suggest that their morpho-functional trophological complex allows
to use the quasi-same food organism spectrums for nutrition. Therefore, these
shrimps occupy similar habitats and use comparable food resources within the
limits of their ranges, i.e., members of identical life forms. Consequently Crangon
shrimps probably occupy a similar position in the food webs of their communities.
There is evidence that, on the one hand, shrimps of the genus Crangon are a food
item of the adult flatfishes and, on the other, they act as predators against the
newly settled flatfish larvae (Hayashy & Kim, 1999; Hanamura & Matsuoki, 2003;
Taylor, 2005).
The composition of food in varying details was studied in 7 species: C. allmanni,
C. crangon,C. septemspinosa Say, 1818, C. franciscorum Stimpson, 1856, C.
nigricauda Stimpson, 1856, C. uritai Hayashi & J. N. Kim, 1999, and C. affinis
De Haan, 1849 (cf. Plagman, 1940; Price, 1961; Wilcox & Jeffries, 1974; Sitts
& Knight, 1979; Siegfried, 1982; Whale, 1985; Hanamura & Matsuoka, 2003;
Burukovsky & Trunova, 2007).
Unfortunately, the methods of quantitative estimation of different food items and
their rations in the food lumps used by different authors are not fully comparable.
It does not allow a detailed analysis of the similarities and differences in the food
composition of all these studied species. Authors often paid attention to quite
specific aspects of feeding, while ignoring what they considered to be unimportant
details. Therefore, further comparison of the food of Crangon-shrimps based on
data of C. crangon and C. allmanni, i.e., two species that I studied using one
methodology (see above and Burukovsky & Trunova, 2007), is not possible. What
could be done in a future study is to extrapolate the results of this comparison to
other species for which the data on food are available.
Crangon crangon is distributed from the White Sea to Morocco coast and the
Mediterranean, Black and Baltic Seas. In the North Sea it is subject to the fishing
industry. The range of C. allmanni is limited by the Northeastern Atlantic from
Iceland and the Bay of Biscay to the White Sea. Crangon crangon inhabits sandy
and silt-sandy bottoms of the upper sublittoral (mainly from the water’s edge down
to 50 m), and it is rare in waters of the outer shelf. Crangon allmanni is distributed
on the inner shelf at depths, mainly 40-100 m.
878 R. N. BURUKOVSKY
Consequently, these very closely related species have ranges that overlap
spatially and somewhat differ bathymetrically and can be considered as vicarious
species that divided the upper part of the North-East Atlantic shelf. Because
bathymetric parameters of their ecological niches are distinguishable, we should
expect a divergence of the food spectra of these species according to the features
of the bathymetric distribution of their food organisms.
The similarity is determined by morpho-functional features of both species,
and the differences by the peculiarities of specific ecological niches. Indeed, both
species are characterized by the presence of sand in almost every stomach. Grains
of sand probably play the role of millstones in their gastric mill: in C. crangon,
the FO value was 91.9% and in C. allmanni 78.6%, while the VFLV was 22.0 and
21-14.3%, respectively. However, in the North Sea the sand VFLV value in the
stomachs of C. allmanni never exceeded 50%, which is probably connected with
the peculiarities of the bottom in its habitat (Burukovsky & Trunova, 2007).
The FO and VFLV of detritus in both species are comparable to those values
of sand. Detritus and sand occurred in almost every stomach and together they
make up almost half of the VFLV, 41.7% in C. crangon and 28.4-52.6% in C.
allmanni. The third common food item for both species were the remains of dead
animals. They were represented in the stomachs by indefinable fragments of tissue
of aquatic organisms with evident signs of post-mortem maceration, and in the
presence of terrestrial invertebrates (mites, adult insects) and also entire specimens
of dead, young fish with a body size disabling the shrimp to ingest these in a living
state.
Both species are benthos feeders with a preference for relatively sedentary prey
(although C. allmanni to a lesser degree). Both species feed on relatively mobile
animals (amphipods in C. crangon, mysids in C. allmanni) only in the later stages
of ontogeny, i.e., a large-sized shrimp.
Differences in food composition clearly originate in different food availability
being habitat-dependent. In the diet of more deep-water C. allmanni, almost no
plant remains, as well as no larvae of chironomids and other aquatic insect larvae
were detected. In stomachs of C. crangon, the remains of echinoderms, crabs and
hermit crabs have never been encountered.
According to the method of food procurment, both species can be classified as
predator-gatherers with elements of detrito- and necrophagia. In the early stages
of ontogenesis, C. crangon predominantly feed on meiobenthos (harpacticoids,
nematodes). Young C. allmanni predominantly show attacking predator behavior,
especially in the North Sea. Both species (mentioned above) demonstrate such
modes of foraging as grazing, gathering and attacking, changing them in the
process of ontogenesis.
FEEDING ECOLOGY OF CRANGON ALLMANNI IN THE NORTH AND WHITE SEAS 879
The application of this approach to other studied species for comparison is
very difficult because different researchers are studying the very same species
had described food items in different ways. For example, Siegfried (1982) notes
that C. franciscorum the FO of inorganic material (i.e., sand) reached 26%, plant
remains 42% and the non-identified (usually crustaceans: “unidentified fragments
of crustacean exoskeletons”; Siegfried, 1982: p. 133), probably dead animals, 44%.
In contrast, Whale (1985) does not mention the presence of sand and unidentified
remains in food lumps of the same species, and the FO of plant remains was only
4%. Comparison of the data of Siegfried (1982) on the food composition of C.
franciscorum gives a substantial similarity with C. crangon and C. allmanni,but
the opposite when compared to the data and description of Whale (1985).
Wilcox & Jeffries (1974) directly compared the food composition of C. affinis
and C. septemspinosa with C. crangon and C. allmanni. They classified all these
species as predators that are capable of swallowing anything (i.e., silt, sand, algae
and detritus). However, they are hesitant to identify some of the components of the
food lumps as the remains of dead animals: in their opinion it is part of detritus or
sometimes it is result of predation. Hanamura & Matsuoka (2003) described almost
the same for C. uritai. However, they specifically stress that plant remains were
absent in stomachs, but they perfectly demonstrated the decrease of the detritus
role and the increasing role of amphipods and mysids in the diet of C. uritai in
parallel with body size growth.
We can conclude that all these members of the genus Crangon are benthos
feeders, predators-gatherers, feeding on relatively sedentary preys. The most
mobile of them are amphipods and mysids, they start to play a significant role
in feeding of the largest shrimps. Probably, shrimps of the genus Crangon all
use grains of sand as millstones in their gastric mill, collecting detritus and plant
remains (the latter at the depths where they are presented), and they are also all
necrophagous to some degree.
ACKNOWLEDGEMENTS
I cordially thank the project coordinator, the late Michael Türkay, for the transfer
of the shrimp stomach samples and their biological data from the Helgoland
Trench; E. Yu. Soljanko and V. A. Spiridonov for collection shrimp samples in the
White Sea; A. V. Trunova and S. Yu. Grigorenko for their help in cameral study of
stomach contents; and Ch. M. Nigmatullin for the MS reading, its translation and
useful comments, as well as M. Sonnewald for editing the scientific English of the
manuscript.
880 R. N. BURUKOVSKY
REFERENCES
ALLEN, J. A., 1960. On the biology of Crangon allmani Kinahan in Northumberland waters. J. Mar.
Biol. Ass. U.K., 39: 481-508.
BLAHUDKA,S.&M.TÜRKAY, 2002. A population study of the shrimp Crangon allmanni in the
German Bight. Helgol. Mar. Res., 56: 190-197.
BURUKOVSKY, R. N., 2009. Feeding and feeding relationships of shrimp: 1-408. (FGOU VPO
KGTU Publishing, Kaliningrad). [In Russian with English abstract.]
BURUKOVSKY,R.N.&A.V.TRUNOVA, 2007. On the feeding of shrimp Crangon crangon
(Decapoda, Crangonidae) in Kandalaksha Gulf (the White Sea) in July and September 2004.
In: Collected papers on memory of famous Russian hydrobiologist B. G. Ivanov “Marine
commercial invertebrates and algae (biology and fishery)”. Trudy VNIRO, 147: 181-203.
(VNIRO Publishing, Moscow). [In Russian with English abstract.]
DEGRAVE,S.&C.H.J.M.FRANSEN, 2011. Carideorum catalogus. The recent species of the
dendrobranchiate, stenopodidean, procarididean and caridean shrimps (Crustacea, Decapoda).
Zool. Meded. Leiden, 85: 195-588.
HANAMURA,Y.&M.MATSUOKA, 2003. Feeding habits of the sand shrimp, Crangon uritai
Hayashi & Kim, 1999, in the central Seto Inland Sea, Japan. Crustaceana, 76: 1017-1024.
HAYASHI,K.-I.&J.N.KIM, 1999. Revision of the East Asian species of Crangon (Decapoda:
Caridea: Crangonidae). Crust. Res., 28: 62-103.
NIGMATULLIN,CH.M.&N.M.TOPOROVA, 1982. Food spectrum of squid Sthenoteuthis pteropus
(Steenstrup, 1855) in the epipelagic zone of the Tropical Atlantic. Collected papers “Feeding
and food relations of fishes and invertebrates in the Atlantic Ocean”: 3-8. (AtlantNIRO
Publishing, Kaliningrad). [In Russian with English abstract.]
PETELIN, V. P., 1967. Granulometric analysis of sea bottom deposits. (Nauka, Moscow). [In
Russian.]
PLAGMANN, J., 1940. Ernährungsbiologie der Garnele (Crangon vulgaris Fabr.). Helgol. Wiss.
Meeresunters., 2: 113-162.
PRICE,JR., K. S., 1961. Biology of the sand shrimp, Crangon septemspinosa, in the shore zone of
the Delaware Bay region. Univ. Delaware Mar. Lab. Contrib., 29: 244-255.
SCHWINN,M.,M.TÜRKAY &M.SONNEWALD, 2014. Decapod fauna of the Helgoland Trench
(Crustacea) a long-term study in a biodiversity hotspot. Mar. Biodiv., 44: 491-517.
SIEGFRIED, C. A., 1982. Trophic relation of Crangon franciscorum Stimpson and Pal aemon
macrodactylus Rathbun: predation on the opossum shrimp, Neomysis mercedis Holmes.
Hydrobiologia, 89: 129-139.
SITTS,R.M.&A.W.KNIGHT, 1979. Predation by the estuarine shrimps Crangon franciscorum
Stimpson and Palaemon macrodactylus Rathbun. Biol. Bull., 156: 356-368.
SPIRIDONOV,V.A.,V.I.SOKOLOV,I.KRÖNKE &M.TÜRKAY, 2008. How is the fragmented
distribution range of the reddish sand shrimp Crangon allmanni Kinahan formed? In: Presen-
tations of the Scientific Conference dedicated to the 70 years anniversary of the White Sea
Biological Station of the Moscow University: 121-123. (Grif & K Publishing, Moscow).
TAYL O R, D. L., 2005. Predation on post-settlement winter flounder Pseudopleuronectes americanus
by sand shrimp Crangon septemspinosa in NW Atlantic estuaries. Mar. Ecol. Prog. Ser., 289:
245-262.
WAHLE, R. A., 1985. The feeding ecology of Crangon franciscorum and Crangon nigricauda in
San Francisco Bay, California. J. Crust. Biol., 5: 311-326.
WALKER, A. O., 1892. Revision of the Podophtalmata and Cumacea of Liverpool Bay to May, 1892.
Proc. Trans. Liverpool Biol. Soc., 6(Session 1891-1892): 96-104.
FEEDING ECOLOGY OF CRANGON ALLMANNI IN THE NORTH AND WHITE SEAS 881
WILCOX,J.R.&H.P.JEFFRIES, 1974. Feeding habits of the sand shrimp Crangon septemspinosa.
Biol. Bull., 146: 424-434.
First received 19 April 2016.
Final version accepted 23 June 2016.
