1. The visible coloration of three species of amphipod Crustacea (Gammarus pulex, G. marinus and Orchestia gammarella) was found to be due almost exclusively to carotenoid pigments, including in each case a proportion of an astacin-like carotenoid acid.

  2. No carotenoid pigments could be demonstrated, even in traces, in three species of the cavernicolous Amphipod genus Niphargus.

  3. Carotenoid pigmentation is thought to be characteristic of Amphipoda and would therefore be the ancestral pigment of the cave forms. The lack of pigment in cave Amphipoda may be due to absence of light; it is not due to lack of carotenoid pigments in the food.

During an investigation of the causes of pigment lack in cave Crustacea, it seemed desirable to study the kind of pigmentation in related above-ground (epigean) forms. The colour of three species of Isopoda was found to be due to a melanin-like pigment in branched cells (Baldwin & Beatty, 1941). Two plant carotenoid pigments—cryptoxanthine and β-carotene—were present throughout the body tissues of one (Asellus aquaticus), but did not contribute significantly to the general colour of the animals. In the present work, it has been demonstrated that the colour of three species of Amphipoda is due wholly to carotenoid pigments, including in each case a proportion of a specific animal astacin-like pigment.

For separation of the carotenoid pigments the standard methods have been used as described by Zechmeister (1934) and Zechmeister & Cholnoky (1938). The petrol ether used throughout was of the 50-60° boiling-point grade. The magnesium oxide used for adsorption was the arsenic-free product of British Drug Houses. Adsorption columns were about 1 cm. in width and 10 cm. in height. Absorption maxima were determined with a Hartridge Reversion Spectroscope with a symmetrical slit and greatest probable error of ± 2 μ μ.

Preliminary investigations have been made of the pigments of three amphipod species, Gammarus pulex L., G. marinus Leach and Orchestia gammarella Pallas. From these species the pigment can be extracted with organic solvents, leaving the bodies entirely decolorized except for the eyes, which remain black. The externally visible coloration of three Isopoda species studied (Baldwin & Beatty, 1941) is due to a melanin, but melanin is absent from the amphipod species unless, perhaps, the eye pigment is a member of this group. In all our observations, typical carotenoid behaviour was exhibited by the pigments extracted, and the animals would therefore appear to owe their visible coloration to carotenoid material almost if not quite exclusively.

A total of 2·9 g. of Gammarus pulex were extracted with acetone and the pigment transferred to petrol ether and partitioned against 90 % methanol. More hypo-than epiphasic pigment was present. The epiphasic materials were examined and the hypophasic fraction rejected, to be studied again later on a separate batch of animals. The petrol-ether solution was chromatographed on calcium carbonate and gave a deep, reddish brown band at the top of the column. This was developed with 2% methanol in petrol ether and gave rise to a pale pink band in the lower part of the column with a small, pale yellow band below it.

The upper band was eluted and found to be still wholly epiphasic ; it was saponified in the usual way, recovered in petrol ether and again submitted to the phase test. Very little pigment was now epiphasic, but the material that passed into the petrol ether at this stage remained epiphasic in the presence of traces of alkali and of acid and was therefore probably a carotene. The material which was now hypophasic became epiphasic again in the presence of traces of acid but passed back into the alcoholic phase in the presence of a trace of alkali. It was found to exhibit a single absorption maximum at 498μ μ in carbon disulphide (cf. the value of 500 μ μ. found for lobster astacin in this solvent by Kuhn & Lederer, 1933). This behaviour is typical of a group of compounds, the so-called carotenoid acids, of which a number have been investigated by various workers (see, for example, the list given by Strain, 1938). These compounds are particularly interesting in that except for a few yeasts and unicellular algae they are not known to occur in plant materials, but appear to be specifically elaborated by the animal organism.

A further sample of G. pulex was extracted, and without previous partitioning or saponification the whole pigment was chromatographed in petrol ether on magnesium oxide. A strong purple band appeared at the top of the column and was developed with 0·5 % methanol in petrol ether. A small yellow band, which probably represents the carotenoid acid and other epiphasic material described in the previous paragraph, separated from the main band and was eluted through the column and discarded. The development was continued with 1·5 % methanol in petrol ether and four bands were obtained in the following order from top to bottom: band 1, pink; band 2, yellow; band 3, pale pink, extensive but containing very little pigment; band 4, rather deep yellow in colour. These were eluted separately and each was found to be wholly hypophasic. After saponification they remained hypophasic in the presence of acids and alkalis and therefore belonged to the xanthophyll group rather than to that of carotenoid acids. The absorption maxima of band 2 after saponification were 506 and 473 μ μ, in carbon disulphide, but these values approximate to those of a number of xanthophylls and the pigment could not be identified further. The material from band 4 showed maximal absorption at 510 and 476μ μ. in carbon disulphide (lutein?). Bands 1 and 3 contained too little pigment for further examination.

The carotenoid pigments of G. pulex therefore include a carotene (probably), and at least four xanthophylls, of which two are present in considerable quantity, and one is perhaps lutein. These are accompanied by a carotenoid acid.

The detection of this carotenoid acid among the products of saponification of those pigments which were epiphasic in the first partition suggested that the original material contained an epiphasic ester of the acid in question. Sörensen (1936) had previously reported the presence in the same species of a hypophasic ester of some carotenoid acid which he considered to be astacin, but in the present work there was no indication of any ester apart from the epiphasic material already mentioned.

There is considerable confusion in the literature on the subject of carotenoid acids, especially in the case of the free and esterified forms of the only one which has been exactly identified and of which the constitution has been definitely established, namely, astacin, a tetra-keto-β-carotene. Kuhn & Lederer (1933) reported that an epiphasic ester of astacin can be extracted from the scarlet hypodermis of Astacus gammarus and a hypophasic ester from the eggs of the same species, both substances being readily transformed into astacin by saponification. The subsequent work of Kuhn & Sorensen (1938) has, however, made it clear that the ‘hypophasic ester’ is in reality not an ester at all, but a dihydroxydiketo-β-carotene to which the name astaxanthine was given. This substance is freely autoxidizable to yield astacin itself in alkaline solutions such as are commonly used for the saponification of xanthophyll esters. The epiphasic ester is probably not an ester of astacin, therefore, but of astaxanthine, astacin itself being an artefact.

Comparing the present results with those of Sorensen it may be suggested that the epiphasic material was an ester of astaxanthine, and that the hypophasic material described by Sorensen was either free astaxanthine (which is very improbable) or else astacin derived from it in the course of his manipulations which, unfortunately, are not described in any detail in his paper. For present purposes, however, the important point seems to us to be the demonstration of the presence of a carotenoid acid, or of some ester of such an acid, in the tissues of Gammarus pulex. Further work on much larger quantities of material will be necessary for its exact identification.

Turning now to the second species examined, several hundred G. marinus, weighing about 5 g., were collected from the shore at Heacham, Norfolk. They were extracted with methanol in the usual way, becoming completely decolorized in the process apart from the eyes, which remained black as before. The fine orangeyellow extract was treated with petrol ether and water and the pigments quantitatively transferred to petrol ether and partitioned against 90 % methanol as usual. More pigment was hypo-than epiphasic.

The epiphasic fraction was washed with 90 % methanol until no more colour was extracted, and the remaining wholly epiphasic material chromatographed on a column of magnesium oxide from solution in petrol ether. A large pink band appeared in the upper part of the column and gave rise on elution to a pigment which showed maximum absorption at 495μ μ in petrol ether. Below it was a smaller brownish red band, the pigment of which showed an absorption maximum at 490μ μ. in petrol ether. After saponification the pigment of the lower band was almost wholly hypophasic in the presence of alkali but epiphasic in the presence of traces of acid. The carotenoid pigments of G. marinus, like those of G. pulex, thus include a carotenoid acid, again in the form of an epiphasic derivative.

Seventy-three large specimens of Orchestia gammarella, weighing 2·7 g., were quietened with chloroform vapour and extracted with methanol. Again the animals were quite decolorized apart from the eyes, which remained black. The pigments were found to be wholly hypophasic, and after recovery in petrol ether were chromatographed on calcium carbonate and the chromatogram developed with petrol ether. Four well-separated bands rapidly appeared, and of these, three were rather small and pale yellow in colour. These were discarded. The remaining band, the third from the top of the column, was large and orange-yellow in colour, and was eluted and saponified. It was then found that the pigment was wholly hypophasic in presence of alkali, whereas in presence of a trace of acid a small part became epiphasic. It may be concluded that some xanthophyll was present together with a little carotenoid acid.

We may conclude therefore that the external colour of Gammarus pulex, G. marinus and of Orchestia gammarella is wholly due to carotenoid pigments. Other pigments are present in traces at most, and the black colouring matter of the eyes is perhaps a melanin. In contrast to the Isopoda (Baldwin & Beatty, 1941), all three Amphipoda gave evidence of containing acidic carotenoid material, i.e. they contain endogenous as well as exogenous carotenoid pigments.

The Amphipoda include many cave types, as may be seen in the summary in Spandl (1926), the majority being either white or colourless. Specimens of the large cave species Niphargus (Stygodytes) balcanicus Absolon and another large Niphargus species from caves in the Popovopolje, Herzegovina (Jugoslavia), were examined. (I am indebted to Dr. I. Gordon for advice that the second species corresponds as far as is known to the still incompletely described Niphargus (Antroplotes) herculeanus Absolon.) They were whitish in appearance and yielded no trace of pigment to the alcohol in which they were preserved. Some specimens of a small Niphargus species from the Postumia Grotte, north Italy, were extracted with methanol, and any possible pigment transferred to a very small volume of petrol ether, but no trace of coloration could be observed.

In the case of Asellus it is evident that the cave form has lost melanin, for traces of the typical branched melanin-bearing cells are to be seen in the integument even of very pale specimens. Among the Amphipoda, however, the absence of pigment is complete, and it is difficult to determine what was the ancestral pigment. From our observations on epigean Amphipoda it is to be noted that three species’, chosen at random, all had a wholly carotenoid pigmentation, including in each case a carotenoid of specifically animal type. A much greater range of species and genera will have to be investigated before it will be possible to state definitely whether carotenoid pigments are or are not of general occurrence in the Amphipoda. It is a matter of general experience, however, that amphipod species have a greenish, yellowish, or red pigmentation, which is rapidly removed when the specimens are preserved in alcohol, and it is to be expected that they contain carotenoid pigments. Details of the colouring and pigmentation of some epigean and hypogean isopods and amphipods are given in Table 1.

Table 1.

Colourings and pigmentation of epigean and hypogean Isopada and Amphipoda

Data for epigean species, Asellus aquaticus cavernicolus and Niphargus from Baldwin & Beatty (1941) and present paper. Data for remaining species from Chappuis (1927).

Colourings and pigmentation of epigean and hypogean Isopada and Amphipoda
Colourings and pigmentation of epigean and hypogean Isopada and Amphipoda

The nature of amphipod pigmentation is of special interest in that there is a certain amount of evidence that it is affected by fight. Viré (1900), quoted by Verne (1926), showed that Gammarus puteanus lost its colour when kept in darkness for 20 days. According to the same author (Viré, 1904), this is also true for G. fluviatilis kept in the dark for 6 months. Viré stated (as quoted by Verne, 1926) that the colourless cavernicolous Niphargus puteanus developed a greenish brown colour after 2 months’ exposure to light, but he found no such effect in Niphargus plateaui (Viré, 1900, 1904). Meek (1929) has made the interesting observation that specimens of the epigean form Gammarus duebeni living underground in the Mill Pit at Blyth are a ‘dull transparent white’, but that after exposure to light they develop the normal colour of the species, including the three characteristic red spots.

It would thus appear that light may be necessary for the development of the carotenoid pigmentation of Amphipoda, and that the darkness of the cave environment may provide an explanation of lack of pigment in cavernicolous Amphipoda. The possibility that scarcity of carotenoid pigment in the food of cave Crustacea accounts for lack of pigmentation in the animals has already been ruled out, for the animals are unpigmented whether or not they have access to carotenoid material in their food (Beatty, 1941).

This work was carried out in the Zoological and Biochemical Laboratories, Cambridge, during the tenure of a maintenance grant from the Department of Scientific and Industrial Research ; it was supervised by Dr Ernest Baldwin, of the Biochemical Department, Cambridge. I wish to express my thanks for facilities offered me by Prof. J. Hadii and Prof. R. Kenk of the Univerza Kralje Aleksandra I, Ljubljana, and by the Director and Staff of the Oceanografski Institut, Split, Dalmatia.

Baldwin
,
E.
&
Beatty
,
R. A.
(
1941
).
J. Exp. Biol
.
18
,
136
143
.
Beatty
,
R. A.
(
1941
).
J. Exp. Biol
.
18
,
144
153
.
Chappuis
,
P. A.
(
1937
).
Die Tierwelt der unterirdischen Gewässer
.
Stuttgart
.
Kuhn
,
R.
&
Lederer
,
E.
(
1933
).
Ber. dtsch. chem. Get
.
66
,
488
.
Kuhn
,
R.
&
Sörensen
,
N. A.
(
1938
).
Ber. dtsch. chem. Ges
.
71
,
1879
.
Meek
,
A.
(
1939
).
Nature, Land
.,
124
,
509
.
Sörensen
,
N. A.
(
1936
).
K. norske vidensk. Selsk. Skr
. no.
1
.
Spandl
,
H.
(
1926
).
Die Tienodt der unterirdischen Gewässer
.
Vienna
.
Strain
,
H. H.
(
1938
).
Leaf Xanthophylls
.
Washington
.
Verne
,
J.
(
1926
).
Les pigments dems l’organisme animal
.
Paris
.
Viré
,
A.
(
1900
).
Quoted in Verne (1926)
.
Viré
,
A.
(
1904
).
C.R. Acad. Sci., Paris
,
138
,
706
.
Zechmeister
,
L.
(
1934
).
Carotinoide
.
Berlin
.
Zechmeister
,
L.
&
Cholnoky
,
L. V.
(
1938
).
Die chromatographische Adsorptionsmethode
.
Vienna
.