62 Mr Potts, The Colour Variations of the Fauna, ete. 
a change in habitat is followed by a readjustment of the pigmen | 
altermg the colour scheme but this power is soon lost. 
The case under discussion is similar in kind though not in 
degree. Synalpheus is capable of producing only the two types 
of pigment—the red and purple. These moreover are only laid 
down in the one pattern in which longitudinal stripes are the | 
dominant note. The paucity of pigments points to a limitation | 
in the power of variability of the animal—but at the same time 
by alterations in the width of the stripes an approximate protective | 
effect is brought about. This effect is plainly a direct response : 
to the action of the environment. The general character of the! 
colour pattern is plainly fixed for both species: possibly by natural ‘ 
selection since most of the other crustacean commensals show the | 
same longitudinal striping. It is however difficult to see how natural ; 
selection can have brought about the power, in the individual, of | 
responding to the colour scheme of the host by developing the | 
requisite quality and quantity of pigment. Moreover in this case 
it would seem that even without a close resemblance Synalpheus | 
and the other commensals are adequately protected by their asso- | 
ciation with the crinoids. For on the approach of predatory animals 
the long arms of the crinoid close over and so defend any com-) 
mensals which happen to be on the disc from threatened attacks. 
These arms principally composed of calcium carbonate would be a | 
disagreeable mouthful for any fish and I never met with crinoids. 
with mutilated arms. Moreover the Alpheids are exceptionally 
active and can thus easily evade any unwelcome attentions. 
The commensal Galatheids are much rarer but they too show. 
an alternation of vertical bars of purple pigment with non- | 
pigmented areas. The new genus of prawns, Pontoniopsis, also 
shows a combination of the striped pattern with yellow (or green) 
and brown pigments which makes it comparatively inconspicuous 
on the light coloured crinoids it inhabits. The Myzostomids may | 
exhibit radiating bars of pigment different to the ground colour— ~ 
their resemblance to the host is sometimes very striking. But im 
the other commensals the striping is not very marked or entirely 
absent. 
It is possibly significant that the larger forms, for which | 
protection is more necessary, are generally speaking those which ” 
show the greatest colour variability and the latter feature may | 
be responsible for the almost universal occurrence of Synalpheus — 
upon crinoids in the Torres Straits area. 
A full account will appear in a forthcoming publication of the | 
Department of Marine Zoology of the Carnegie. Institute of | 
Washington. | 
