BRACK YUEA FKO:\r T0R11E8 STRAITS. 47 



forms one side of tlic cavity in which tlio jmrasite was lodged, the other side being formed 

 by an expansion of the opposed j^art of the main l)i-anch, at the sides of which lobate 

 projections are beginning to dcAelop. A considerably more advanced gall is shown in 

 fig. 39. Here the gall is terminal in position and is roughly lenticular in form, the two 

 digitate lobes which compose it being perforated by fissures and only touching each other 

 here and there at the edges. A still older gall (fig. lOj, occupying a lateral position on 

 a branch, is closed except for two or three very small apertures at or near the margin. 

 These apertures are not placed regularly opposite to each other as Semper states. On the 

 outer surface of the gall the calicles are rather smaller than those on the normal branches 

 and are not arranged like them in series, but appear to be otherwise well developed. On 

 the inner surface of the gall the calicles, as Semper states, are small, shallow, and have 

 the septa only feebly developed. They are also in some parts distorted and drawn out to 

 an elliptical outline, but this distortion does not appear to l)e definitely related to the 

 marginal apertures of the gall as described by Semper, who attributes it to the action of 

 the current of water caused by the crab. Semper also found on the inner surface " very 

 distinct scars, whicii are evidently produced by continual scratching in (me spot," and he 

 concludes that the crab usually remains in one position within the gall. Such scars are 

 not visible in our specimens. In the older galls the outer surface rises into rounded, 

 irregularly placed swellings and short branches, as if the coral were about to resume the 

 normal habit of growth disturbed by the intrusion of the 2)arasite. 



From Semper s earlier note we gather the not uninteresting detail that the polypes 

 on the mner surface of the gall are colourless. 



As regards the further habits and life-history of Kapalocarcmus we have no informa- 

 tion. The fact that each gall is inhabited by a solitary female, while the male is as yet 

 unknown, would seem to iiidicate that both sexes are at first free-living, and that it is 

 only after impregnation that the female becomes imprisoned in a gall. The fact that 

 the youngest gall observed is of ample size to contain a full-grown Rapalocarciniis tends 

 to confirm this suggestion. 



As regards the systematic position of Iliqxifocarcinm, we have to note in the tirst place 

 its close affinity with the CryptocJiirus (^oralliodijtets of Heller (" Beitr. z. Crust. Famia d. 

 roth. Meeres," SB. Akad. Wien, xliii. (1) 1861, p. 3(56, pi. ii. figs. 33-39). As Semper 

 has shown ('Animal Life,' pp. 217, 221-223)*, Cri/jitoc/uriis, like HapalocarciiiKs, is 

 parasitic on living corals. In this case, however, no closed " galls " are formed, the 

 crab living in massive corals (ex. Ooniastuea,) at the bottom of a funnel-shaped 

 depression, due to an arrest of the upward growth of the coral. The affinity between the 

 two genera is most clearly shown by the third maxillipeds, Avhich in both cases are 

 peculiar in having the merus-joint very narrow and the exopod rudimentary. The 

 structure of the facial region is somewhat similar in both, the antennules not being 

 retractile into fossettes, while the antemaae are very small and the orbits ill-defined. 

 The abdomen of the female in both genera is much enlarged, but in Cryptochirm it lies 



» Simpers figure of Cn/idochinis differs considerably from those given !.y Tleller. He states, however, that the 

 Philippine form " ajjpears to l)e in no respect specifically different '' from that found in the Red Sea (o/;. (it. ]>. 2.S1). 



