284 THE TANGANYIKA PROBLEM. 



and nearly quadrilateral. Only a single tooth, however, is present on the antero- 

 lateral margin, in addition to that at the outer angle of the orbit. With this the 

 description of Platythelphusa, which actually comes from Lake Tanganyika, agrees 

 in the main, but the antero-lateral margins are, in contradistinction, multi-dentate. 

 Several figures of this form are given, but they are not, unfortunately, all one could 

 wish. The figure of the antennae suggests that we are dealing with a simple undis- 

 torted condition of the joints, such as I have seen nowhere else but in Limno- 

 thelphusa, but the right and left antenna? do not even agree one with another, 

 according to the drawing. The fourth pair of walking legs presents a peculiarity in 

 being rather short, while the terminal joints are somewhat flattened and expanded, 

 presumably for swimming purposes. 



Of the manner in which Limnothelphusa attained its present distribution in Lake 

 Tanganyika there are two possible views. Either from a land Thelphusan it has 

 become converted gradually into a wholly aquatic type, or it may have entered the 

 lake more or less directly from the sea, at some time when a connection between 

 them was far more close than at present. It is generally accepted that the Land- 

 Crabs have descended from ancestors with a littoral habit, so that there would be no 

 direct objection to the supposition that this creature has merely retained its primi- 

 tive aquatic character, rather than regained it after adaptation to a terrestrial mode 

 of existence. We can only come to a conclusion on this head by estimating how 

 far the general structure of the animal suggests simplicity on the one hand, or, on 

 the other, specialization. The arched or vaulted condition of the branchial regions 

 of the carapace in Thelphusa is evidently a specialization in connection with aerial 

 respiration. That such prominent vaulting does not here exist is not surprising, but 

 though it is perhaps conceivable that this character, once attained, might be lost 

 again on change of environment, it is, I think, more probable that such a condition 

 was never reached by Limnothelphusa. Again, as regards the less prominent 

 deflection of the front in the latter, the condition appears rather primitive than 

 secondarily acquired ; while the simple nature of the second antennal joint, as 

 compared with that of Thelphusa, which so much suggests a distortion produced by 

 the frontal downgrowth, also supports this view. The greater number of spines 

 occurring on the antero-lateral margins is a further feature, capable, however, of 

 two possible interpretations. The carapace in but few species of Thelphusa bears 

 more than one, and that a less prominent spine. If, then, we are dealing with a 

 multiplication of marginal spines, we have an indication of greater specialization 

 than that met with in Thelphusa, an indication contrary to the tendency of the 

 other evidence. The other possible explanation, then, that a reduction towards the 

 extreme condition of Thelphusa is in progress, would seem far more probable ; and 

 it is a noticeable fact that the marine and littoral Crabs, from which we may suppose 

 this form has been derived by comparatively slight modifications, are far more 

 spinous than any of the modern terrestrial or fluviatile forms. Thus, while Platy- 

 thelphusa and Limnothelphusa would appear to be the most primitive of these Old 

 World genera, Parathelphusa, in the less pronounced arching of the carapace and 

 the more numerous lateral spines, would come as intermediate between them and 

 the most specialized condition of Thelphusa. 



On the causes which have contributed to the present-day distribution of these 

 genera, a word or two may be said. It is no very recent conception that Mada- 



