1907.] CRUSTACEA OF THE THIRD TANGANYIKA EXPEDITION. 263 



varieties, and, on the other hand, there can be no doubt that many 

 so-called species of other writers are not more distinct than these 

 varieties of Schenkel. If it is possible to trace a gradual 

 transition from one variety to another, as appears to be the case 

 with the form celehensis, it becomes evident that certain established 

 species are of similarly little value, and that species of this genus 

 ought never to be constituted on such slender grounds. 



This matter is dealt with at such length, because the conviction 

 has forced itself upon me that the number of African species, at 

 all events, has been multipHed to an excessive degree. Where 

 two or more well-marked species exist side by side, there can be 

 no hesitation in the matter ; but our knowledge of the Crabs of 

 the African continent is still very scanty, and over a vast area 

 restricted to a few isolated records. Thus it is impossible to 

 determine yet whether this or that species may not pi-ove to be only 

 a slight local modification of some widely distributed form. _ This 

 being the case, I have not ventured to name the two specimens 

 fi'om Nyasa, which are placed under the head-line above. 



Although it was during my study of the two species described 

 as new in this paper that I became impressed with the unsatis- 

 factory character of some species of Potamonautes, it was not 

 until i examined these two Nyasa specimens and a third from 

 Tanganyika that I found a case where the most careful comparison 

 only resulted in confusion with several closely allied species. The 

 two specimens from Nyasa — a rather small male and a small 

 female — are in the collection of the British Museum (Reg. No. 

 97.4.29.2-3), and the following note is given as to locality: — 

 "Kondowe to Karonga, N. Nyasa, June 1896. A. Whyte 

 collector — pres. by Sir H. H. Johnston." 



Using as a basis the key in Miss Rathbun's recent revision, 

 we find they belong to group c, where the lateral margins extend 

 less than an orbit's breadth beyond the external orbital angle. 

 Further, they would seem to belong to division e', where the 

 post-frontal crest is little advanced in the middle, and to sub- 

 division /', with fingers of the cheliped but slightly gaping. 

 Thus they woidd come nearest to the species hilgendorfi (Pfeft'.) and 

 johnstoni (Miers). The specimens difier from hilgendorfi (Pfeff.) 

 principally in possessing a furrow on the ischium. From johnstoni 

 they difier in being dis'tinctly longer in proportion to the breadth, 

 in the sculpturing being less marked, and in the absence of the 

 fairly stout spinules on the anterior margin of the carpus of the 

 cheliped— a single distal spine only being present. 



They agree to some extent also with suprasidcaius pseudo- 

 perlatus, which, if it exist at all as a true variety, is in a measure 

 intermediate between hilgendorfi (Pfefi".) and johnstoni. The 

 length in proportion to the breadth is, however, more extreme in 

 the case of siqmisidcatus pseudoj)erlatus . From another important 

 type in group c—perlatus itself— these individuals do not diflier 

 greatly, "the carapace being of much the same proportions as 

 regards length and breadth!" The width of the front is, however, 



