610 MB. F. E. BEDBABD ON THE [May 22, 



The large number of reputed species of this genus is due to two 

 principal causes. In the first place, it has been tacitly assumed that a 

 Perichceta coming from the New World would be certainly different 

 from an Eastern form. "We now know that this is not the case, 

 and that in all probability there are no indigenous species outside 

 of the Malayan Archipelago and the adjacent parts of the Asiatic 

 mainland. In the second place, sufficient attention has not been 

 paid in the past to the great variability which is found among the 

 species of this genus. I direct particular attention as regards 

 this matter to the large series of variations seen in such a protean 

 species as Amy Mas haivayanus or A. stelleri. Gradated specimens 

 connect the extreme forms which, when described singly, have been 

 referred to different species, a quite necessary conclusion in view 

 of the facts then known. 



I have naturally, therefore, found great difficulty in deciding 

 upon the value of specific characters in the following survey of 

 the species of the genus. There are some characters which seem 

 to be of first class importance and to be not subject to variation. 

 Those characters are : — (1) The seta? on the clitellum, when these are 

 present on every segmeut and form continuous rows. Otherwise, 

 in those cases, such as A. hawayanus, where the seta? are at best 

 but few, they vary in number and in the segments upon which they 

 occur. In such forms as A. violaceus, A. taprobance, there is no 

 variation. (2) The presence or absence of a rounded muscular 

 sac, often containing a protrusible penis, into which the duct of the 

 spermiducal gland opens. This is either present or absent in a 

 given species, definitely present or absent. (3) The caeca again 

 seem to be always either present (in the majority of cases) or 

 absent, with the apparent exception (?) of A. merabahensis. 

 (4) The spermathecse again are, with not many exceptions, 

 constant in number of pairs and position of segments. In one or 

 two instances (A. dyeri) their very dorsal position seems to be a 

 valid specific difference. (5) The close approximation of the male 

 pores in a few species (e. g., A. iris, A. hahnahera) is a good 

 character. But a slight difference in the remoteness of the two 

 pores must be used with the greatest care, if at all. I have not 

 made use of it here, as I am not convinced that it is possible to 

 eliminate from such slight differences the effects of differing degree 

 of contraction. (6) The position of the last "heart"; this is 

 always in segment xii. (A. taprobance) or in xiii. (the vast majority 

 of species). The only species which varies at all is A. stelleri. 

 (7) The presence or absence of septa in the segments occupied by 

 the gizzard. As a rule those septa (viii./ix., ix./x.) appear to be 

 absent. But in some species, one or both are present and even 

 quite thick. In other cases, there is one very thin septum or not 

 more than traces. The character must therefore be used with 

 great care : its use is marred by the fact that there are not 

 always exact observations upon the point on record. (8) Barely, 

 e. g. A. neoguinensis, the testes, funnels, and sperm-sacs are present 

 to the number of one pair only. This seems to be a non-variable 

 characteristic of those species in which it is met with. (9) The 



