CLASSIFICATION 6F THE CEPHALOCHORDA. 291 



arrangement in A. bassanum. In view, therefore, of its 

 resemblance to A. eultellum on the one hand and to A. 

 bassanum on the other, it seems well to recognise A. 

 maldivense as a distinct species. A. lucayanum is also 

 a well marked species, bnt it is doubtful if A. caudatum 

 is morei than a variety of A, lucayanum. A. caudatum 

 is only known from two specimens both of which are 

 larger than the average length of A. lucayanum which 

 is 19 mm. according to Kirkaldy and 16 mm. according 

 to Andrews, while A. caudatum measured 20 mm. to 28 

 mm. F. Cooper 1 finds that the length of A. lucayanum 

 from the Indian Ocean is 18-20 mm., which is about the 

 size of A. caudatum as observed by Willey. The differ- 

 ence in size is, therefore, of no account, and this brings 

 the two species still closer to one another. 



A comparison of the formula? of the two species shows 

 a close affinity. 



A. caudatum 40, 9, 11 = 60 Willey. 

 44, 9, 11 = 64 Willey. 



A. lucayanum 42, 8, 12 = 62\ 



43, 8, 12 = 63 



44, 8, 12 = 64 r Andrews. 

 44, 9, 11 = 64 

 44, 9, 12 = 65' 

 44, 9, 13 == 66 Andrews and F.Cooper. 



The second formula of A. caudatum agrees exactly with 

 A. lucayanum, while the first may only be an individual 

 variation. 



The two species agree exactly in other characters 

 except that the rostral and caudal fins of A. caudatum 

 are more deeply constricted off than in A. lucayanum. 

 I do not think that these slight differences can have 

 weighed very much with Dr. Willey in forming this new 



i Fauna and Geography of Maldives and Laccadives. Vol. i., Part 4. 



