216 PEECY SLADEN TEUST EXPEDITION. 



Only quite a small proportion of the specimens are accounted for in the last table. 

 By far the larger number were obtained from anchorages close to land. Althougli 

 material uas only found in 5 out of 17 localities situated ia shallow water, these 5 

 yielded in all 148 A. pelagicus, 9 A. valdivue, and the two metamorphosed animals. 

 Particularly favoured was the anchorage off Salomon Atoll, in the Chagos i^.rchipelago, 

 where plankton nets taken in every hour during two nights gave the following 

 numbers : — 



JuneSOth, 6.30 p.Ji.-July 1st, 6.30 a.m. (10 fathoms);— 7, 1, 3(1), 2,0, 5(3), 4(2), 10, 6(1), 2. 1, 0. 

 July 4th, 7.15 r.M.-July 5th, 7.15 a.m. (surface) :— 5, 12[1], 0, 0, 14, 4(2), 0, 0, 11 [1], 35, 7, 2. 



Thus over three-quarters of the whole material Avas taken from one place. 



If Anqjhioxides is only the larva of an Amphioxus-\\kQ form, destined ultimately to 

 metamorphosis and a littoral existence, we should expect to find it in the greatest 

 numbers in the immediate proximity of its breeding-places, which upon this assumption 

 we must look for in the littoral region. That it does occur in great numbers in just 

 such a locality is a strong argument for its larval nature ; at any rate it forces us to 

 abandon the view that it is a j^urely pelagic organism, to which the earlier observations 

 seemed to point. All Goldschmidt's material, as well as the few other specimens 

 obtained before 1905, excepting Cooper's, Avere taken in the open sea ; hence this view 

 seemed to be fairly well founded. 



Cooper (1903) obtained his specimens in stormy Aveather, and regarded them as larvae 

 of some pelagic species driven in accidentally to laud; but that the animals obtained 

 off Salomon Atoll were the chance jetsam of a pelagic species Ave clearly cannot suppose. 

 The fact that Ampldoxkles has only as yet been taken iu large numbers from one 

 locality need not surprise us, since the occurrence of all Cephalochorda is notoriously 

 very local. 



Among the specimens taken close to land, animals between 8 and 9 mm. in length 

 Avere much the most frequent, as the following figures show : — 



Length in ram.— 5-5-5 . 5-5-6 . 6-6-5 . 0-5-7 . 7-7-5 . 7-5-8 . S-S-5 . 8-5-9 . 9-9-5 . 9-5-10 . 10-10-5 

 No. obtained.— 2.8. 7.4.8.8 . 40 (2)[1] . 59(2) [1] . 14(3) . . (1) 



Probably growth becomes considerably less rapid with the attainment of the full 

 number of myotomes and larval gill-slits ; possibly the animals then enter upon a 

 resting stage of some duration, in which growth is inhibited, previous to metamorphosis. 



In connection with the pelagic occurrence of Amphioxides — Ave know that the 

 Branchiostoma larva is also to a certain extent jielagic ; but, so far as I am aware, it 

 has never been taken at any very great distance from land. Not improbably the 

 prolonged larval existence of Amphioxides is an adaptation to the highly jjelagic life 

 Avhich it may be compelled to lead, a life to which the delicate tissues of the Branchiostoma 

 larva Avould be ill-suited ; if so, it is an adaptation which has become fixed in the 

 species, not merely a possible resource of individuals driven out to the open sea. 



The eA'idences for the larval character of Amphioxides will be discussed later. In 

 the foUoAviug account of certain points in its anatomy, this character w^ill be taken for 

 granted. 



