TRANSACTIONS OF SECTION D. 



753 



him in liis most valuable speculations. In one matter, however, I would venture to 

 express a diil'erenco of opinion. He regards the littoral forms of invertebrates as 

 miirratinjr into the deep sea by the followin<r process : Their free-swimming larviB 

 are supposed to be carried out by currents far from land, and then, having com- 

 peted their development, to sink to the bottom, wliere a very few survive and 

 thrive. It is hardly to be conceived that any animal, especially in a young and 

 tender cmidition, could suddenly adapt itself to the vast change of conditions 

 t'litailtd in a move from littoral to deep-sea life, ft seems to me much more 

 likelv that the move of animals from the shallow to the deep sea has been of the 

 most pTiulual kind and spread over long series of generations, which may have 

 mi<rrated downwards, perhaps a fathom or so in a century, partly by very slight 

 miirrations of tiie adults, partly by very short excursions of larvte. Thus alone 

 l)v almost insensible steps could animals, such as those under consideration, be 

 «>iiabled to survive an entire change of food, light, temperature, and surroundings. 



The following Papers were read : — 



1. On flu' Georjraj^Mcal Dii^frihufi'on of the Mncrv,rous Crustacea, 

 By C. Si'ExcE Bate, F.E.8. 



The Man'ura may conveniently be classified into three natural groups of* 

 appiireiitly equal importance in size and structural character. These differ from 

 oaili otlier in anatomical details, even where they approximate in external form ; 

 iiuil they are essentially distinct in consanguinity by a large history in tlu'ir develop- 

 Mier.t and growtli. 



The distinction is more conveniently exhi])ited in the structure of the branchial 

 apparatus, which differs importantly, and assumes three different forms. 



In the Astaciden — the lobster and crayfish division — the branchiiB are formed 

 its a mass of cylindrical tubes varying in hnigth, but thickly grouped together in 

 a ;;Teater or less mnnljer of rows opening from ont> central stalk. 



In the I'ouridiut the branchia? consist of filaments dividing into branches, 

 snuetinies dicliotoniously and sometimes unequally. These tilanients are generally 

 cyiindriciil in form, but in some genera they are so closely arranged that they 

 liecome eomijrersed into plates of considei'able tenuity, still retaining, however, 

 their divided or branching character. 



hi the Cm-idea — the prawn and shrimp section — the branchia? are in the form 

 e.f broad plates of extreme tenuity, arranged similarly to those that exist in crabs 

 ";■ the short-tailed Crustacea, the entire plume being suspended by the centre to 

 that portion of the animal to which it belongs. 



With eacli of tliese there is a corresponding condition of the walking legs. 



In tlie Astaoidca all have a tendency to possess a chelate condition, of which 

 the tirst pair is the largest and the third the smallest, while the posterior is 

 usually chelate only in the females. 



A departure froiu this exists in those genera that l)elong to the aberrant family 

 lif Stewipidcp, which possess the branchia of the Astaciden, the pereiopoda of the 

 I'nmiha, and the develojiment of the Coridea. 



In the Penroidoa the three anterior pairs of pereiopoda only are chelate, and this 

 feature increases in importance posteriorly, and in the depauperised species the first 

 and second pairs depart from that condition, whereas the third, however feeble, still 

 retains the chelate character; the fourth and fifth pairs are never chelate, gener- 

 ally enfeebled, rudimentary, or obsolete. 



In the Caridea there are never more than two pairs of legs chelate, and the 

 P'v*terior three pairs are invariably simple and, generally, robust and eflicient organs 

 ■>' a pedifonn character. 



With each of these three divisions there is more or less constant condition of 

 development. 



Inthe^-ls/rtc/dpfftheembrvo always leaves the ovum in a more or less perfectly 

 developed niegalop condition^ with tlie exception of those genera that belong to the 

 'Sfoinpidfp which are hatched in the zo.nca stasre. 



18^1.. ^' 3 c 



