792 



SCIENCE. 



[Vol. II., No. 4ti 



wliich is purely respiratory. Thoy also differ from 

 Nebalia in the iMSsession' of tli.at distinctively mala- 

 costracan organ, a tail-fin, made up of a teKson and a 

 pair of swimmerets. 



The relationship of Nebalia to the Malacostraca on 

 the one hand, and to the phyllopods on tlie other, 

 h.as long been I'ecognized, and Clans has even gone 

 so far as to hold that this form is a true malacoslra- 

 can; but Boas believes that it is neither a true 

 malacostracan, nor the phyllopod from which the 

 Malacostraca originated, but simplj' the nearest liv- 

 ing ally of this ancestral form. 



He believes that the pi-esenceof a great mantle-like 

 carapace, of eight iinspecialized, broad cormopods 

 with leaf-like exopodites, of a furcated abdomen with- 

 out tail-fins, and of eight abdominal somites, show 

 that it is not a malacostracan, but a phyllopod. As 

 many phyllopods, such as Limnetis and the Clado- 

 cera, have, like the Malacostraca, an exopodite on 

 the second antenna, we must believe that the Mala- 

 costraca have inherited this feature from their phyl- 

 lopod ancestor; and, as it is absent in Nebalia, this 

 form cannot be the direct ancestor of the Malacos- 

 traca. So, too, the fifth and sixth pairs of abdominal 

 feet are rudimentary in Nebalia, while they are well 

 developed in nearly all Malacostraca. As most of the 

 phyllopods, and some of the Malacostraca, leave the 

 egg as a free-swimming nauplius, we must believe 

 that this was true of the phyllopod ancestor of the 

 Malacostraca; but as Nebalia does not pass through 

 a free nauplius stage, but leaves the egg in a more 

 advanced condition, it cannot be in the direct line of 

 evolution. Boas therefore concludes that Nebalia is 

 a true phyllopod, and that the Malacostraca have 

 originated from a form somewhat different, although 

 Nebalia is the closest living ally of this ancestral 

 form. 



Having thus tracedthe decapods back through the 

 Euphausiacea to a phyllopod ancestor very similar to 

 the recent Nebalia, we have now to trace the ancestry 

 of the other Malacostraca. Boas holds that the squil- 

 loids are a branch from the Euphausiacea, and that 

 the Mysidacea have been derived from the Euphau- 

 siacea along still another line of descent, and have, in 

 their turn, given rise to all the remaining groups of 

 Malacostraca. 



The Mysidacea differ from the Euphausiacea and 

 the decapods in many features which they show in 

 common with the Cumacea and the amphipods and 

 isopods; and it is not diiBcult to show, that, in these 

 points of difference, the Euphausiacea are the primi- 

 tive group, and the Mysidacea the modified group. 



In Euphausia, as in the swimming decapods, the 

 body and abdomen are compressed; while they are 

 flattened and rounded in the Mysidacea, and the tip 

 of the abdomen is directed backwards, lacking the 

 peculiar bend of Euphausia and Penaeus. 



The structure of the mandible is very instructive. 

 In Mysis, as well as in the Cumacea and amphipods 

 and isopods, the mandible is forked, the cutting part 

 being widely separated from the crushing part; and 

 between the two there is a row of setae, and a pecul- 

 iar accessory appendix. In Euphausia and the deca- 



pods the appendix and row of setae are absent, and 

 the chewing part is hardly separated from the crush- 

 ing part. In Mysis, as in Cuma and the amphipods 

 and isoi>ods, the palp and exopodite of the first max- 

 illa are absent, and the laciniae are turned forwards 

 as well as inwards ; and in all these forms the laciniae 

 of the second maxilla are directed forwards. They 

 overlap, and the lacinia interna is undivided. In 

 Euphausia, the decapods, and squillas, there are no 

 brood-pouches; but these structures are present in 

 Mysis, as well as in the Edriophthalmata, and they 

 are formed in essentially the same way in all, — by 

 plates which are developed on the bas.al joints of cer- 

 tain of the cormopods. In all these forms the young 

 pass through a long metamorphosis within these 

 pouches. The liver is comparatively simple. There 

 are no sperm atophores, and the spermatozoa have 

 tails. The Cumacea are regarded by Boas as a greatly 

 modified offshoot from the Mysidacea; and the am- 

 phipods and isopods are derived from an ancestral 

 form somewhat like, but more primitive than, the 

 living Cumacea. 



As regards the position of the amphipods and iso- 

 pods, Boas's view is directly opposite to that which 

 has been generally accepted ; as he regards these as 

 the most highly specialized and divergent of the 

 Malacostraca, instead of low and primitive forms. 

 The conspicuous segmentation of the nervous system, 

 the absence of a carapace, the sessile position of the 

 eyes, the great number of similar somites, the worm- 

 like shape of the body, and the elongation of the heart, 

 — all seem at first sight to show that these forms are 

 ancient and low. Boas points out that the nervous 

 system gives no proof of a primitive condition, as 

 there are as many independent ganglia in Mysis as 

 there are in the sessile-eyed Crustacea. It is true that 

 the heart is longer than it is in Mysis ; but there are 

 only three pairs of ostia, and the length of the heart, 

 as compared with that of the mid-body, is no greater 

 thair it is in Mysis. As the eyes are stalked in Neba- 

 lia, the nearest ally of the Malacostraca, all of the 

 latter must have inherited stalked eyes from their 

 phyllopod ancestors, and the sessile eyes of the Edri- 

 ophthalmata must be due to secondary modification. 

 So, too, regarding the absence of a carapace. As the 

 Malacostraca inherit this structure from the phyl- 

 lopods, those forms in which it is absent must h.ave 

 lost it by secondary modification. The same thing is 

 true of the absence of a scale on the antenna. There 

 is, therefore, no proof that these animals are primitive ; 

 and the many points of resemblance to the Mysidacea 

 which we have just noticed show the close relation- 

 ship between these groups. But as the Mysidacea, 

 like Euphausia and the decapods, have stalked eyes, 

 a carapace, and a fused mid-body, exopodites in first 

 maxillae, exopodites and palpi in second maxillae 

 and on cormopods, and as a seventh abdominal seg- 

 ment. is present, we must believe that the Mysidacea 

 are the more primitive group, and the Edriophthal- 

 mata their recently modified and highly specialized 

 descendants. 



Boas believes that most of these differences are due 

 to the fact that the Edriophthalmata have become 



