INVERTEBKATA, CRYPTOGAMIA, MICROSCOPY, ETC. 945 



The ancestor of the two groups is to he found in the Mysis-stage 

 of the Decapoda (Fritz Miiller). The ancestors were pelagic forms, 

 with three flagella to the superior antennas, and with biramose ab- 

 dominal swimming feet in both sexes. So far as the present existing 

 Mysidse are concerned the third flagellum is only retained in the 

 male of Podopsis. This is somewhat remarkable, as this genus is 

 one of the most retrograde of its group. In the other genera there 

 is no indication of it whatever. Among the Macrura, Palcemon has a 

 third flagellum, more or less well developed, and in the more lowly 

 rejiresentatives it is still well marked. The abdominal feet are still 

 swimming organs in the Macrura, but in forms of Mysidae which were 

 examined, those parts never had that function in both sexes. In the 

 former the right and left mandibles are equal, but in all Mysidae they 

 are unequal, and are generally very different. 



Basing his argument on the conclusion to which he has arrived, 

 that where the male differs most from the primitive form, the group to 

 which it belongs is progressing, and that, on the contrary, where the 

 female exhibits the most marked divergence, the group is retrograding, 

 the author concludes that the Mysidae are degenerating. This may 

 be shown by the abdominal appendages, for in the male there is a 

 gradual series of atrojjhy, while in the female they are nearly always 

 comi^letely rudimentary ; so, too, the male Mysidae often retain their 

 pelagic habitat. 



The author concludes with an indication of the characters by 

 means of which the relations of the different genera, and their history, 

 are to be made out. 



Nest-building Amphipods* — Mr. S. J. Smith, in a paper on some 

 Amphipods described by T. Say, states that the tubes which certain 

 species make to live in are to a great extent formed of pellets of their 

 excreta. 



In 1874 he watched carefully the process of constructing the 

 tubes in several species of Amphipoda. Microdentopus grandimaniis 

 (M. minax Smith) was a particularly favourable subject for observa- 

 'tion. 



When captured and placed in a small zoophyte trough with small 

 branching algae, the individuals almost always proceeded at once 

 to construct a tube, and could very readily be observed under the 

 Microscope. A few slender branches of the alga were pulled toward 

 each other by means of the antennae and gnathopods, and fastened by 

 threads of cement spun from branch to branch by the first and second 

 pairs of perajopods. The branches were not usually at once brought 

 near cnougli together to serve as the framework of the tube, but were 

 gradually brouglit together by jjulling them in and fastening tliom a 

 little at a time, until they were brought into their proper position, 

 where they were firmly held by means of a thick network of fine 

 threads of cement spun from branch to branch. After the tube 

 had assumed very nearly its completed form, it was still usually 

 nothing but a transparent network of cement threads woven among the 



♦ ' Trans. Connect. Acud.,' 1860. Soo ' Nature,' x.\ii. (1880) p. 51)5. 



