196 



SCIENCE. 



[Vol. I., No. 7. 



water, as if feeling and fishing for pre}', while 

 the open mouth projects vertically' upwards. 

 It also climbs the sides of the aquarium, using 

 its feet like a starfish. 



One of the most interesting deep-sea forms 

 is Tesserantha conuectens, one of the simplest 

 and oldest representatives of the family Tes- 

 saridae. 



In his System der Medusen, Haeckel has 

 given his reasons for regarding this family as 

 the primitive form from which all the Acras- 

 peda are descended, and he has given a figure 

 of this species in the same place. Tesseran- 

 tha is little more than a Scyphostoma, which 

 becomes sexuall}' mature in this larval stage. 

 Unlike a Scj'phostoma larva, it is a locomotor 

 form, which Iras become adapted to a free- 

 swimming life by the change of its oral disk 

 into a sub-umbrella, and its basal peduncle 

 into an apical process. In place of the sim- 

 ple peripheric gastric space of the Scyphos- 

 toma, it has a chamber divided hy partitions 

 into four radial pouches. This interesting 



Pectythis asteroides, anchored on its back, maanifled ten diame- 

 ters. Outline-sketcli from Haeckel's drawing of the living 

 animal, Deep-sea Medusae, Pi. 8, Fig. 7. 



medusa, which is undoubtedly a deep-sea form, 

 was captured in the South " Pacific in 2,160 

 fathoms of water. 



A magnificent specimen of Periphylla mir- 

 abilis, a mature male, was captured by the ex- 

 pedition, near New Zealand, in 1,100 fathoms 

 of water ; and it has furnished Haeckel with 

 the material for a minute and valuable descrip- 

 tion (illustrated by eight plates) of the anatomy 

 of this remarkable family, which shows manj' 

 points of close relationship to the very simple 

 and primitive Tessaridae and to the Cucernar- 

 iadae, although it is in other respects the 

 most highly organized of the coelenterates. 



Half of the eighteen species of Medusae 

 in the collection were Craspedotae, and half 

 Acr'aspedae ; and, as they represent eighteen 

 genera and thirteen families, the}' pre'sent a 

 great range of diversity, and represent most 

 of the important types of medusa structure. 

 Haeckel has thei-efore pi'efaced his descrip- 

 tion by a general introduction, which sets forth 

 briefly and clearly the present state of our 

 knowledge of the anatomj', histology, embry- 



ologj', and systematic zoology of the Medusae 

 as a whole. This introduction, written in 

 English, is of great value to those who are not 

 specialists, but yet wish to know the results of 

 modern research on this subject. It is only 

 proper to point out to such readers the fact, 

 that the paper contains man}' statements which 

 are not accepted, without qualification, b}' all 

 naturalists : such as the assertion (on p. xxv) , 

 that, " as the formation of the gastrula by 

 invagination of the blastula in the Medusae 

 has been observed in very different groups, 

 we may assume that it happens universall}' in 

 this class; and supposed exceptions (e.g., 

 Gerj-enia) are founded on erroneous observa- 

 tion.." Most embrj'ologists would certainly 

 hesitate to believe, without verification, that 

 MetsehnikoflF's careful study of the develop- 

 ment of Liriope involves a fundamental error ; 

 and manj' would be disposed to douljt whether 

 the statement on p. sv, that the Ctenophorae 

 are derived from an Anthomedusa (Ctenaria) , 

 is full}' proven. 



The presence of a number of uncorrected 

 typographical errors also detracts from the 

 value of the paper for general readers. For 

 instance : p. viii contains the stateinent, that, 

 " as regards the two sections or sub-classes, 

 the Craspedotae are more probably of mono- 

 ph}'litic origin ; the Acraspedae, of polyphy- 

 litic ; " while other sections especially devoted 

 to this point (11 and 14) show that the author 

 really holds the opposite view, and believes 

 that the Craspedotae are of polyph3-litic, and 

 the Acraspedae of monophylitic origin. 



Haeckel's very extensive and minute ac- 

 quaintance with all forms of Medusae qualifies 

 him, to an e.xceptional degree, for speculating 

 upon the origin and ancestral relationship of 

 the various orders and families ; and his at- 

 tempt to trace the evolution of the various 

 forms is therefore interesting to all zoologists. 

 In sects. 10-14 he gives a phylogenetic classi- 

 fication of the Medusae, the outline of which 

 is essentiallj- as follows : the scypy-polyps and 

 hydro-polyps diverged from each other ; and 

 the latter became evolved along three diver- 

 gent lines, thus giving rise to the tubularian 

 hydroids, the campanularian hydroids, and a 

 third imaginary ' trachylariau ' hydroid, before 

 any true Medusae were evolved. The Acras- 

 pedae are the descendants of the scypy-polyps, 

 of which their Scyphostoma larva is the onto- 

 genetic recapitulation ; while the three great 

 groups of Craspedotae are the independent 

 descendants of the three kinds of hydro-polyps, 

 — the Anthomedusae (e.g., Margelis) , from the 

 tubularian hydroids ; the Leptomedusae (e.g., 



