TYPE TR AGUE ATA. 523 



The Phylogeny of the Tracheata. It has been the custom to unite 

 together the Crustacea, Arachnida, and Tracheata in a single group, the 

 Arthropoda, characterized by the possession of a chitinous cuticle, by the 

 occurrence of jointed limbs, and by the masticatory organs being modified 

 limbs ; and furthermore it has been customary to consider the Arachnida 

 and the Tracheata as closely related on account of the occurrence in both 

 groups of tracheae. The early processes of development in the three 

 groups also show many points of similarity, though a closer examination 

 shows decided differences in the details. How far convergent evolution 

 may have operated to produce the similarities is the problem to be settled, 

 and it can be settled only by a consideration of all the facts at our disposal 

 which indicate the phylogeny of the various groups, a discussion which 

 would prove entirely beyond the limits of a text-book. 



It has been pointed out that the probable ancestry of the Crustacea is 

 to be found in the Annelida, and that the Arachnida have in all probability 

 descended from Eurypterus-like ancestors, which were certainly Crustacean 

 in their affinities. Are the Tracheata then also descended from the Crus- 

 tacea and from forms which possessed tracheae? Our present knowledge 

 of the group negatives any such supposition ; it seems impossible that the 

 Tracheata should, like the spiders, have descended from Earypterns-\\\iQ 

 ancestors ; it must rather be concluded that the similarities between them 

 and the Arachnida are due to convergent evolution, the embryonic simi- 

 larities to the acquisition of comparatively large amounts of food-yolk in 

 the ova, distributed in a similar manner, and the similarities of the adult 

 to the exigencies of a terrestrial life. The occurrence of tracheae in both 

 groups seems at first an important point of similarity to be accounted for 

 only by a community of descent, but, when it is considered that in the 

 terrestrial Isopoda tracheae also occur in the branchial opercula, and that 

 their occurrence in these forms is a purely secondary adaptation, without 

 any phylogenetic significance, it is evident that their importance as indi- 

 cations of affinity is much reduced. It may also be pointed out that the 

 Malpighian tubules of the Arachnida and Crustacea are endodermal, whereas 

 in the Tracheata they are ectodermal, arising from the ectodermal hind- 

 gut. 



There is little room for doubt but that Peripatus is closely related to 

 the Annelida, and its relationships to the Myriapoda are also pronounced, 

 so that the conclusion seems inevitable that the Tracheata have been de- 

 rived from Annelid-forms, and have therefore a phylogeny practically inde- 

 pendent of that of the Arachnida. However, it is possible that the Anne- 

 lid ancestors of Peripatus and those of the Crustacea were more or less 

 closely related, and that certain of the general similarities of all the three 

 groups are thus to be accounted for, though to what extent we are not at 

 present in a position to judge. One point, namely, the occurrence of com- 

 pound eyes of similar structure in both groups, seems worthy of considera- 

 tion, since it seems to be unexplainable by this hypothesis and to be a re- 



