ZOOLOGY AND BOTANY, MICROSCOPY, ETC. 567 



general facts of colony-forming and classification, in a third chapter 

 the anatomy, in a fourth the conditions of life. The detailed classifica- 

 tion and the phylogenetic probabilities form the subjects of the conclud- 

 ing chapters. Herr Krapelin does not believe in the existence of a 

 continuous phylogenetic series including all modern forms. The 

 ctenostomatous genera Victorella, Poftsiella, and Paludicella stand in 

 close relationshijD ; the group of Phylactoleemata has arisen from 

 Paludicella-1\]LQ Ctenostomata, starting from Fredericella. Among the 

 higher Phylactolsemata parallel diiferentiation may be observed ; thus 

 the genera Lophopus, Pectinatella, and Cristatella form each in their way 

 the terminal point of a series. The greatest advances in the phylogeny 

 are marked by the families Fredericellidte, Plumatellidse, and Crista- 

 tellidee. The genera Plumatella, LopJiojnis, and Pectinatella are insepar- 

 able, and must all be referred to the family PlumatelUdas. A diagnostic 

 table of the genera is appended. 



Arthropoda. 



Embryology of Insects and Aracknids.* — The late Mr. A. T. Bruce, 

 from his observations on the development of Insects and Arachnids, was 

 led to certain views as to the relations of tracheates. He was of opinion 

 that Peripatus and the Myriopoda, from the absence of wings and other 

 primitive characters, may fairly be considered the most primitive 

 tracheates. Some Myriopods exhibit indications of a hexapod stage in 

 their development, and they may, therefore, be related to the wingless 

 Hexapods. The mode of origin of the endoderm is not very important 

 for classificatory purposes, as it is very likely modified by the presence 

 or absence of food-yolk. The mesoderm of Peripatus grows forwards 

 from an undifferentiated cell-mass at the posterior end of the embryo ; 

 the mesoderm arising from the " primitive cumulus " of Spiders also 

 grows forward from an undifferentiated cell-mass at the posterior end of 

 the embryo. But this resemblance must not be taken to indicate any 

 close relationship, for in the Crustacea the mesoderm has a similar mode 

 of growth. In the higher insects the yolk-cells appear to rej)resent the 

 inner layer of the gastrula, and are consequently equivalent to the 

 endoderm of lower forms ; the true endoderm is functional only during 

 embryonic life in absorbing the yolk, and takes little or no part in the 

 formation of the digestive tract. In these tracheates the layer which 

 corresponds to the mesoblast of Arachnids and of Peripatus has usurped 

 the functions of the true endoderm. 



In endeavouring to separate the different divisions of the Arthropod 

 phylum, anatomical characters as well as embryological phases must be 

 taken into consideration. The possession of a single well-develoj^ed 

 pair of antennse, of tracheal invaginations, and of embryonic membranes, 

 together with the existence of a hexapod stage in their development, 

 afford sufficient ground for regarding Myriopods as lowly-organized or 

 degenerate Insects. Peripatus perhaps belongs to the same category, 

 but its embryonic membranes do not appear to correspond fully to those 

 of Insects. Arachnids, in all probability, never possessed antennte, for 

 all their appendages, like those of Limulus, are at one period post-oral, 

 and are not innervated by the supra-oesophageal ganglion. 



* ' Observations on the Embryology of Insects and Araclmids,' 4to, Baltimore, 

 1887, 31 pp. and 6 pis. 



2 R 2 



