The Protochordata 465 



in all the warm seas. In Spengel's recent paper, "Die Benen- 

 nung der Enteropneusten-Gattungen," three families are recog- 

 nized with nine genera and numerous species. At least seven 

 species are now known from the Pacific Coast of North America. 

 Family Harrimaniidae. In Harrimania maculosa, lately de- 

 scribed by Dr. Ritter from Alaska, the 

 eggs are large, with much food yolk, 

 and the process of development is 

 probably, without Tornaria stage. A 

 second species of Harrimania (H. kup- 

 feri) is now recognized from Norway 

 and Greenland. This genus is the sim- 

 plest in structure among all the Enter- . 



FIG. 277. Hammama macu- 



opneustans and may be regarded as the losa (Ritter), the lowest of 



1nwpc;t of Vnnwn PVinrrlatpc tVip mnct chordate animals. An En- 



raates, t OSt teropneustan from Alaska. 



worm-like of back-boned animals. (After Ritter.) 



In Dolichoglossus kowalevskii the species studied by Bateson 

 on the Virginia coast, the same simplicity of development 

 occurs. This genus, with a third, Stereobalanus (canadensis) , 

 constitutes in Spengel's system the family of Harrimaniidce. 



Balanoglossidae. The family Glandicepitida contains the genera 

 Glandiceps, Spengelia, and Schizocardium. In the Balanoglossida 

 (Ptychoderidcs of Spengel) the eggs are very small and numerous, 

 with little food yolk. The species in this family pass through 

 the Tornaria stage above described, a condition strikingly like 

 that of the larval starfish. This fact has given rise to the 

 suggestion that the Enteropneusta have a real affinity with the 

 Echinoderms. 



The Balanoglossid(E include the genera Glossobalanus, Bala- 

 noglossus, and Ptychodera, the latter the oldest known member 

 of the group, its type, Ptychodera ftava, having been described 

 by Eschscholtz from the Pacific Coast in 1825, while Balano- 

 glossus clavigerus was found by Delia Chiaje in 1829. 



Low Organization of Harrimaniidae. Apparently the Harri- 

 mamida, with simpler structure, more extensive notochord, and 

 direct development, should be placed at the bottom as the most 

 primitive of the Enteropneustan series. Dr. Willey, however, 

 regards its characters as due to degeneration, and considers the 



