136 R. W. GLASER. 



For the above work, I chose two species of cockroaches for 

 the reason that conditions seemed simpler. Cockroaches can be 

 easily obtained and reared. They are large insects when com- 

 pared with aphids, coccids, psyllids, aleurodids, etc., and can be 

 readily dissected. Moreover, Mercier (1906) reported that he 

 had cultivated the so-called " Blochmann bodies," thus putting 

 an end to the tiresome discussions of Cuenot, Prenant and Hen- 

 neguy, who supposed that the rods were metabolic products and 

 not bacteria. Mercier cultivated these organisms on nutrient 

 agar, gelatine, potato, milk and bouillon. He described only the 

 morphological characters and named the organism Bacillus cue- 

 noti. The fact that one species of cockroach had yielded an 

 organism that could be cultivated also led me to choose these 

 insects in preference to others. Lastly, cockroaches are mono- 

 symbiotic; in other words, conditions are not, so far as known, 

 complicated by the presence of more than one microorganism. 



Before proceeding to the cultural and physiological section of 

 this paper it might be well to glance over the morphological re- 

 sults in Blattids as shown by Blochmann, Heymons, Buchner and 

 others. 



i. THE BACTERIOCYTES AND BACTERIA IN BLATTIDS. 



In Blattela gennanica and Periplaneta orientalis certain of the 

 fat cells are filled with bacteria measuring 6 to 8/x, long, straight 

 or somewhat curved. The bacteria divide by binary fission. The 

 cells occupied by the microorganisms do not contain any fat or 

 urates and the nucleus is always normal. 



The cockroaches present the bacteriocyte condition ; there are 

 no mycetomes nor complexes of mycetocytes nor, more strictly 

 speaking in this case, bacteriocytes. 



In Blattela gennanica several rows of bacteriocytes are found 

 in the fat body. The youngest eggs in the oviduct are free from 

 infection. Somewhat older eggs, however, show several organ- 

 isms on their surface. With the growth of the egg the number 

 increases so much that later several layers surround the periph- 

 ery. The direct wandering of the bacteria out of the host cells 

 through the follicular membranes into the eggs, has not been ob- 



