A NUTRITIONAL STUDY OF INSECTS 39 



on which larval growth depended, but found that the microor- 

 ganism when extracted with hot alcohol could no longer serve as 

 food for the insect. The addition of those special substances nec- 

 essary to higher animals did not take the place of the substance 

 extracted from yeast. The insects could not be reared on the 

 normal salts, sugars, and amino-acids or proteins sufficient for 

 higher anunals, viz., cane-sugar, MgSO^, NaCl, and CaCl2, with 

 casein, edestin, egg albumin, or a mixture of leucine, alanine, gly- 

 cine, asparagine, tyrosine, tryptophane, and histidine, or with 

 milk. Twelve successive generations of the flies were raised 

 aseptically on yeast, water, and citric acid. It should also be 

 mentioned that Loeb and Northrop raised aseptic flies on aseptic 

 unfermented banana, but were unable to secure a second genera- 

 tion from them even after feeding the adults on yeast, as both 

 sexes were sexually sterile. 



My experiments show that Drosophila can be reared normally 

 on yeast nucleoprotein, sugars, and salts, therefore any 'special 

 substance' required by the larvae must be present in this mixture. 



As previously mentioned, I have been able to rear sterile larvae 

 on sterile hot aqueous extract of banana agar and obtain adults 

 which appeared to be sexually sterile, as they did not oviposit on 

 the banana during six days (the usual preoviposition period being 

 twenty-four to forty-eight hours), but when half of the number 

 were transferred to an aseptic 4 per cent yeast-agar medium, the 

 females oviposited in one to three days. The larvae that emerged 

 reached a length of 5 mm. in three days; the females remaining 

 on the banana did not oviposit. Guyenot ('13, b) has explained 

 this as a nutritional phenomenon. He observed that normal 

 females from yeast-fed larvae placed upon a poor food, such as 

 carrot, after a few days deposit eggs which though fertilized no 

 longer develop to maturity, but die as partially developed em- 

 bryos. If the same female recopulates after a period, it at first 

 deposits normal fertile eggs, then abnormal fertihzed eggs, and 

 finally unfertilized eggs. The following experiments of Guye- 

 not's ('13 d) will serve further to illustrate this point. He reared 

 adults from aseptic larvae fed on sterile potato, but found that 

 most of them were ahnost sexually sterile. Oviposition did not 



