BUREAU OF CHEMISTRY. 275 



very considerably in gossypol content. In consecjuence, experiments 

 upon rats were carried out in which different varieties, notably Egyp- 

 tian. Lone Star, Price, and Durango cottonseed meal, were fed. It 

 was found that the toxicity of different varieties of seed runs roughly 

 parallel with their gossj^pol contents. It was also found that heating 

 and extraction with certain organic solvents, such as ether, removes 

 the gossypol. It follows, therefore, that in these observations is to be 

 found the explanation of the discordant results that various investi- 

 gators have ol)tained in feeding cottonseed meal, either to farm stock 

 or to laboratory animals. It also follows that it should be feasible to 

 so treat cottonseed meal commercially as to remove or destroy prac- 

 tically all the gossypol content of the meal. Recently plants have been 

 constructed for the extraction of cottonseed cake Avith organic sol- 

 vents so as to remove the 4 to 6 per cent of cottonseed oil that remains 

 in the press cake. It is likely that this treatment if carried out thor- 

 oughly will either completely or very largely remove the gossypol 

 from the cottonseed cake so treated. If this is found to be the case 

 such cake should be more suitable for feeding than ordinary' com- 

 mercial cake and should be relatively non-toxic. It may be possible 

 to feed it in very large quantities. 



The work to determine the effect of feeding small quantities of 

 toxic substances over long periods of time by the methods described 

 last year was continued and extended to include several heavy metals, 

 such as lead and zinc, us well as a number of other substances occur- 

 ring as adulterants in foods. During the war, in cooperation with the 

 Bureau of Biological Survey, an investigation of the effectiveness of 

 various toxic agents in the extermination of rats was undertaken. 

 Strychnine was not found very satisfactory. Barium carbonate was 

 found on the whole quite satisfactory. It was determined that when 

 it Avas present in the proportion of 20 per cent of the total quantity 

 of the food submitted to the rats practically 100 per cent of the 

 animals died. A manuscript embodying this part of the work has 

 been prepared for publication. Observations were made upon the 

 relative toxicity of different forms of arsenious oxid. It was found 

 that in the finely powdered or amor]:)hous form it was much more 

 toxic than in the form of well-defined crystals. It v.'as about twice 

 as efficient as barium carbonate. Various observations were made 

 which should be of very great value both in the use of barium car- 

 bonate and in the use of arsenic as rat exterminators. The work upon 

 arsenic has practically l)een com])leted. Papers have been prepared 

 for publication under the following titles: The Toxicity and Physi- 

 ological Action of Arsenic and Zinc; Observations upon the Intra- 

 peritoneal Injection of Fixed Oil ; Toxicity of Strychnine for Some 

 of the genus Mus; A Comparison of the Effect of Certain Saponins 

 on the Surface Tension of Water with their Hemolytic Power. 



INSECTICIDES AND FUNGICIDES. 



In Department Bulletin 795, The Adulteration of Insect Powder 

 with Powdered Daisy Flowers {Chri/sanfhemum Leucanthemum L.). 

 are described tbo nictliods foi- tlio detection of this form of adultera- 

 tion: in Department Bulletin S24, Insect Powder, are described the 

 methods for the detection of the addition of powdered insect flower 



