not possible in many caa'ss to det.ermiR'2 as: sccui'ately as cue would Like the 

 strength of the poisori. If anyone nii'^ht vilsh to carry on further experi- 

 ments of this sort, the author recommends the use of injections by the 

 ejctraction method described in a later c/iapter„ 



TJie followin,<» facts should be noted with regard to the tables of animal 

 eocperltnents inserted in the various sections of this report: 



(1) I5xperiraental animal. M is the abbreviation for'^iouse". No. is 

 the number of the cage. The cages were di\n.ded into tv?o co/npartments and 

 the animal used in an experiment is desio;nated as "right" or "left". The 

 mice were not given individual serial numbosrs and for this reason when the 

 two mice in one cage v;era used in different experirients the caf,e namber is 

 the same. The cats 'sero not given serial numbers except for temjjorary ones 

 assigned when a large number of animals were being kept at one time. 

 Animals for which no body weights are recorded are these for vshich the data 

 werf lost, and for these the weights run about 10-20 grams for the mice and 

 500 grams — 1 kilogram for the cats. 



(2) The figure on the line below the individual serial number of a 

 fish indicates the total length, 



(3) The ratios of mixtures where fishmeal, starch, and so forth were 

 mixed with the test material is shown as 1:1 or as * '^4.^ - ^ , The true 

 quantity should be determined by subtracting the amount of material added, 



(4) "Flesh", "head", "tail", and so forth indicate muscle tissue 

 taken from those parts. 



(5) "Fresh" (or "raw") or "cooked" indicate respectively unpre- 

 pai^d material and material prepared as described above, 



(6) Amount of evaporation has not been subtracted from the amount 

 eaten. As a standard for determining this connection the time at which the 

 remaining food was weighed lias been recorded. Time was lackinrr to make 

 these calculations for each experiment, but the following examples are given 

 to serve as a reference for evaluating the data of all the experiments. 

 These measurements were made in the laboratory at Salpan, 



-11- 



