ULWES' AND GILBERT'S EXPERIMENTS. 



123 



" In the selection of the animals," say Messrs. Lawes 

 & Gilbert, " it was only sought to get such as resembled 

 each other in character, age, and weight, in the several 

 pens ; and, with this view, a competent person was em- 

 ployed to go to the various sties and markets in the 

 neighborhood to purchase animals suited to our object. 



" Forty pigs were purchased, as nearly as possible of 

 the same character, and all supposed to be about nine or 

 ten months old. The pigs were weighed and marked, and 

 thirty-six of them selected out, and divided into twelve 

 lots, of three each, in such a manner as to give equal 

 weights in each lot, but it was found that, in selecting 

 them by weight alone, ' they did not secure animals of 

 sufficiently equal feeding quality in the several pens.' On 

 the following day, therefore, they were changed from pen 

 to pen, so as to provide, as much as possible, a similarity 

 in this respect betwe"en pen and pen, and, at the same 

 time, to retain a near equality in weight also. This being 

 done, the weights stood as follows : 



TABLE I. SHOWING THE WEIGHTS OP THE PIGS WHEN ALLOTTED 

 TO THE PENS, FEB. 3, 1856. 



"The allotment thus completed, all the pigs were fed on 

 a mixture of one part bean meal, one part lentil meal, 

 two parts Indian corn-meal, and four parts bran these 

 being the foods fixed upon for the subsequent experiment. 

 The pigs were allowed as much of this food as they 

 would eat." " Upon this mixture," say the experimenters, 

 " all were kept for twelve days, prior to commencing the 



