338 oscar riddle. 



The Experimental Production of Fault-Bars. 1 



Professor Whitman's suggestion that fault-bars are due to the 

 malnutrition of the feathers during their growth, was put to a 

 direct and vigorous experimental test. The method employed 

 varied with the nature of the experiment and with the material, 

 which was of four kinds. Of course, one had to work with 

 growing feathers. Those experimented upon were (1) the 

 juvenal feathers of ring doves {T. risorius) ; (2) the later plum- 

 ages of ring doves produced at the regular season of moult ; (3) 

 feathers, chiefly remiges and rectrices, of ring doves obtained at 

 any season by previously removing some of these ; (4) juvenal 

 and adult plumages of chicks {G alius). All of these yielded 

 entirely comparable results ; the young birds merely showing a 

 greater sensitiveness to lack of food. Five types of experiment 

 were tried : (1) Reduced feeding or starving ; (2) feeding with 

 Sudan III.; (3) mechanical crumpling of the germs; (4) effect of 

 light, temperature, bad sanitary condition of the nest, parasites, 

 etc.; (5) amyl nitrite. 



Reduced Feeding. — To reduce the feeding of the young doves 

 one had only to limit the feeding of the parents, the latter refusing 

 to regurgitate the food for the young when insufficiently fed. 

 By this means one could not be certain of the amount eaten by 

 the young unless the parents were not fed during a couple of 

 days. In many of these cases the young died. In those cases 

 where a partial starvation of the young was evident, one invari- 

 ably found later, one or more fault-bars to correspond to it. In 

 some cases the old birds were fed normally and one of their 

 young was left with them as a control while the other bird was 

 placed either in an incubator or under other nesting birds. The 

 experimented bird could then be replaced with its parents and 

 fed by them as little or as much as the experiment demanded. 

 Twenty-four hours without food was invariably accompanied by 

 the production of pronounced fault-bars (of type 1) in these 

 birds. The control birds usually did not show these obvious 

 defects (PI. XIII., Figs. 6 and 7). 



1 Incomplete results of these experiments were communicated (1905) to Professor 

 Duerden, who in his paper already cited (1906), wrongly credits the work to Drs. 

 Strong and Whitman. 



