72 WILLIAM H. COLE 



The work was done under the direction of Professor G. H. 

 Parker, to whom I wish to express my sincere thanks for guid- 

 ance and suggestions throughout its progress. 



EXPERIMENTS 



1. Effect of gravity in the dark. — The first experiments were 

 carried out in a dark box modelled after the one described by 

 Carpenter, except that no heat screen was used. 1 The glass 

 cylinder employed was 18 cm. long and 4 cm. in diameter, and 

 was marked off by fine ink lines into six regions of equal length, 

 to facilitate locating the flies at the end of the experiments. A 

 small number of flies were put into the cylinder and attracted 

 to the top end by a strong light. Quickly but carefully the 

 cylinder was placed, this end down, inside the box. After a 

 period of one minute the door was opened, the lights turned 

 on and the position of the flies noted. Observations were also 

 made with a single animal, with smaller and larger cylinders of 

 celluloid as well as of glass, but since the results were always 

 the same it is not necessary to describe these modifications in 

 detail. 



The results of 58 trials involving 26 different animals showed 

 that an average of 82 per cent went to the uppermost third of 

 the cylinder after it was inverted, that 4.8 per cent remained 

 in the lowest third and that the others stopped creeping in the 

 middle third. The individual readings for those at the top 

 varied from 67 to 92 per cent. In other words the animals re- 

 acted negatively to the stimulus of gravity in the dark. Whether 

 or not this response is due entirely to gravity without regard 

 to the mechanical stimulus of turning them over will be con- 

 sidered later. 



One of the sets of records in this series of experiments is given 

 in Table I. 



2. Effect of gravity on flies equally illuminated from above and 

 below. — The dark box was converted into a light box by the intro- 

 duction of two electric lights, one at each end. These were either 

 carbon-filament lamps of 16 candle power or 15 -watt Mazda 

 lamps. As before, the flies were attracted to the top of the 

 cylinder, which was then inverted and placed in the light box. 



1 Carpenter's heat screen, because of the thinness of the water layer, was prob- 

 ably of no great value in preventing the action of the heat on the flies. 



