88 HABITS 



numerous imprints of flies' proboscides on a layer of Indian 

 ink, and Fig. 3 one of the imprints more highly magnified, 

 on which the tracings of the pseudo-tracheae at the end of 

 the proboscis can be clearly seen. If the flies had been fed 

 previously on carmine syrup, red patches were frequently 

 observed at the margins of these proboscis marks, either due to 

 the deposition of carmine which had remained on the proboscis 

 or to the regurgitation of carmine stained material to moisten 

 the dried milk. The latter explanation is probably the correct 

 one in most cases, for a single fly will leave many (100 or more) 

 carmine-stained proboscis marks, and moreov^er carmine stains 

 are more common when the layer of milk is rather dry, and 

 requires more fluid to moisten it, than when it is less dry. In 

 one experiment, made two hours after feeding on carmine syrup, 

 half the proboscis marks showed carmine stains, and in another 

 made 22 hours after feeding several of them showed carmine 

 stains. 



" It was also frequently noticed that flies which had the 

 opportunity of feeding on either fluid or partially dried milk 

 often chose the drier portions. Under natural conditions they 

 can often be seen sucking the dried remains near the top of a 

 milk jug. 



" If flies are carefully observed under natural conditions, or in 

 captivity in a cage, it is seen that they are constantly applying 

 their proboscides to the surfaces over which they are walking, 

 apparently attempting to suck up nutritive material. Under 

 suitable conditions the imprints of their proboscides can often be 

 made out." 



(3) Soluble solids. 



" Flies will feed readily on crystals of brown sugar. The 

 mode of feeding can be very accurately watched by placing one 

 or two flies in a small cage with a crystal of brown sugar on the 

 bottom. The cage may be easily so arranged that the lens of a 

 Zeiss binocular microscope can be focused on the sugar. The 

 oral lobes of the proboscis are ver)^ widely opened and closely 

 applied to the sugar. Fluid seems to be first deposited on the 

 sugar and then strong sucking movements take ])lace. When 



