1911] Pemberton—The Sound-Making of Diptcra and Hymenoptera 115 



certain gnats, and a few other flies; also with a bumble bee, honey 

 bee and certain wasps and hornets. With all he came to the same 

 conclusion, namely, that besides the humming sound produced 

 by the vibration of the wings in the air, another finer, higher- 

 pitched sound was produced, not by the wings but by an appara- 

 tus connected with the spiracles and tracheae, usually the thoracic 

 spiracles only. He found that by stopping the openings of the 

 spiracles humming ceases, or becomes so feeble as to be hardly 

 perceptible. 



In one family of flies, however, (Stratiomyidae), in which the 

 thoracic stigmata are not strongly developed, no humming sounds 

 could be detected other than those produced by the wings. The 

 stigmata were similar in structure to all others examined, but not 

 so strongly developed. 



It is interesting to note that Landois seems to find in those 

 insects which hum loudest with their spiracles, the strongest 

 developed spiracles to be found in all. These are, according to 

 Landois, excellently developed for sound production. The external 

 opening of the spiracle is covered with hairs and leads to an 

 enlargement at the commencement of the trachea and immediately 

 behind the spiracle. This enlargement serves as a sounding box. 

 A folded membrane, called the vocal membrane and forming lips 

 or curtains more or less plaited and fringed, is placed between the 

 edges of the spiracle and the sounding box and, when vibrated by 

 violent rushing of the air, produces sound. This theory has been 

 carefully worked out by Landois and has been generally accepted 

 as fact. 



Having performed a number of experiments with certain Diptera 

 and Hymenoptera, and having obtained results in exact contra- 

 diction to those of Landois, I venture to state these results, offering 

 them to entomologists for what they are worth. 



Despite the weight of testimony which seems to favor the 

 theory of the spiracular voice, I cannot avoid the conclusion, from 

 these observations of my own, that there has been some curious 

 mistake about it all. My experiments with several species of 

 Syrphidse {Eristalis tenax in particular) the house fly, honey biee 

 and the bumble bee, show that these insects do not produce audible 

 sounds by vibration of any portion of their spiracles or tracheae, 

 but that all sounds of the nature of humming or buzzing produced 



