104 PROCEEDINGS OF THE ACADEMY OF [March, 



as having a fin-like or rudder-Uke function, mentioning no 

 respiratory function whatever. Dufour (1852) and Roster (1886) 

 ascriVjed to them the functions of respiration and locomotion. 

 Sharp (1895), remarking that the nymphs lived after removal of 

 the caudal gills, thought their respiratory function must be of an 

 accessory nature. Heymons (1904) held that the entire body surface 

 serves for respiration in addition to the gill-appendages. 



A respiratory function was first attributed to the rectum in this 

 group bj^ Roesel von Rosenhof (1749). Dufour (1852) described 

 the gills in the rectum and from this time on a respiratory function 

 has often been assigned to them. Calvert (1915) has given a detailed 

 account of the manner in which water is taken into and expelled 

 from the rectum, and from these observations, we must undoubtedly 

 ascribe to the rectum a relation to the respiratory functions of the 

 nymph. Tillyard (1916) asserts that in early larval life, immediately 

 after hatching, for a period of half an hour or so, the rectum 

 contracts vigorously and to these contractions assigns a respirator}'- 

 function, especially since he thinks the caudal gills of the very 

 young larvae are incapable of carrying on respiration. This con- 

 clusion is undoubtedly true, but in later periods of the larval life, 

 he attributes to these intermittent contractions only a possible 

 respiratory function, governed ''by the efficacy of the total of other 

 respiratory activities. Respiration through caudal gills, through 

 lateral abdominal gills when they occur and through the integument 

 in such cases as it may occur, all differ from rectal respiration in 

 being practically continuous, carried on without special efforts on 

 the part of the larva. Consequently, if the caudal gills have a poor 

 tracheal supply, or a very tough integument, or if the general 

 integument of the larva be unsuitable for respiration by diffusion, 

 the call on the larva for rectal pulsations must be heavier." 



If these statements be true, larvae which are old and hence have 

 thicker chitin over body and gills ought to show rectal pulsations 

 to a greater degree than the normal larvae. Such a condition, 

 however, has not been found to be the case in the larvae examined, 

 but such larvae have contractions quite like those younger and 

 with softer chitin. Tillyard continues: "If the caudal gills are 

 accidentally lost, we should expect that the rate of rectal respiration 

 would be increased." When the gills are removed one at a time, or 

 all together, no change in rate of rectal respiration has ever been 

 noticed, hence we can be quite sure of the non-respiratory function 

 of these caudal gills. 



