1918.] NATURAL SCIENCES OF PHILADELPHIA. 107 



surface, and especially at the time of transformation, respiration 

 must be carried on by this method, since the larvae have then come 

 to the end of their aquatic existence and no longer need those means 

 of respiration previously employed but rather are ready to begin 

 an aerial existence, hence respiration by means of spiracles. 



Experiments. 



Various experiments made during 1916 and 1917, seem to throw 

 some light upon the possible respiratory mechanisms of Zygopterous 

 larvae. The larvae used were of species of Ischnura and of 

 Enallagma, obtained from small ponds at Primos, Pennsylvania. 

 Many individuals were employed, the experiments were repeated 

 several times and the data for each set of experiments represent 

 average results. 



The normal larva (with three caudal "gills") when observed 

 under a binocular microscope and in a Sj^racuse watch glass, show 

 rhythmic pulsations in the rectum, in the manner described by 

 Calvert (1915) for the larva of Hetaerina americana, i. e., each 

 pulsation consisted of three to four successive contractions of the 

 rectum, followed by a pause. No uniformity, however, in the 

 number of contractions per minute seemed to exist ; this was possibly 

 due to age, size and environment of the larvae under examination. 

 Particles of carmine and of lampblack were also used and these 

 were seen to go in and out of the rectum. 



The current of water taken in and expelled was not so forcible 

 that a possible locomotor function could be ascribed to it. The 

 currents were not strong but rather weak and of only sufficient 

 strength to be taken into and expelled from the rectum so that a 

 change of water could be brought about. 



The three caudal gills were then removed as follows: the larva 

 was placed under a binocular microscope as above and a fine needle, 

 bent at a right angle was used to remove the gills. The needle was 

 placed on the particular gill to be taken off and the larva pulling 

 away under this stimulus, broke off the gill at its attachment to the 

 posterior end of the abdomen. By this method the three gills were 

 successively removed. The results obtained were the same whether 

 one, two or three gills were removed at the same time or at 

 intervals. The rate of rectal contractions was not increased or 

 decreased at the time of operation and several hours afterward the 

 contractions had not changed their regular normal rhythm. 



These larvae, after the removal of the gills, behaved normally 

 and lived in a regular way, and moulted as did larvae with gills.^ 



