632 CAUDAL (ilLLS OF ZYGOPTERID LARV.f:. 



i\ote on the Problem of the Origiii of the Constricted or T^vo- 

 joint.ed GiU: — The known facts for tlie solution of this incom- 

 pleted piobleni are as follows : — 



(1) In the Ontogeny of the Nodate and Subnodate Types 

 {Aiistroayriou and Ischtiura) no sign of a division of the gill 

 into two parts is to be seen up to the seventli instar, 



(2) In the fourth or fifth instar of Neosticta, the gill is a two- 

 jointed or constricted saccus, but the distal joint is very small. 



(3) Half-grown larvae of Calui(/rion hillinghnrsti (NodateType) 

 show the constriction almost as marked as in Isosticta. These 

 larvae are probably of the tenth or eleventh instar. 



►Statements (1 ) and (3) might appeal- to be antagonistic; since, 

 at the most, only four instars could intervene for the change 

 from a simple to a constricted gill. But, as a matter of fact, 

 the node in Caliagrion is much more strongly developed (in the 

 full-grown larva; than it is in Auatroagrion, though both are 

 clas.sed as of Nodate form. Consequent!)", if a series of instars 

 of the rare larva of (kdiayrion could be obtained, say, from the 

 fourth upwards, the problem would very probabl}' be solved. It 

 would also be necessary to breed the larva? of Neosticta or Iso- 

 sticta from the egg, and raise them to the fourth or fifth instars, 

 to complete the chain of evidence. 



Thus we see that the solution of this interesting problem now 

 rests only with him who can obtain two lots of very rare larva?, 

 and keep them alive through a sufficient number of instars. So 

 far, I have not succeeded in doini; this. 



[Printed oft; December 19th. 191 



