lyniAX DliAGONFLlE^. 4G5 



appear as long, ropy masses similar but on a smaller scale to thoso of 

 some of the Batrachians. It is not known how long after deposition, 

 they take to develop, bnt it appears that th^re is no delinito incubation 

 ]>eriod, as the eggs of any one partietdar batch are ftmnd to hatch ont over 

 a period or succession of days, the eggs on the periphery of the mass being 

 the lirst to hatch out and so on towards the centre of the batch. Micros- 

 oojncal examniation of the eggs shows that they are in diflerent stages of 

 development. 



It has already been stated that the majority of the Anisopteridif 

 deposit their eggs freely in water, the Aeschnid:e and Vetalurinie appear- 

 ing to be the only exceptions to this rule. It has now been satisfactorily 

 proved that the majority of dragonflies, including the latter two families and 

 th'.> whole of the Zygopteridte, deposit their eggs actually in the tissues of 

 plants or in decaying pieces of wood in or a little above or in the vicinity of 

 water. As would be suspected, the former class possess a very incomplete 

 ovipositor and their eggs are soft and globular. The latter class on the 

 other hand, have a very highly organised ovipositor and their eggs are of a 

 shape and consistency fitted for their passage into the nest. They have a 

 much firmer shell, are oval in shape and pointed more or less at the 

 extremities. 



The ovipositor is a complicated organ consisting of a pair of vulvar scales 

 which functionate partly as a covering for the boring organs and partly as 

 tactile organs for grasping the eggs and guiding them into the holes bored 

 by the other processes. These latter are two paired organs lying between 

 the vulvar scales, one of which is a stillette-shaped process for making the 

 punctures in the stems of reeds and water-plants and the other a saw-like 

 organ for enlarging the punctures to a size sxutable for admitting and 

 accommodating the eggs. The vulvar scales are further fitted with small 

 tactile organs or styles which are furnished at their summits with a tuft <ir 

 pencil of stift" bristles. With these latter, the female insect palpates the 

 stem of the plant for a suitable spot and thereafter makes a series of 

 punctures in a more or less irregular line. Into each hole, one or more 

 eggs are guided by the vulvar scales and stillettes. 



In the Moolah IJiver at Poona, specimens of Mieromerus lineatus and 

 Art/ia f/uadri»;acu'ata were observed depositing their eggs in reeds which 

 had bent and fallen into the water and were lying flush with the surface, 

 swept by the current but firmly anchored by their stems. Sections of these 

 reeds showed manj"^ hundreds of j)unctures on their surfaces, easily observ- 

 able with the naked eye as minute black points arranged in very irregular 

 and broken rows. The larvic from these eggs hatched out on the fifth day, 

 but the incubation period is not always as short and in a colder climate 

 would possibly be of several weeks duration. 



The incubation period depends largely on the climate and the tempera- 

 ture and in this country, on the occurrence of the wet and dry seasons. 

 The majority of dragonflies leave the larval state some time before the onset 

 of tlie rains, this being a natural provision to preserve the species, as most 

 ])')ols are at their lowest ebb and many tanks are completely drying up. 

 The ovse are deposited in the deeper pools at the onset of the rainy season 

 so that the larvye are given the fullest time in which to develop before the 

 water supply fails. Although this is stated as a general rule, it is by no 

 means applicable to all the species. Tn the hills, the imagines are seen to 

 ascend the mountain ravines at the onset of the rains, followinc: the course 

 of the streams and the opposite phenomenon occurs at the end of the wet 

 season. Many of the ovse deposited in these swift running streams must 

 be carried immense distances before they develop into larvie. 



