22 
Indian Forest Records. 
[Vol. VIX. 
produced, and the number of moths obtained was therefore low. The 
emergence dates in the 1915 column of Table 2 are based on some 400 
shoots, and obviously do not give a correct indication of the normal 
emergence period of the fifth generation moths ; the earliest pupae and 
moths were undoubtedly killed off by frost. 
1916. Toon shoots collected during January, February and March 
1916 contained mostly mature and pupating larvae. The emergence 
records of the moths of this brood are given in the 1916 column of Table 2, 
and represent a truer indication of the emergence period than do the data 
obtained in previous years. Plate Y shows graphically the curve of 
emergences of the 1916 generation constructed from the figures in Table 2. 
The earliest recorded emergence of moths of the fifth generation is 
the 18th February and the latest the 12th April, with the optimum 
period of emergence from the beginning of the 2nd week in March up to 
the beginning of the 4th week in March. 
Larvae of the first generation. 
1914. As all attempts at mating the moths of the fifth generation 
failed (vide p. 19), no insectary work was possible on the eggs and larvae 
of the first generation. During the early spring of 1914 as soon as the 
toon had commenced to flower, selected trees were periodically examined 
for signs of larval activity. Shoots and inflorescences were dissected on 
alternate days in the laboratory, but though the results of attack were 
frequently detected no larvae were discovered. 
1915. In the following year operations were repeated but with 
unsatisfactory results. New first stage larvae were found on April 1st, 
1915 in damaged flowers, but all subsequent collections failed to give 
evidence of the development of the brood. 
1916. In 1916 attempts at mating fifth generation moths having 
again failed, recourse was made to field-work to obtain material of the 
first generation larvae. During March and April a very thorough exami¬ 
nation was made on specially selected trees, and practically every inflore¬ 
scence of two trees was removed and scrutinised in the laboratory. The 
results were much more satisfactory and yielded material sufficient to 
determine the development of the larval stages. 
The first batch of larvae was obtained on toon flowers on the 29th 
March, including a few early first stage individuals. Toon flowers were 
collected regularly from the middle of March onwards and records ob¬ 
tained of the progress of the generation. Throughout April, up till the 
6 th May, robusta larvae of all stages w^ere observed feeding in the inflore¬ 
scences ; eggs were also obtained on the flowers. In the latter part of 
April mature fourth stage larvae were most abundant, and after the 
6 th May larvae of all stages, feeding on flowers, became scarce. 
L 167 ] 
