THE GLIM AT AL CHANGES Ob' MELAN1TTS LED A. 71 i 



of the larval stage under ordinary conditions is about a month. They pupate 

 either on the stalk of the leaf or beneath it. 



They are subject to innumerable enemies. Ants are extremely ford of their 

 eggs, and will also carry off the larva? ; a green and particularly obnoxious 

 spider fastens itself on to the anal orifice and sucks the unfortunate larva dry. 

 The immature forma of the preying Mantis hide themselves among the plants, 

 and very speedily reduce their numbers. It is better, therefore, to keep them 

 on growing plants of the food plant, and protect them by placing them over 

 water and surrounding them with muslin ; to save them from their numerous 

 foes is a task which has taxed my energies to the utmost. 



The pupsa are at first bright translucent green, changing, as the butterfly is 

 being formed, to an opaque whitish green, particularly about the wing covers ; 

 it is oily 24 hours before emergence that the colours of the butterfly begin to 

 show, they rapidly become more visible in clearness of outline and intensity of 

 colouring, the white apical spot b ing very conspicuous six hours before emer- 

 gence. It is remarkable in the hot weather how very little the pur 83 vary in 

 their time of emergence. One may be almost certain that pup® reared under 

 the sam3 conlitions will hatch out together. 



Th3 pupal stage lasts a fortnight, — the life history from the hatching of 

 the egg to the emergence of the butterfly in the hot weather boh g appri xi- 

 mately six weeks. 



The few hours before emergence is the critical time for the butterfly, and 

 owing to my high mortality in future experiments, I intend removing the pupae 

 from artificial to Latural conditions on the first sign of the brown colouring 

 matter being formed. 



My fh-st endeavour was to ascertain the definite relation between the seasonal 

 forms and meteorological conditions, and for this I was happily situated. The 

 observatory is at Curepipe, and all the insects recorded were caught in the Etme 

 lane within 200 yards of the observatory. The meteorological conditions under 

 which they were born and bred are therefore exact. I may statu an objection 

 which may be raised regarding the tables below, and that is that an insect caught 

 early in the month can scarcely be affected by the climate of the end of the 

 month. This no doubt is true, but I have gone on the supposition that the 

 seasonal changes take place in the last few days of pupal life, and as I was 

 careful to take fresh specimens (Melanitis soon tatters itself), the meteorological 

 data for each specimen can easily be worked out, and it will be found also that 

 tha results are the same for each specimen as I have given for the total number 

 for the whole month. 



I have kept female specimens of Melanitis alive in captivity for three 

 weeks, but the males for a much shorter time, most of my captures were of 

 this sex, and I should put the life of a male Melanitis as cot much more than a 

 fortnight, and in really gool condition as inside of a week. 



The following meteorological observations were taken at Curepipe, 1,860 feet 

 above sea level, at 9 a.m. and 3 pm. The insects were taken within 2<j0 yards 

 of the observatory. W=" wet " form. D=" dry " form. I=" intermediates." 



