F. B. Browne 367 
the things which make all the difference in that relationship. Another example 
of this is to be seen in the life-history of the fossoriai wasp Sapyga quinque- 
punctata which I have recently investigated. 
In this case the usual host of this parasite in my garden in Cambridge 
was the Blue Osmia ( 0 . aenea (caerulescens )) while Osmia rufa was immune 
and it appeared that the latter species owed its immunity to the fact that 
the pollen paste stored in the cells by the mother bee was too dry for the 
larvae of the parasite. Eggs of Sapyga were occasionally laid in the cells of 
this species but the larvae always perished and, in laboratory experiments, 
where the Sapyga eggs were placed in the cells of 0. rufa , the result was always 
the same. If, however, the pollen paste was made suitably moist by the 
addition of honey or even treacle—the latter, as experiment showed, being 
quite useless as a food for the Sapyga larva—the parasite could be brought 
to maturity without any difficulty. 
XII. Winter rearing of Melittobia. 
I have already said that by keeping the species during the winter in an 
incubator at a summer temperature 20°-22°C. (about 70° F.) it was possible 
to raise successive generations. 
I first tried this because I was held up in the autumn in the middle of 
experiments which I was very anxious to continue and I therefore placed 
batches of hibernating larvae in glass cells in an incubator early in December. 
Different batches of larvae behaved differently. Some responded within 
a few days to the warmth and began to void excrement preparatory to 
pupation; others remained apparently quite unaffected by the warmth for 
much longer periods. Moreover a whole batch of larvae did not necessarily 
behave in the same way and a few extreme examples will make this point clear. 
Batch 1. Batch 2. 
Placed in incubator 10. xii. 21 In incubator 10. xii. 21 
Some casting excrement 26. xii. 21 One pupa 24. xii. 21 
A few pupae 31. xii. 21 All pupae 31. xii. 21 
A few imagines 5. ii. 22 
Mostly pupae, a few larvae 15. i. 22 
Pupation period 31 days and upwards Pupation period 50 days 
and upwards 
Batch 3. 
In incubator 10. xii. 21 
One pupa and the) 
male appeared j 
About half have ) 
pupated f 
Many still larvae 
6. i. 21 
5. ii. 22 
19. ii. 22 
During the summer, the normal pupal period is only about seven days 
and females hatched in the incubator at summer temperatures gave rise to 
generations which had pupal periods of about the normal summer duration 
and yet in the case of those hibernating larvae which pupated in the incubator, 
the pupal period varied from 18 to 57 days in duration. Now a long pupal 
period is not an innovation to the species, since over-wintering pupae do occur 
under natural conditions and therefore the explanation of the long pupal 
period in the case of these incubated hibernating larvae seems to be that the 
effects of the incubator are gradual and that it takes some time before the 
insect reacts to the increased temperature, or it may be that the reaction of 
24—2 
