96 INSECT TRANSFORMATIONS. 



earth, or or on the branches of trees ? Yet, when the 

 spring had tempered the air, these eggs produced as 

 they usually did after the mildest winters. Since that 

 period there have been winters more severe. In 

 France, during December, 1788, the thermometer 

 fell considerably lower, and in several other tempe- 

 rate European climates. 



' I have exposed eggs to a more rigorous trial 

 than the winter of 1709. Those of several insects, 

 and among others the silk- worm, moth, and elm 

 butterfly ( Vanessa polychloros ?) were enclosed in a 

 glass vessel and buried five hours in a mixture of ice 

 and sal gem (rock salt), the thermometer fell 6 

 below zero. In the middle of the following spring, 

 however, caterpillars came from all the eggs, and 

 at the same time as from those that had suffered no 

 cold. In the following year, I submitted them to an 

 experiment still more hazardous. A mixture of ice 

 and sal gem with the fuming spirit of nitre (Nitrate 

 of Ammonia), reduced the thermometer 22 below 

 zero, that is 23 lower than the cold of 1709. 

 They were not injured, as I had evident proof by 

 their being hatched. 



' Combining all these facts, we conclude that cold 

 is less noxious to germs and eggs, than to animalcula 

 and insects. Germs in general can support 2 be- 

 low zero; whereas of animalcula some die at the 

 freezing point, and some at about 20. The eggs 

 of many insects continue fertile after being subjected 

 to a temperature of 22 below zero, while insects 

 themselves die at 16 and 14. This I have ascer- 

 tained in the eggs of the silk-\vorm moth and of the 

 elm butterfly; and although there are caterpillars 

 and chrysalides able to resist great cold, I have uni- 

 formly found it to be in a less degree than what can 

 be resisted by their eggs. What can be the cause of 

 so great a difference? Insects killed at 16 and 14 



