A. W. Rymer Roberts 129 



no food other than humus and decaying vegetable matter. Graf (13) 

 records an experiment made by H. M. Russell on Limonius californicus 

 to test the capacity of the larvae for subsisting, as he supposed, without 

 food. " Ordinary soil " alone was given, and, for the first thirteen months 

 of the experiment, it was regularly watered. At the end of this time 

 seven larvae were alive and healthy. The soil was then allowed to 

 become dry for some two months and during this time all the larvae 

 died except one, their bodies being removed as they died from desiccation, 

 to prevent them from being used as a source of food by the survivors. 

 Watering was then resumed and the one survivor continued to live until 

 it died through an accident twenty months after the commencement of 

 the experiment. This treatment was not of course in any sense normal 

 to the larvae and the species dealt with of a different genus from any 

 of our native species known to be injurious. Nevertheless the experiment 

 throws light on the capacity of "wireworms" to withstand starvation 

 by fallowing, per se, or by the use of mustard as a " starvation crop " on 

 infested land. I have made no experiment lasting over so long a period, 

 but from observations made in the laboratory and in pot cultures, there 

 can be little doubt that Agriotes larvae would succeed in maintaining 

 life in similar conditions for a very considerable time, if not actually 

 for so long as Limonius californicus. Larvae have been kept for some 

 months on a diet of vegetable compost during the spring and summer 

 and in the laboratory have frequently been kept in boxes for prolonged 

 periods with no food other than decaying roots and the humus contained 

 in the soil. They do not thrive under such a treatment as the last named, 

 but life is maintained and it is certain that nourishment is derived from 

 the organic matter contained in the soil. On the other hand, provided 

 that sufficient moisture is supplied, Agriotes larvae (and those of other 

 Elaterids) are able to fast for a considerable time without apparent 

 inconvenience. In order to test this, three larvae of different sizes were 

 isolated in wide glass tubes containing only moistened sand which had 

 been previously sifted and treated with acid. One larva was able to 

 reach the cork of its tube and was found gnawing it about a w^eek after 

 the commencement of the experiment. At the end of one month all the 

 larvae were found to be alive and active, having apparently suffered no 

 ill effects from the treatment. It should be added that the experiment 

 took place in February and March, so that there was probably no ex- 

 cessive reserve of fat in the bodies of the larvae such as might have 

 been the case at the end of summer. 



In the course of experiments to test the toxicity of certain substances 



