1912.] Sterilisation of Soil for Glasshouse Work. 819 



dera radicicola, and of the free-living Rhabditis terricola and 

 brevispina, Diplogaster sp., together with some Cephalobus, 

 Plectus granulosus, &c. All these forms are killed by heating 

 the soil to 140 0 F., and when we heated the soil to 200 0 F. 

 and grew tomatoes, we obtained the other root shown in 

 Fig. 5c 



We have never observed reinfection of Heterodera in the 

 steamed soil, although we took no special precautions to 

 exclude it; we did, however, find that the free-living 

 Rhabditis and Diplogaster got in again unless the soil was 

 kept carefully covered up. The point is of considerable 

 importance to the horticultural investigator, because it shows 

 that the subsequent presence of these easily detected free- 

 living forms is no necessary proof that a certain treatment 

 has failed to kill eelworms, nor does it imply that the really 

 dangerous Heterodera is present. 



Toluol and carbon disulphide readily kill all free-living 

 forms, but not all the Heterodera. These, however, have 

 more opportunities of escape from a poisonous vapour than 

 the others, since some of their eggs are enclosed in dead 

 females, the body walls of which make a good protective 

 membrane. 



Other Animals. — The untreated "sick" soils received from 

 the tomato and cucumber houses also contained numbers of 

 other animals, including white worms (Enchytraeidae), wire- 

 worms, springtails, millipedes, centipedes, mites, woodlice, 

 &c, but all were killed at 200 0 F., together with their eggs, 

 so that none of them subsequently reappeared from the soil. 

 There was, of course, some reinfection from outside, spring- 

 tails, &c, getting in from neighbouring pots. Lower 

 temperatures (130 0 F.) were in our experiments also effective, 

 but we do not recommend them for practical work. Carbon 

 disulphide killed all woodlice and other animals, but toluol 

 was less fatal. 



Fungoid Pests. — As we had no outbreak of any fungus 

 disease, in spite of our use of "sick" soils, we were not able 

 in the present series of experiments to study the effect of 

 partial sterilisation on the fungus spores. Mr. W. Dyke has 

 been studying the wilt fungus (Neocosmopara Vasinifecta) 

 that attacks the roots of cucumbers and melons, doing con- 



