248 M. A. de Candolle on Germination 
5°-4. to 6°. The mean temperature every two days was 5°68, 
say 5°°7. 
All the species were sown on the 9th of March, upon a basis 
of sand, in a large box. They were wetted with water of the 
surrounding temperature. 
At higher temperatures evaporation would lower the mean of 
the soil in which the seeds were sown, which circumstance was 
taken into account, in the following experiments, by measuring 
the temperature of the soil instead of the air. At 5° or 6° this 
cause could be of but little importance, but it would give rise to 
the presumption that the mean was a little below 5%7. 
The following were the results to the 11th of April :— 
Collomia. Some seeds germinated on the fourteenth day ; 
the others failed. 
Lepidium. Germinated abundantly on the fifth day. 
Linum. Germinated abundantly on the sixth day. 
Maize. Did not germinate. 
Nigella. Germinated on the twenty-seventh day. 
Sesamum. Did not germinate. 
Sinapis. Germinated abundantly on the fourth day. 
Iberis. Germinated on the fourteenth day. 
Trifolium. Germinated on the tenth day. 
Melon. Did not germinate. 
From the 11th of April to the 9th of May the temperature of 
the cellar gradually rose to 8°. Still the seeds of maize, Sesa- 
mum, and melon did not germinate. Those of the Sesamum had 
perhaps suffered from the damp; but those of the maize and 
melon were hardly swollen, and some of them were mouldy. 
At a temperature of about 9°. 
In the middle of May the temperature of the cellar had risen 
to nearly 9°. I took advantage of this to sow, at 1 o’clock, on 
the 17th, all the species, in a broad box, exposed to the free air. 
On the 18th, at half-past 2, I watered them, and allowed the 
experiment to contiue. From the 18th of May to the 2nd of 
June the thermometer in the open air only varied 0°°6. In the 
sand containing the sowings the variation was 0°8. The hu- 
midity caused by the watering always lowered the temperature 
of the soil relatively to that of the air, which induced me to 
determine, as exactly as possible, the temperature of the upper 
layer of soil. On making every correction with the standard 
thermometer, I found 9°2 to be the most probable temperature 
to which the seeds had been subjected. The following are the 
results :— 
