380 BOTANICAL GAZETTE [JUNE 
Wishing to determine the effect of natural color and surface 
of buds upon the absorption of heat, I carried through several experi- 
ments with horsechestnut buds, which gave some interesting results 
as follows. 
On February 8, in bright sunshine, a large horsechestnut bud 
was obtained and the scales dissected away, care being taken that 
they were not unnecessarily injured. Two thermometers previously 
tested as to accuracy were obtained, and over the bulb of one the 
bud scales were carefully imbricated and firmly held in place by 
a few turns of black thread. There was enough resin present to 
cement the scales firmly together and thus form an artificial horse- ° 
chestnut bud with the thermometer bulb in place of the normal 
shoot. The instruments were then placed on a table out of doors 
’ and in the shade where they were allowed to lie. As soon as the 
readings were nearly the same,'® the table and instruments were 
carried to a place in full sunshine, care being taken that the two 
bulbs projected about 6.5°™ beyond the edge of the table so as 
not to be affected by direct radiation from the surface of the latter. 
The readings were taken as follows: 
‘TABLE I. 
Horsechestnut bulb and naked bulb, from shade out of doors to sunlight. (See jig. I.) 
Maked bein. | Momoeneh to ee ° Difference 
ge" F. pire 0:00 oO PS (axe €) 
33 32 0:10 I mo: 
34 33 0:30 I ae 
34.5 34 1:10 0.5 0.2 
35 35:7 bee fe) 0.7 0.4 
36 a5 2:00 I 0.5 
36 38 2:30 2 L-6 
or 39 3:00 2 EO 
37 40 3:30 3 to" 
37 42 4:00 5 2.8 
38 44 4:30 6 ae 
38 45 5:10 7 3.9 
38 48 6:00 10 5.5 
38 49 7:00 II 6.1 
38 50 10:00 rz 6.6 
30 5L 14:00 12 6.6 
39 52 19:00 13 7.2 
3) 53 24:00 14 7.7 
39 54 29 :00 Ig 8.3 
‘© They could not be made to read the same because the overcooling point in the 
bud had been reached. 
