672 EXPERIMENT STATION RECORD. 



The conclusions reacbed are as follows: 



"While the several experiments here detailed agree in showing that torches placed 

 and burned in the manner described did exert only a feeble influence on the temper- 

 ature of thermometers withiu the protected areas, or none at all, yet the writer feels 

 that these experiments do not warrant the couclusiou that frost torches may not, at 

 times, be used to advautage. They do demonstrate conclusively, however, that 

 there are conditions under which they are of no avail. 



"It must be remembered, however, that all of these trials except the first, when 

 there was no frost, occurred at times either later in the season or earlier than they 

 are likely ever to be needed. Now it may be shown that during these seasons the 

 temperature of the air does not materially increase upward, and if this were true at 

 the times of the experiments here recorded it is evident that the sucking of the 

 layers of air up from the ground by the torches only helps to bring other air of the 

 same or possibly lower temperature down upon it. In such a case only the smoke 

 and cloud formed above the field could be effective in lessening the loss of heat by 

 radiation. 



"If, however, it is true that earlier in the fall and later in the spring, when 

 destructive I'rosts do occur, the temperature of the air at night increases rapidly 

 upward, then the influence of the torches might be much greater. But many trials 

 similar to those here described need to be made at such times before the real value 

 of torches and smudges in preventing damage by frost can be demonstrated." 



An experimental rainfall, L. Errera {del et Terre, 17 [1896), 

 Aug., X). 353; ahs. in U, S. Dept. Agr., Weather BureaUj Monthly 

 Weather Review, 24 {1896), No. 10, pp. 373, 574).— The method pro- 

 posed is as follows: A cylinder of Bohemian glass about 20 cm. in 

 height and 12 cm. in diameter is tilled half full with strong alcohol (92 

 per cent), covered with a porcelain saucer, and warmed nearly to the 

 boiling j)oint over a water bath until the whole apparatus is in thermal 

 equilibrium. It is then removed from the bath and, without agitating 

 the liquid, placed upon a wooden table. After some minutes the porce- 

 lain cover cools and the vapors rising in the cylinder commence to con- 

 dense in the form of distinct clouds. These condense into very fine 

 drops, which fall steadily, vertically, and in countless numbers into the 

 liquid. The drops were found to have an average diameter of from 

 0.04 to 0.05 mm., sometimes larger, but more frequently smaller. This 

 interesting spectacle may last for half an hour. At first the vapors 

 rise quite up to the porcelain cover, but as the apparatus cools the 

 level of condensation naturally falls until a clear space is noted above 

 the zone of clouds. 



"The experiment has been repeated at the Weather Bureau with 

 success, but it should be carried out on a large scale, with a very tall 

 jar and great care as to uniformity of temperature, if one desires to 

 get satisfactory results." 



Instructions for taking phenological observations, L. H. Bailey 

 {V. S. Dept. Agr., Weather Bureau, Monthly Weather Review, 24 {1896), 

 JVo. 9, pp. 328-331). — The author states that phenological observa- 

 tions are of two general types, those which simply record the external 

 features of the passing life in x^lants and animals, and those which 

 attempt to discover or construct some vital connection between life 



