38 MASS. EXPERIMENT STATION BULLETIN 150. 



section where a rise would, as a rule, be expected . This fall of the barometer 

 in the northeast often causes the wind to keep up all night when other 

 conditions would lead a forecaster to expect almost a dead calm. There 

 seems to be no way of figuring on this action of the barometric waves, 

 except by more extensive observations of conditions in the eastern 

 Provinces of Canada than are at present carried out by the Weather 

 Bureau. The officials of the Weather Bureau are planning to take special 

 afternoon barometric observations in that section, in order to forecast 

 our frost conditions more accurately. 



Another puzzUng factor is the occasional occurrence of cloudiness on 

 mornings when, from the general weather conditions, no cloudiness 

 would naturally be expected. It seems quite possible that we may not be 

 able to fully understand the causes of such cloudiness without a study 

 of the conditions of the upper atmosphere. This is of course a very im- 

 portant matter, for the presence of clouds always makes a difference 

 of several degrees in the temperature of a cold night, 



The readings of the maximum and minimum shelter and bog thermome- 

 ters, and the amounts of precipitation, were telegraphed to the office of the 

 United States Weather Bureau at Boston, every morning after May 4, 

 during the spring and fall periods of frost danger. Triple register (for 

 sunshine, wind direction and wind velocity), thermograph, and barograph 

 records were taken in the usual way throughout the season, from early 

 May until the last of October. As the hydrograph did not work satis- 

 factorily no records were made with that instrument. 



There seems to have been this year a general increase in confidence, 

 on the part of the growers, in the forecasts sent out from the Boston 

 office of the Weather Bureau. This increased confidence is probably fully 

 justified, for the forecasts seem to have been much more accurate dur- 

 ing the past season than formerly. 



2. FROST PROTECTION. 



Careful considei'ation has been given to the different possible methods 

 of frost protection where water is not available for use in the usual way. 



It was suggested in last year's report that it might be possible to use 

 the Skinner system, or some other sprinkling arrangement, with an engine 

 and pump only large enough to pump water for one section at a time, the 

 idea being to draw the frost out of the vines by one or two applications 

 of cold water in the morning, before they were thawed out by the heat of 

 the sun. The practicability of this method was tried out on a small 

 scale last May with a spraying outfit, and the results seemed to be far 

 from satisfactory, for the sprayed areas afterward appeared to show 

 distinctly more injury than did the surrounding unsprayed portions of 

 the bog. As the whole matter now stands, therefore, it does not seem 

 at all probable that the Skumer system can very well ever be made practical 

 use of for frost protection. Its undesirability from the standpoints of ex- 

 pense and nozzle clogging were discussed in last year's report. All other 



