STATE AGBICULTURAL SOCIETY. 325 



formerly disagreeable by reason of fog in Summer and continuous rain in 

 Winter, lias so moderated in the last few years that fog is comparatively 

 infrequent, and the rainfall has diminished, how much it is difficult to say, 

 as observations conflict. 



It is proper, in this place, to notice the difficulties which arise in the 

 study of rainfall on this coast, and the many sources of error to be avoided. 

 The United States Signal Service has established stations along the coast, 

 where the temperature, velocity, and direction of the wind, and the rain- 

 fall are recorded. As a rule, the men chosen to keep this record are those 

 who have shown a desire, as well as a natural adaptability, for such work, 

 and their reports are reliable. 



On the other hand, the reports from certain military posts are to be looked 

 on with suspicion. The observers usually selected are the Post Surgeons, 

 and, too often (I speak from personal knowledge) the duty of recording 

 observations is delegated to the hospital steward, who, from the recesses 

 of his inner consciousness, draws up a report that reads well, but which 

 has not the slightest foundation in fact. 



Another most important source of error, and one difficult to correct, is 

 that of accurately measuring the rainfall. For this purpose the Signal 

 Service Bureau has recommended to each observer a receiving can, which 

 is to be placed in the most open spot available. When the water falls 

 perpendicularly this can will be an accurate register. As a rule, the coast 

 rains are accompanied by winds of high velocity. This wind will vary 

 the amount of rain caught in the receiver; for it is evident that more rain 

 would fall in a receiver, the plane of which is perpendicular to the plane 

 followed by the drops, than in one where these planes are at an angle less 

 than ninety degrees, and the decrease would vary directly as this angle. 



If the surface of the earth were absolutely level this might represent the 

 actual rainfall, but on every hand incline planes arise, such as sloping hill- 

 sides, and other elevated obstructions, forming inclined planes. These 

 planes, making greater angles with the lines of incidence, receive a double 

 portion, and the result is a raise of the average rainfall. This error can 

 be overcome by taking the velocity of the wind in any given storm and 

 calculating its disturbing effect. 



Again, the wind, meeting with obstructions, throws the drops into vor- 

 texes and currents, materially interfering with their fall. Should the 

 receiver happen to be in this current, the result recorded would be far too 

 great. Although the Signal Service Bureau has recommended the adop- 

 tion of a standard gauge, placed at a certain height above the ground, 

 their suggestions have not, in all cases, been adopted. 



For these reasons the reports of the Signal Service Office should be taken 

 with many grains of salt. Even were the gauges absolutely accurate, only 

 approximal deductions could be drawn from the reports sent in. The only 

 general law that has so far been established is that the rainfall of the coast 

 increases toward the north. The stations are too far apart. 



Before we can give a true report of the rainfall of any given section, rain 

 gauges should be placed on at least every square mile of the territory 

 reported. A failure to avail itself of all sources of information obtainable, 

 and an implicit reliance on figures reported, without taking into considera- 

 tion modifying and extenuating circumstances, has, in certain instances, 

 led to serious mistakes. 



A most notable example of this is the one hundred and five inches 

 given as the rainfall of Crescent City, and presumably, of the surrounding 

 country. 



This leads all other localities by four inches, and all other coast stations 



