FISHERIES, GAME AND FORESTS. 503 



flow I can not but think he is entirely wrong, possibly due to the fact that the New 

 Jersey data on which he chiefly bases his conclusions are too restricted in their scope 

 to give any certain conclusions on the point in question whatever. By way of 

 illustrating this statement I may merely refer briefly to two points. 



(i) Mr. Vermeule announces in the paper in question, and also in several of his 

 reports appearing in the annual reports of the Geological Survey of New Jersey, 

 what he calls the discovery, that there is a definite, certain relation between tempera- 

 ture and the total loss of water from a drainage area, which he has properly 

 designated under the general term evaporation, the word evaporation, in this case, 

 meaning absorption by plants, evaporation proper, and various other losses. 



{2) Mr. Vermeule also states that he knows of no more accurate way to compare 

 the relative total evaporation from forested and deforested areas than by measuring 

 the rainfall in comparison with the run-off, the difference of the two making total 

 evaporation as defined in the preceding. In this way he says we obtain natural 

 conditions, and include in our computations not only direct evaporation from the soil 

 but also the water absorbed by vegetation, much of which is exhaled into the 

 atmosphere. He also holds that this method is far preferable to the attempts to 

 measure evaporation on a small experimental scale which have frequently been made. 

 Working on this line he ignores all of the valuable data of forest meteorology which 

 have been obtained abroad, and which cannot be safely ignored by any person 

 attempting to study this question on its merits. However, by ignoring such data and 

 basing conclusions only on a comparison of rainfall with run-off as determined by 

 several series of gaugings of streams extending over a number of years, Mr. Vermeule 

 is forced to conclude that " the effect of our New England and Middle state-forests 

 upon the total run-off of streams, hence upon evaporation, is not important enough to 

 be shown in the measurements of stream flow." I can not but think that in stating 

 this proposition so strongly Mr. Vermeule has in reality — although without doubt 

 inadvertently — written that which on the whole tends to perpetuate an exceedingly 

 mischievous error. I hold indeed that had he studied the subject more broadly he 

 would, with his fine turn for philosophical analysis, have arrived at quite different 

 conclusions. In discussing this proposition, Mr. Vermeule depends upon data derived 

 mostly from observations in New Jersey. My first criticism of his data is that with 

 the exception of the records applying to the drainage areas of the Passaic and 

 Hackensack Rivers, the periods covered are entirely too short for any final indication 

 whatever. Certainly, they are far too short for any such sweeping generalization as 

 that just quoted. The Passaic record which he cites in his report on forestry in 

 Northern New Jersey, in the annual report of the State Geologist of the State of New 

 Jersey for the year 1895, covers a period of seventeen years, while the Hackensack 



