370 University of California Publications in Agricultural Sciences [Vol. 3 



Pendleton's residence in India since the paper was written by him 

 has rendered satisfactory changes and deletions practically impossible. 

 Under these circumstances, with the burden of preparing the paper 

 for the press and the reading of the proof falling to me, the author 

 cannot well be held responsible for the inaccuracies and the infelicities 

 of expression which have been carried over from the original manu- 

 script without change. Moreover, the investigation was carried out 

 under my direction, and the plan of attack on the problem, together 

 with the methods employed, were suggested by me. Much that, in the 

 light of present knowledge, is superfluous or patently inexact or 

 erroneous in the paper is due to points of view held by me in 1915, 

 but now happily discarded. For all these, I assume the entire 

 responsibility, and absolve Mr. Pendleton in that regard. 



On the other hand, the work having been carried out at my sugges- 

 tion and under my direction. I feel constrained, in .justice to myself, 

 to say that the views expressed in this paper, and the conclusions 

 drawn are wholly Mr. Pendleton's and are not in agreement with those 

 held by me. I fail to see the cogency of the arguments set forth for 

 soil classification and mapping at this juncture in soil studies, and 

 cannot admit the pertinence of the analogy between classification of 

 other objects and of soils which the author of this paper employs. 

 My own general conclusion from the results obtained by Mr. Pendle- 

 ton is that they cast grave doubt on the validity of the Bureau of Soils 

 method of soil classification and mapping, and, incidentally on all 

 methods devised for that purpose to date. I cannot see how such 

 methods can serve us in scientific work at all. and. from the practical 

 standpoint, it would surely seem that guides for the purchaser of land 

 could be arranged more cheaply and less elaborately than by the soil 

 mapping methods extant. This statement has particular reference to 

 the subdivision of types very minutely, such as. for example, sandy 

 silty clay, clay loam adobe, etc. Such minute classification and sub- 

 division in view of the present state of our knowledge of soils, is 

 analogous, in my opinion, to carrying figures out to four decimal 

 places when it is known that the accuracy of the method makes it 

 impossible for them to be correct beyond the first decimal place. In 

 support of this seemingly radical conclusion, the reader will find much 

 of interest in the recent studies of this laboratory on variability in 

 soils, which have already appeared in this same series. 



Chas. B. Lipman. 



INTRODUCTION 



For several years the University of California has been cooperating 

 with the United States Bureau of Soils in the mapping of the soils of 

 the agricultural portions of the State of California. The system of 

 mapping used is that developed by the Bureau of Soils. During the 

 year 1914—1915 the writer, representing the University of California, 

 was engaged in some of this soil survey work. In that year, in the 

 field, many questions arose regarding the criteria used, the methods. 



