With reference to climate in particular, the statistical method was employed by Quételet, Bertillon, Leffingwell, Ferri, Holzendorf, Guerry, Curcio, Lombroso, and others, who established a parallelism, or coincidence, between certain climatic features and the criminal conduct of man.
Delimited aspects of environment, relating again more to climate than any other phase of the milieu, were made the objects of observational or experimentally observational studies by Dexter, Brunhes, and Hellpach, the last two giving the most recent comprehensive summaries of our knowledge in this field. And they are among the best we have.
The next part of this study will continue the survey of the history of this theory in the above mentioned sciences as well as in literature.
APPENDIX
Since the foregoing study was completed, E. Huntington’s stimulating book—vide supra, p. 79, n.—on Civilization and Climate has appeared. He continues what Dexter began. Lack of definiteness in observation, argumentative conviction, reasoned out opinion, are superseded by scientific exactness in ascertaining the action of climate. Chapters 4–7 (pp. 49–147) concern us here. In these chapters he investigates “the exact effect of various climatic factors upon selected groups of people” (p. 49).
Huntington subjects to statistical analysis the daily records of about 550 factory operatives, pieceworkers, employed in three factories in three New England cities. The records, most of them for a complete year, are distributed over the four years from 1910 to 1913 (p. 53).
He computes wage averages. He finds for each working day the average hourly wage for each group of operatives. When the daily averages had been found, they were averaged together by weeks. To give each individual an equal importance, the figures of each group have been reduced to percentages. Finally, the different groups were combined (p. 57). His final computations are represented in curves. A curve, graduated in twelve parts (one for each month), for a given year shows the earnings in percentages at any point and thus reveals the time of the weakness or efficiency of the worker; it shows the time of his wages from least to most, thereby indicating the time of his work and energy from poorest to best.
Huntington worked up similarly the records of 65 operatives in a North Carolina factory, of 240 operatives in four cotton mills in South Carolina and Georgia, of 57 carpenters at Jacksonville, Fla., and on a different basis the work of 2700 cigar makers in two cigar factories in Florida. On the first basis he also computed a series of data from a large factory at Pittsburgh, Pennsylvania, based on the work of about 950 operatives in 1910, of about 750 in 1911, of 69 in 1912, of about 7000 in 1913. He figured the monthly or bi-weekly averages of hourly earnings of these pieceworkers in Pittsburgh.
Discussing the curves in Figure 1 (p. 59), he mentions (p. 61) five features revealed by the curves that show no sign of disappearing. They are: “an extremely low place in midwinter, and a less pronounced low place in midsummer; a high point in June, a still higher point at the end of October, and a hump in mid-December....
“Before we discuss the causes of the variability of the summers let us consider the meaning of the curves as a whole. In the first place, it is evident that, although details may vary from year to year, the general course of events is uniformly from low in the winter to high in the fall with a drop of more or less magnitude in summer. To what can this be due?...