To the Editor:—
A communication from Dr. Paul D. Rosahn (The Journal, October 16, p. 1294) analyzed statistically the data on the seasonal incidence of acute coronary artery occlusion presented by Drs. Dack, Jaffe and myself in an article entitled "Factors and Events Associated with the Onset of Coronary Artery Thrombosis" in The Journal, August 21. We divided the seasons into autumn-winter (October to March inclusive) and spring-summer (April to September inclusive). The difference in the incidence of coronary artery occlusion in these two groups was only 2.6 per cent. Dr. Rosahn states that "when, however, a slightly different division of the published data is made, a wholly different conclusion results." This "slightly different division" consists of winter-spring (December to May inclusive) and summer-autumn (June to November inclusive). Dr. Rosahn pointed out that when this is done the difference in incidence becomes 8.8 per cent, which may be significant from a statistical standpoint. However, this grouping is not, in our opinion, quite relevant to the problem of the influence of cold weather in New York City. We could not present, within the space limits of our original communication, data on the monthly mean temperature for New York for the years 1930-1935, published by the U. S. Department