tions have been substituted, as indicated by footnotes. (See Appendix C for the complete list of these cities and the nearby stations which have been substituted.) No temperature daca are given for 19 cities as there were no nearby representative stations with a sufficient period of record to provide mean values. Items 456–457. Mean temperature for January and July.

Source: Weather Bureau. Decennial Consus of United States Climato—Amcaly Normals of Temperature, Precipitation, and Houting Degree Days (1932–1967), 1962: Climatological Data. monthly and annual; and Zoeni Cimaroingwai Datu, monthly and annual.

For most stations mean temperature figures are based on the period 1932-1960. The means are adjusted to represent observations taken at the present standard location.

See general note for Dems 43643.

Items 468–459. Record high and low temperature.

Source: Same as for Items 436–457

Recori temperature igures present the highest and lowest temperatures ever recuried. While these figures are related the length of record (which is not strictly uniform for all cines shown, they are considered adequate for usual comparisons. It should be borne in mind, however, hat der indicate exteme conditions and that temperatures approaching these figures secur

Items 460–463. Mean annual precipitation, possible sunshine, hourly wind speed, and degree days.

Source: Same as for Items 476-K7.

Mean annual precipitation figures are derived in the same manner as the mean temperature figures (see text for Items 656-67).

Mean annual percent figures of possible sunshine are for stations having automatic sunshine recorders for a considerabie period of time and for which sunshine records have been summarized. These percents represent the relationship between the n'imber of hours of sunshine as recorded by instrument and the number of hours between sunrise and sunset for each day during the year.

Mean annual hourly wind speed figures were taken from recoris at stations equipped with wind instruments. The means repre sent averages of the hourly wind speed for the period of record. Mean annual fegree days, 65° base, is the mean of the annual sums of the diferences between 45° and the faily mean temperatures below 65° Temperatures of 65° or higher do not afect the degree day total. The data were compiled from the tocal number of degree lays each month for the period 231-1960. The fegree day irure is based on the assumption that the amount of frei or heat used per day 50 heat buildings a empericare of approximately 30° is proportional to the number of degrees that the outside temperature falis below 15°

See general note for Items 456–43.