It is interesting to read the weather forecasts from the early 1900s. They are almost too general, they come with very little specifics. I am no meteorologist, but I wonder if it is because a) the science of forecasting the weather was not fully developed, and/or b) there was little reason for most people (beyond farmers, perhaps?) to be terribly concerned about the detailed specifics of the weather.

Here's an example, from the July 30, 1906 NYT:

The Weather Bureau forecast for Monday and Tuesday in the cotton States is as follows:

North Carolina - Showers Monday; Tuesday, partly cloudy; fresh west to northwest winds.

South Carolina - Occasional showers Monday and probably Tuesday; light to fresh southwesterly winds.

Georgia - Showers Monday; warmer in northwest portion; Tuesday, fair; warmer in northern, showers in souther, portion.

Eastern Florida - Showers Monday and probably Tuesday; variable winds.

Western Florida and Alabama - Showers Monday; Tuesday partly cloudy and warmer.

Mississippi - Partly cloudy Monday; warmer in southeast portion.

Louisiana - Generally fair Monday and Tuesday; warmer Monday in southeast portion.

Eastern Texas - Fair Monday and Tuesday.

Western Texas - Fair Monday and Tuesday.

Now, July/August in Texas is usually anything but "fair." What with temps in high 90s and low 100s for days on end, to term that "fair" is either northeastern bias or testament to the hardiness of the folks who lived in the south without air conditioning.

In a separate article, there is a table of temperatures in New York City - one of the first I have come across. The notes to the table suggest the following:

The average temperature yesterday was 76; for the corresponding date last year it was 74; average on the corresponding date for the last twenty-five years, 73.