No. 9.-Table of the Occurrence of Frost, (Earliest and Latest) from Observations made at Glenwood, near Clarksville, Montgomery County, Tennessee, 1851 1873 inclusive.

The following table, No. 10, which has been kindly supplied by Prof. Payne, presents similar data for the years 1871 to 1873 inclusive, at Knoxville:

This table is good so far as it goes, but the observations of at least a decade of years are required before very satisfactory comparisons can be made.

According to the Glenwood table, the length of the growing season in the northern part of Middle Tennessee is 189 days. It is seen, too, that April and October respectively are pre-eminently the frost months of spring and autumn. From the third week in April to the middle of October the farmer can afford to risk the occurrence of frost; it may come within these dates, but the probabilities are against it. In the southern part of the State the period of no frost is twelve days, or two weeks, longer, and embraces, therefore, 200 or 203 days. This difference is of considerable importance to the cotton region of the State.

Rain and Snow. In general, the quantity of rain (including snow) which falls upon the surface of Tennessee, although not so great as that precipitated upon the States further south, is amply sufficient. The summers are sometimes too dry, but severe droughts are excep tional. It may be said that the rainfall is just enough to ensure a vigorous growth of vegetation and not too much to interfere with the proper cultivation of the earth.

The average annual fall of rain upon the surface of the globe is about sixty inches. In the Torrid Zone, which, by reason of its greater circumference, has a far larger surface than either the Temperate or Frigid Zones, the annual fall of rain is 96.5 inches, in the Temperate Zone 36.5, and in the Frigid Zone 12.25. If the whole amount of rain which falls annually were collected into a single place, it would be sufficient, according to Commodore Maury, to form a lake 24,000 miles long, 3,000 broad, and sixteen feet deep.

4.7

The following table, No. 11-another of Prof. Stewart's-is our main dependence for conclusions as to the rainfall of the State. It will be seen from this that the mean quantity of rain and melted snow for twenty-one years, at Glenwood, is 45.715 inches, or in round numbers, 46 inches. Of the months, April has the highest average, inches; December follows, 4.6; then March, 4.4; February, 4.2; May, 4 inches, &c. The dryest month is October, 2.5 inches, while September has but little more, 2.9. The greatest rainfall in any one year was in 1865, when it amounted to 60 inches; the least was in 1872, it being 34 inches in round numbers.

No. 11.

Table showing the Monthly Rainfall for the years 1852 to 1872 inclusive, at Glenwood, Montgomery County, Tennessee, given in inches and thousandths. The Rain-water is supposed to have remained where it fell, without running

MONTHS.

January..
February
March.
April.....

off or evaporating, the Snow (melted) being estimated as Water.

Means

1852. 1853. 1854. 1855. 1856. 1857. 1858. 1859. 1860. 1861. 1862. 1863. 1864. 1865. 1866. 1867. 1868. 1869. 1870. 1871. 1872. for 21

Years

In. In. In. In. In. In. In. In.
2.010 0.740 5.277 4.667 1.381 1.988 4.211 2.572 6.143 2.210 5.136 8.582 2.428 2.378 4.365 1.612 5.030 2.867 8.385 4.833 1.330 3.721
6.700 5.200 1.751 1.531 2.897 2.760 3.176 8.227 5.269 2.701 8.230 6.456 1.127 3.970 4.145 8.988 1.310 3.137 3.027 6.514 2.400 4.262
2.220 2.410 9.506 3.851 1.640 0 506 4.004 4.068 1.478 2.151 4.767 5.032 3.911 7.983 5.067 8.282 5.519 5.068 3.465 9.776 2.470 4.437
3.050 4.046 1.367 2.598 4.176 2.707 6.565 5.179 2.832 5.440 8.015 3.271 3.567 11.330 4.396 4.897 7.605 5.283 2.656 4.761 5.680 4.734
5.610 4.383 7.010 3.783 6.299 5.997 4.587 5.462, 2.933 5.838 2.457 2.744 1.880 3.386 2.110 4 555 3.217 2.465 3.763 4.330 1.705 4.024|
1.330 0.900 2.887 3.923 1.231 4.234 6.546 3.464 3.524 2.488 4.971 8.862 2.979 2.675 2.843 2.533 1.915 4.395 8.108 4.014 2.637 3.641
3.510 3.671 2.370 3 572 1.857 4.614 4.125 0.935 1.445 7.026 5.751 2.243 1.757 5.153 4.929 4.052 4.410 2.132 5.970 2.575 7.903 3.809
August.. 4.170 4.525 1.256 4.562) 3.952 5.077 1.692 4.654 2.252 7.850 1.204 1.758 6.207 4.923 0 675 1.300 3.018 1.000 4.275 2.410 1.330 3.242
September, 3.640 4.163 2.231 7.298 0.716 0.258 2.112 2:120 2.888 3.230 2.489 1.997 2.056 6.186 5.601 0.545 3.802 2.545 3.145 2.550 2.235 2.943
October. 1.810 2.675 5.696 1.400 2.430 3.018 2.586 1.069 2.706 2.090 0.666 6.041 3.904 1.034 2.240 1.975 1.556 2.736 2.150 2.085 2.675 2.502
November.. 3.320 2.440 2.388 4.657 6.238 6.591 2.601 5.651 5.692 5.082 2.675 2.598 7.627 1.223 4.722 5.935 2.024 3.400 1.170 1.546 1.720 3.776
December 11.500 1.815 1.290 2.265 6.680 5.644 11.476 5.188 2.706 1.941 3.280 4.762 3.867 9.763 3.712 4.249 5.371 4.135 3.070 2.544 1.800 4.621

In. In. In. In. In. In. In. In. In. In. In. In. In.

May..
June.
July..

Annual

In. In. In. In. In. In. In. In. In. In. In. In. In. In. In. In. In. In. In. In. In. In.
Amount 48.870 36.968 43.029 44 107 39.500 43.394 53.681 48.589 39.868 48.047 49.641 54.346 41.310 60.004 44.805 48.923 44.777 39.163 49.184 47.938 33.885 45.712

Mean for twenty.one years, 45.715.

The annual means of the above table have been arranged by Prof. Stewart in a diagram, for the purpose of exhibiting any relation that may exist among them. The diagram is given below. Each space between the horizontal lines represents ten inches. The lowest line begins with 30 inches, as no mean falls below this. The annual means are represented by the points in the vertical columns, each under its respective year. Successive points are connected by straight lines, so that a zigzag is formed, running through the diagram. The dotted line represents the average of the oscillations, and shows that in the period beginning with 1851 and ending with 1865, there was a general tendency to an increase in the rainfall, and that since, the tendency has been in the opposite direction.

Prof. S., in a note with reference to this diagram, says: "In looking over my records, I can trace an ascending line (in the tables of precipitation) in the advancing years; very much zigzagged by the oscillations from year to year, but still generally ascending. Taking the year 1851 as the minimum, there is a general increase in the annual quantities of rain which culminates in a maximum in 1865. The tables give, it is true, a double curve, but the general mean line is unmistakable. It would appear, therefore, that the period of these observations covers two-thirds of a curve of a great oscillation (21 years), and that a whole oscillation, from maximum to maximum, would require thirty years. The tables appear to show that, since 1865, the curve has been falling to another minimum. Whether this will be realized or not, will remain to be determined by future observations. As it seems almost certain that in other meteorological phenomena there are such periodical oscillations, it would be highly interesting to determine whether this is the case with the aqueous meteors."

In order that all the data we have may be recorded, the following tables are appended. No. 12 is from Prof. Payne, and presents results of observations made at Knoxville in 1873; No. 13 from Mr. A. C. Ford, and presents results obtained at Nashville for the same year; No. 14 is from Glenwood, for same year:

The table below, No. 15, is from Dr. Safford's Report:

No. 15.—Quantities of Rain and Melted Snow for the Seasons and the Year, in inches and hundredths of an inch. The Years of Observation are given under each Station.

Tables 12, 13 and 14 supply data for comparing the rainfall of 1873 at three different points, one in East, the others in Middle Ten