Two farmers are grumbling about the weather. The scene is Ohio, the time July, and the prevalent crop corn (i. e., maize).

Farmers have grumbled about the weather from time immemorial. The point of interest in this particular case is that the two grumblers do not agree about what is wrong. Farmer A thinks the corn needs rain. Farmer B declares that at this stage rain would do more harm than good. Plenty of warm sunshine is, he thinks, the right prescription to insure a “bumper” crop. Of course Providence will do as it pleases, and whatever weather comes, since it cannot be cured, must be endured; but it is a matter of practical as well as academic interest to get some inkling betimes as to how your crop is going to turn out, and the weather is likely to be the decisive factor. Moreover, it is a very significant fact that our two farmers are not of the same mind about which atmospheric blessing is in default. It is painful to reflect that an enormous amount of grumbling about the weather on the part of the rustic community must, at one time or another, have been misapplied. It is a plausible assumption that farmers have sometimes worried themselves to death over meteorological events that were either harmless or actually beneficial to their crops.

How can we arrive at the facts? Admitting, as everybody does, that the weather has a preeminent influence upon plant life, is this influence susceptible of analysis? Is there anything definite about it? Are not the effects of various atmospheric conditions so entangled with one another, and with the effects of soil and methods of cultivation—to say nothing of insects and plant diseases—as to baffle all attempts to gauge them separately?

There is a new branch of applied science that teaches farmers how to grumble right about the weather. It is called Agricultural Meteorology. As a coherent branch of knowledge, this subject is so new that the first formal textbook about it in the English language was published in the year 1920. It happens that the author of this book, Professor J. Warren Smith, of the United States Weather Bureau, began his investigations in the new field by making a careful study of the relation of weather to the yield of corn in Ohio. Let us see what light his studies shed upon the question at issue between our friends A and B.

Day after day, and year after year, the principal atmospheric conditions are observed and measured at a great number of points scattered over the State of Ohio, as they are elsewhere throughout the Union, and the records thus obtained are carefully compiled, summed up, averaged and otherwise discussed by officials of the Weather Bureau. Thus a great fund of detailed statistical information about the weather is available for comparison with the statistics gathered by other agencies concerning the yield of crops and their condition at different stages of growth.

Professor Smith’s analysis of the Ohio records revealed a fact of so much practical importance that this discovery alone suffices to place agricultural meteorology among the most fruitful branches of knowledge cultivated by mankind. He discovered that the success of the Ohio corn crop depends chiefly upon the amount of rain that falls during the month of July. The normal rainfall of that month for the State is 4 inches, while the average yield of corn during the past sixty years has been 34.5 bushels per acre. Comparing the values for individual years, it is found that the yield is strikingly sensitive to variations from the normal July rainfall, and especially so when the rainfall is a little more or less than 3 inches. Near this critical rainfall point, a variation of one-fourth inch of rain in July means a variation in the value of the corn crop of Ohio of nearly $3,000,000, and a variation of one-half inch makes an average variation in the value of the crop of more than $7,500,000. When the rainfall for July averages over 5 inches the probable yield of corn will be more than 27,000,000 bushels greater than it will be if the rainfall averages less than 3 inches. In other words, this difference of 2 inches in the rainfall for the month of July adds $13,650,000 to the income derived by Ohio farmers from corn alone.

Variations of the temperature in July, in Ohio, have been compared with variations in the yield of corn, with the result that the temperature of the month appears to have little effect upon the crop. Thus we find Farmer A to have been right and Farmer B wrong; but both were merely expressing personal opinions based upon an insignificant sum-total of experience. Science rests upon a surer foundation.

Although the case of the Ohio corn crop is probably simpler than most of those that agricultural meteorology has to deal with, for the reason that a single meteorological element is, in this case, of decisive importance, it illustrates a rule of quite general application that has recently come to light; viz., that in the growth of any particular crop there is usually a rather brief “critical period,” when it is most sensitive to the influence of weather. For corn, in a considerable area of the northern United States, this period is July, or more specifically, in Ohio, the interval from July 11 to August 10. The rainfall and temperature of other months have, however, definite though minor influences, which can be evaluated for the same regions.

With respect to the American “corn belt” in general, it is not certain how far the rules deduced for Ohio are applicable. Professor Smith has been inclined to look upon July rainfall as the dominating factor for the whole of that region; so that, for example, a difference of 1 inch in the rainfall (viz., a total for July of 4.4 inches or more, as compared with 3.4 inches or less) has been held responsible for an increase of 500,000,000 bushels of corn in the eight principal corn-growing States. It has also been stated that in the four States of Indiana, Illinois, Iowa and Missouri an increase of half an inch of rain in July meant an increase of $150,000,000 in the value of the crop. These figures have, however, been challenged, and the subject is still under discussion.

The study of the critical periods of different crops, and the determination of the amounts of heat and moisture most favorable to the success of the crop at such periods, may be regarded as the leading task of the agricultural meteorologist. The most elaborate researches of this character have been made in Russia by Professor P. Brounov, who founded in 1896 a meteorological bureau, attached to the Ministry of Agriculture, with an extensive network of stations scattered over the Russian Empire. This bureau was quite distinct from the ordinary meteorological service, under the direction of the Central Physical Observatory in St. Petersburg. Just before the war Professor Brounov had 150 stations in operation; most of them for observing the effects of weather on the leading cereal crops, though some studied the corresponding relations of horticulture or the animal industries. Each agricultural station comprised a small plot of land, on which a certain sequence of crops was grown year after year under conditions of cultivation as nearly uniform as possible, the only variable factor being the weather. Meteorological instruments were installed in the immediate proximity of the plants under investigation. Prior to 1914 Brounov had determined the critical periods of most of the crops grown in Russia, and had published a great deal of information on this subject that could be turned to practical account by Russian farmers.

It will perhaps not be immediately apparent to the reader just how such information can be utilized. Its practical applications vary, in fact, according to circumstances. First of all, a knowledge of critical periods and of the weather requirements of crops at these periods enables the farmer to select his crops and time his farming operations on the basis of climatic statistics. Brounov published a series of charts showing the probability of dry weather, as deduced from many years of observations, for each ten-day period throughout the agricultural year for every part of European Russia. With such charts at our disposal, and knowing how long after planting each crop arrives at its critical period with respect to moisture, we can readily estimate the probable success of a given crop planted at a given time and place; at least, so far as this is determined by rainfall. If temperature or other meteorological conditions are of special importance at the critical periods, we shall need additional climatic charts. Of course, the weather in any particular year may differ widely from the climatic averages; but in the long run crops will prosper in proportion as their critical periods coincide with the occurrence of favorable weather as shown by the climatic record. It will be seen that this is quite a different idea from the traditional one that a certain crop needs a “moist climate,” another a “hot climate,” etc. The agriculturist now asks the man of science to tell him when, between planting and harvesting, heat or moisture is of vital importance to the crop, and how much of each will produce the biggest yield.

In regions where irrigation is practiced it is obviously advantageous to the farmer to know at what stage of its growth a crop becomes sensitive to the amount of moisture received. During the greater part of its life the plant may be quite indifferent to moisture, and at such times irrigation would be wasteful. The farmer needs to know not only when the critical period has arrived, but also what the water requirements are at that period. Too much water may be as bad as too little.

Even when agricultural practice ignores the rules laid down by the agricultural meteorologist, a knowledge of these rules may be applied with great advantage to the prediction of crop yields. It is hardly necessary to tell any farmer or business man that accurate crop forecasts are an economic desideratum of the utmost importance. The United States Government maintains an army of more than 200,000 volunteer crop reporters, supervised by a staff of experts, for the purpose of determining month by month the condition of every agricultural crop and its prospective yield. With regard to the monthly announcements of the Bureau of Crop Estimates, Professor H.L. Moore, of Columbia University, says:

“The commodity markets are in a state of nervous expectancy as the time approaches for the official forecasts, because great values are at stake. It has been estimated that in the case of the cotton crop alone an error in the forecasts which should lead to a depression of one cent a pound in the price of cotton-lint would—assuming a crop equal to that of 1914—entail a loss of eighty million dollars to the farmers. The vast values at stake and the dangers when no official estimate is available of the manipulation of the markets in the interest of speculators are held to justify the large recurrent annual cost of the employment of the numerous correspondents, clerks, and experts.”

Professor Moore is one of those who have pointed out that the forecasts based upon the actual condition of the growing crops can be vastly improved by a mathematical analysis of the weather reports from the various regions in which the crops are grown. In fact, he goes so far as to assert that much better forecasts can be made from the weather reports alone than from reports on the condition of the crops. Whether or not this view is unduly optimistic, it goes without saying that the precise data which agricultural meteorology is now acquiring cannot fail to enhance greatly the accuracy of crop forecasts.

Of course, the weather has always been watched with keen interest by everybody concerned with the purchase or sale of agricultural products and has been one of the chief factors determining the rise and fall of prices. At produce exchanges throughout the United States daily weather bulletins are received from the agricultural districts, and at many of them a large weather map is drawn every morning by an employee of the Weather Bureau detailed for this purpose. The Bureau has made various other arrangements for supplying the information that is so eagerly desired concerning the weather as it affects crops, as well as the animal industries. During the “growing season” in the cotton, corn, wheat, sugar, rice, broom-corn and cattle-producing areas, designated centers receive telegraphic reports of rainfall and the daily extremes of temperature from substations in the regions concerned, and these are distributed in bulletin form. Each local center, besides publishing detailed reports from its own area, issues condensed reports from all the others. The Bureau also issues every week during the agricultural season a “National Weather and Crop Bulletin,” with text and charts setting forth the current conditions of moisture, temperature, etc., and the state of the crops in all parts of the country.

The use which dealers and farmers make of these weather reports is, however, very far from having been reduced to science. Some of these persons, it is true, are frequently able, by a purely instinctive process of deduction, to make successful forecasts of crop yields from a close study of the weather, and others have worked out crude rules of their own for the same purpose; but the agricultural meteorologist approaches the problem in a different way. Immense progress has been made in the past decade in applying the mathematical theory of correlation to this problem. This branch of mathematics, originally developed chiefly for statistical studies in biology by Galton, Pearson, and others, is now extensively used by meteorologists not only for studying the effects of weather on crops, but also for finding out what correspondences or relationships exist between variations of weather in different parts of the world, as well as between weather and sun spots, weather and vital statistics, etc.

Sometimes, when the farmers do not disagree on the subject of favorable and unfavorable weather for the crops, they hold opinions in common that agricultural meteorology is unable to substantiate. An illustration is found in the idea that a good covering of snow during the winter is favorable to the yield of winter wheat. Apparently this is one of the host of popular ideas that are based merely on the delusive foundation of “everybody says so.” Smith has investigated the statistics of wheat for Ohio and C.J. Root those for Illinois. In both cases their results negative the prevailing opinion. Professor Smith finds “some evidence to indicate that wheat has a better prospect if it is not covered by snow during the month of January,” while Mr. Root states that, in general, “the winters of light snowfall are followed by good wheat yields and the winters of heavy snowfall by light yields.”

The study of the relations between weather and crops is really a branch of a science of broader scope, known as phenology. This science is devoted to the investigation of all periodic phenomena of plant and animal life that are controlled by the weather. There are, in some parts of the world, large corps of phenological observers, who maintain records year after year of the leafing, flowering, and fruiting of both wild and cultivated plants, the migrations and first songs of birds, and various other events of a biological character that recur with the seasons. In the course of time it becomes possible to compute from such records the normal dates of these events; and then, in any particular year, a comparison between the actual dates and the normal shows whether the season is early or late, and by how many days. Phenological observations on plants also make it possible to draw charts showing the normal march of the seasons over a country, expressed in terms of plant life, and such charts are often more valuable to the agriculturist or horticulturist as a guide in selecting varieties for cultivation and in timing his operations, than any charts that can be compiled from ordinary climatic data. Some admirable charts of this kind have been drawn for parts of Europe.

There are many practical applications of phenology to agriculture, and there would be more if phenological observations had been made more extensively throughout the world. Good phenological charts of different regions would, for example, greatly facilitate the work of foreign plant introduction carried on by the United States Bureau of Plant Industry. In the United States phenological observations were made systematically between 1850 and 1863, but only desultory work has been done in this line subsequently. The most comprehensive individual record is that maintained by Thomas Mikesell from 1873 to 1912, at Wauseon, Ohio, and published in full by the Weather Bureau in 1915.

The old rule of American farmers, inherited from the Indians, that the time to plant corn is when the leaf of the white oak is “the size of a mouse’s ear,” illustrates the use that can be made of so-called “index plants” of the native flora as guides for farming operations. Professor A.D. Hopkins writes on this subject:

“If such guide plants do not occur on the farm, they can be found among the ornamental trees and shrubs and hardy flowering plants of other localities or countries and transplanted. The periodical event of the falling of the flower catkins of the Carolina poplar has been found to be one of the best guides to the general early or late character of one season as compared with the average, while the opening of the leaf buds and unfolding of the leaves serve as reliable guides to the progress of spring. The various magnolias in their succession of flowering events serve as excellent guides to the rate of progress of spring and the time to do various kinds of work. The ornamental SpirÆas, Deutzias, Diervillas, climbing roses, and Clematis among the ornamentals, and the dogwood, service tree, redbud, and oaks among the native trees of the middle and eastern regions of the United States are more or less constant in their response to prevailing local influences which are indicative of the time to plant certain field and garden crops. The opening of the leaf and flower buds and the flowering of the common fruit trees and shrubs of almost every farm serve as more or less reliable guides to the time to spray for certain insect and plant diseases.”

Dr. Hopkins has worked out an interesting rule known as the “bioclimatic law,” according to which the periodical events of plant and animal life advance over the United States at the rate of 1 degree of latitude, 5 degrees of longitude, and 400 feet of altitude every four days—northward, eastward, and upward in spring, and southward, westward, and downward in autumn. Thus, when the date of any phenological occurrence is known for one locality, it may be approximately determined for any other. This law has enabled the Department of Agriculture to publish rules of general application concerning the best time to plant winter wheat in order to escape the ravages of the Hessian fly, thus saving many millions of dollars to American farmers. The same law is susceptible of various other profitable uses.

The United States Weather Bureau has been a branch of the Department of Agriculture since 1890, and a very large share of its routine work is devoted to the agricultural interests of the country. The climatological statistics that it has assembled are indispensable in many departments of agricultural research, besides furnishing varied information of practical value to farmers. The Bureau has developed a number of special types of forecasts for the rural industries; such as predictions, three or four days in advance, of favorable weather for cutting alfalfa; forecasts of weather unfavorable for sheep-shearing; notices to fruit growers of dry-weather periods in which fruit trees should be sprayed; and warnings of the occasional summer showers that would do so much damage to the great raisin-drying industry of California but for the vigilance of the forecasters and the efficient arrangements made by the industry itself for disseminating and acting upon the warnings. Of course, the ordinary daily weather forecasts, storm warnings, and cold-wave warnings are valuable in many ways to agriculturists, and the Bureau has made great efforts to give such information prompt and general distribution in the rural districts. The forecasts are generally displayed in post offices, and in many cases the rural telephone exchanges are pressed into service to distribute weather information regularly to all their subscribers. Some exchanges sound a signal every morning when the forecast is ready for distribution. Lastly, the wireless telegraph and the wireless telephone, which, in the immediate future, will form part of the equipment of every up-to-date farm, afford ideal channels for the dissemination of weather news and are already extensively used for this purpose.

There remain to be mentioned the various steps the Weather Bureau has taken to protect the rural industries from the night frosts of spring and autumn, in the shape of special forecasting arrangements, the publication of frost charts, and a wide range of scientific investigations. The Bureau’s undertakings in this line are merely a part, though a leading one, of a great campaign of frost protection that is being carried on by scientific and official agencies in this country on a larger scale than anywhere else in the world.

Frosts, classified according to their severity as “light,” “heavy,” and “killing,” are most likely to occur in spring and autumn, when an extensive area of high barometric pressure brings its usual accompaniment of clear skies and calm nights. They are predicted on a general scale from the weather map, and locally from indications of temperature and humidity and a knowledge of important topographic influences, such as those due to hills and valleys and neighboring bodies of water.

In agricultural usage the term “frost” is applied to the occurrence of a temperature low enough to kill or injure tender vegetation, such as growing vegetables or the buds, blossoms, and fruit of fruit trees. The occurrence of a frost, in this sense, is not necessarily identical with the deposit of ice crystals known as “hoarfrost.” Different species and varieties of plants are, of course, susceptible in very different degrees to the effects of low temperature; i. e., they differ greatly in “hardiness.” In the case of fruits and vegetables the danger point generally lies a little below the freezing point of water (32 degrees F.).

The occurrence of frost is favored by the rapid cooling, by radiation, of the earth and its plant covering, which goes on at night under a clear sky and in still air. Under these conditions a layer of stagnant, cold air forms close to the ground, with warmer air lying above it. The difference in temperature at different levels is often so pronounced that fruit on the lower branches of a tree is killed while that growing on the higher branches remains uninjured. Similarly, frost will occur in the bottom of an inclosed valley but not on the surrounding slopes. In the case of a valley the layer of cold air that forms at the bottom is commonly deepened by additional cold air draining down from the hills.

Many large orchards have their “warm spots” and their “cold islands” or “north poles,” well known to the orchardist; due in some cases to the nature of the soil rather than to topography. Certain mountain regions in North Carolina are famous for their “thermal belts” or “verdant zones”; i.e., areas part way up the slopes that escape the frosts occurring both above and below them. These frostless belts, which have been the subject of numerous investigations for three-quarters of a century, seem to mark the upper level of the pool of cold air that collects in the valley by drainage from the mountainsides. A detailed temperature survey of the thermal belt region of North Carolina was made during the years 1912–1916 by the United States Weather Bureau and the North Carolina State Board of Agriculture. In some places the minimum temperature at night was found to be 15 or 20 degrees higher in the thermal belt than at the bottom of the valley, a few hundred feet below.

Clouds, by checking radiation from the earth, and wind, by mixing the colder and warmer layers of air together, both prevent frosts that would otherwise occur. Artificial methods of protection include covering plants with screens of wood, paper, or cloth, building smudge fires to provide a blanket of smoke (a method of doubtful value), and, above all, heating by means of wood fires or various types of “orchard heater,” burning either oil or coal. An elaborate technique of orchard heating has been developed, having in view especially the most economical use of fuel and labor consistent with the object to be attained. In many cases orchards are provided with alarm thermometers, which ring a bell when the temperature approaches the danger point in the orchard. The local prediction of frost from the readings of meteorological instruments is a problem that has not been fully solved. The idea formerly prevailed that the temperature of the dew point, as determined from readings of the dry-bulb and wet-bulb thermometers in the early evening, was a safe guide to the fall of temperature to be expected during the night, but this belief has not stood the test of accurate observations. At the present writing certain formulas involving data of both temperature and humidity are being used experimentally by Weather Bureau specialists for predicting the lowest temperature of the night when the general conditions indicate that frost is possible. A comprehensive discussion of this subject has been published by the Bureau as Supplement No. 16 of the “Monthly Weather Review.” (Washington, 1920.)