4. The wheat sown September 29 and October 6, 1880, which headed out December 30 and February 19, was injured as to the heads by the subsequent frost.

5. The seeds sown October 20, 27, and November 3, 1880, flowered between the 4th and 8th of June, 1881, but at this time there was experienced at Paris a spell of very cold weather, the minimum daily temperature being 3.1° C., and even white frosts were reported, so that wheat which was then in flower was badly injured.

6. In general, the dates November 10, 1880, to December 15, 1880, are those indicated as most favorable for sowing wheat in that year, and the crop of 1881 may be predicted as likely to be small, but of excellent quality.

The grape and wine crop.-In a short study on the relation between the vine and the weather, Marié-Davy (1882, p. 290) states that, in general, the annuals, such as the cereals, concentrate all their energy in the formation of the ear and the seed or grain. Their work is then finished and they die. The next year's crop of these annuals is largely under the control of the husbandman, who can obtain seed from more favored regions if his own crop was inferior.

On the other hand, the work of the vine, like all perennials, is not merely to ripen its fruit and seed, but to preserve its own individual self for usefulness in future years. Therefore it elaborates out of its own sap not merely leaves and fruit and seed, but a store of woody fiber. Corresponding to this more complex system of growth the relations of the perennials to the climate are apparently more complex than the relations of the annuals, and, it may also be added, the range of geographical distribution, whether by nature or by cultivation, is more restricted.

Our studies will be confined to the data furnished by the observations at Epernay (1873-1881), to which Marié-Davy adds other data computed from the observations made at Montsouris, in which latter computation certain laws of growth of the vine as established by Gasparin were adopted.

In the neighborhood of Paris the leaf buds of the vine burst open in May when the mean daily temperature has permanently passed above 11° or 12° C. Assuming that the mean of twenty days, as observed at Montsouris, will give this date (which was unfortunately not observed at Epernay), we obtain the figures in the first three columns of the following table. In some of these years the early leaf buds were undoubtedly killed by nocturnal frosts, but they were soon replaced by other buds, and the dates here given must be adopted in the absence of actual observations, especially when we remember that the quantity and quality of the final crop of grapes depend not only

on the meteorological conditions, but, equally or more, on the condition of the woody fiber of the stock and stems. Similarly the date of flowering is calculated by assuming, with Gasparin, that the sum total of the mean daily temperatures in the shade must be 466° C., counting from the date of leafing. This number is more especially applicable to the vineyards of Champagne and Burgundy, and is not necessarily strictly applicable to Epernay or to the vineyards of the south of France. The fifth and subsequent columns of the table give the mean climatic conditions for a period of sixteen days, in the middle of which is the calculated date of flowering. There appears to be no simple relation between this latter data and the resulting wine crop, nor has the crop any apparent relation with the total sky radiation during this period. In fact we may conclude that up to the time of flowering the energy of the vine has been devoted to building up its own structure as a preparation for the work that is to come. The development of the grape does not depend upon nitrogenous particles stored away in the vine, but on the power to elaborate the sap which is to become the juice of the grape, and that power depends upon the leaf surface, the roots, and the stock during the first stages of the growth of the grape, but eventually upon changes that go on within the grape itself.

These facts are brought out by the study of the conditions prevailing during the last stages, viz, from flowering to maturity:

According to Gasparin the grape reaches its maturity when it has received a sum total of mean daily shade temperatures of 1,926° C., counting from the date of flowering, but the grape ceases to be nourished or to ripen when the mean daily temperature falls below 12.5°. If, now, we ascertain the date of maturity by summing up the daily temperatures, as required by Gasparin's rule, we obtain the results given in the first column of the following table. If we

2667-05 M- -17

a

take the mean values for the twenty days previous to the date of maturity we obtain the data in the second, third, and fourth columns, and we notice that although in the warmer seasons there may be great variety in the value of the crop, yet in the cold seasons, 1877 and 1879, when the mean temperatures fell below the limit (12.5°) required for ripening, the crop was very poor or failed altogether. If now the total radiation from sun and sky is computed according to Marié-Davy's method for the period between leafing and flowering and again from flowering to maturity we obtain the figures in the columns five, six, and seven. Here we see, as before, that the variation during the flowering period was of little importance, whereas that during the ripening period has a direct relation to the character of the wine crop, such that in general the larger the total radiation the better the crop, provided the temperature of the air has not fallen too low.

In general, Marié-Davy concludes that the number of grapes to the bunch and the number of bunches to the vine do not seem to have any clear relation to meteorological conditions, except in the case of spring frosts, which can destroy a crop. Besides the conditions as to pruning the vine and dressing the soil, the number of grapes that have set (on which principally depends the quantity of the crop that will be produced) is a result primarily of the meteorological conditions during the previous year and of the state of preparation of the woody stock. On the contrary the final size of the grapes and the quality of the juice depends on the meteorological conditions of the crop year and those that accompany the flowering and succeed it up to the time of maturity. A final sum total of radiation is not sufficient; it is necessary to take account of its distribution with reference to the

phenological periods and of other accompanying circumstances. Thus in 1877, with a low mean temperature and a high radiation during maturity, and in 1879, with a low temperature and a low radiation during the last phase, both alike gave a poor crop, but the sunshine of 1877 was able to make a large quantity of sugar as compared with the small quantity of sugar in 1879.

Sugar beets.-Marié-Davy (1882) and Pagnoul (1879) give the data of a research into the relation of climate to the development of sugar beet as cultivated at Arras, the agricultural station of the Department of Pas de Calais. The following table gives the results of meteorological observations and chemical analyses of sample beets taken up every ten days during the season. The beets were sown April 5, 1879, averaging six plants to the square meter. They were of a poor variety, but of the kind ordinarily planted in that section; they were of a rosy color, and were planted a great distance apart in order that they might grow more rapidly.

The influence of sunshine is to be found by studying the fourth column of the sum total of daily average cloudiness at Arras, as resulting from twelve daily observations of the amount of cloudiness. The clearness of the sky, as given in the fourth column in percentages, is

the complement of the cloudiness and represents the relative duration of sunshine, but owing to the varying altitude of the sun can by itself alone give no idea of the intensity of the radiation received by the plant. To obtain thisst item and as no actinometric observations were made at Arras I give in the fifth column the results of observations at Montsouris, expressed in actinometric degrees.

The beets are reported to have sprouted very late and very unequally; this was due not to dryness, since the rain during March and April was in excess of its normal value, but was directly traceable to the low temperature, which was especially low in April.

The study of the development of sugar, week by week, as given in the last two columns of the above table shows that after September 9 the sugar crop increased slowly, became stationary, and then fluctuated very much as the weight of the leaves fluctuated. The rainfall had at that time become light and the development of the beet seemed to depend mostly on the temperature, so that it may be concluded that the beet ceases to increase in its quantity of sugar after the mean daily temperature falls below 13.1° C., and that there is no probable advantage in leaving the beets in the soil after that date, which in this case is September 29, 1879.

Marié-Davy points out that the actual increase per decade of the weight of the roots coincides with the increase of the rainfall and the temperature, but the proportion of sugar increases with the degree of radiation or total sunshine; the sunshine precedes the formation of sugar, since its action is slow and indirect, being through the assimilation that takes place within the leaves. It is therefore not an excess of water, but a deficiency of light and heat that causes rainy autumns and summers to give poor crops of sugar. Therefore, if during dry, clear, warm summers having large radiation, one could irrigate the fields properly one would realize the best conditions for a good crop. Therefore, every ray of sunshine that strikes the ground instead of the leaf is a loss to the formation of sugar and by helping to evaporate the moisture of the soil it also causes further great loss of sap to the plant. These conclusions agree with other experiments made by Pagnoul, who raised beets both in darkness and under a transparent bell glass, and again in the free air, and found the amount of sugar to increase with the strength of the sunshine.

The following table gives a general survey of the beet crops in Pas de Calais and the corresponding climatic data at Montsouris, which is about 90 miles south of Arras. The numbers given in the columns for quantity and quality of the crops are the estimates obtained from many planters and are recorded on the following scale: 1, very small