integuments; finally, the fronds of certain ferns have a green color, even when they grow in complete darkness. With regard to the seeds of Acer, Astragalus, Celtis, and Raphanus, it has been shown by J. Böhm that when they germinate in darkness they do not acquire any green color; Flahault (1879) has obtained the same result for the seeds of the Viola tricolor, the Acer pseudoplatanus, and the Geranium lucidum. Similarly as to the other seeds above enumerated the studies of Sachs and Flahault render it probable that in most cases there was stored up in the seed certain reserve nutrition, which reserve, originally formed under the action of light, can subsequently in the act of germination temporarily replace the further direct action of light. It would thus seem that in no case can dark heat truly replace the action of sunlight.

On the other hand, light can replace heat in the process of vegetation. This was first shown by De Candolle, and a striking illustration is quoted by Moleschott (1856), who shows that by the influence of light during the resplendent nights of the polar regions the harvests ripen in a short time, while many days of our autumn heats in lower latitudes scarcely suffice. It is the quantity of light and the quality of the radiations that these plants receive that enable certain cereals, such as barley and oats, to be cultivated as far north. as 70° of latitude. The observations of Schleiden on the potato, of De Candolle on the radiola, and of Haberlandt (1866) on oats, show that there exist decided differences in the quantities of heat necessary to the development of different species of vegetables under different latitudes, and that the most important cause of these differences is the quantity of light which these plants receive. De Candolle, in his botanical geography, says the effect of light is shown in the northern limits of certain species; thus the radiola is perfected by a total supply of heat represented by 2,225 day-degrees in the Orkneys at 59° north, but by a total of 1,990 day-degrees at Drontheim, latitude north 63° 25'; the difference (235) corresponds to the fact that the longest day is 14 hours longer at Drontheim than in the Orkneys, which increased sunlight enables the plant to complete its growth better under the same temperature.

Wheat furnishes a still more striking example. It begins to vegetate when the temperature in the shade is about 6° C., and observation has shown that it requires the following day-degrees to ripen: At Paris in 138 days, total shade temperature 1,970° C.; at Orange, 117 days, total shade temperature 1,601° C.; at Upsala, 122 days, total shade temperature 1,546° C.; at Lynden (North Cape), 72 days, total shade temperature 675° C. Or, if we use, not the shade temperatures, but those of a thermometer exposed to the full sunshine, as is done by Gasparin, then the above figures become at Orange, 2.468 day-degrees; Paris, 2,433 day-degrees; Lynden, 1,582 day-degrees.

These remarks of De Candolle with reference to germination are equally applicable to the whole period of growth of the piant.

As to the method of calculating the sum total of temperatures De Candolle found that it may be conducted in two ways, either by adding together all the mean daily temperatures above 0° C. or by omitting the useless degrees and adding all the others. This last method would seem to be the most logical, but can rarely be employed, owing to our ignorance of that minimum temperature below which all must be omitted. On the other hand, if we consider that a plant which vegetates between 10° C. and 30° C. has a maximum at 20° C., and if we seek the coefficients of growth corresponding to each successive degree of temperature, we find, as Boussingault has shown, that these coefficients vary for each degree as we depart above or below the temperature most favorable to vegetation.

Similarly De Candolle (1865) has shown that near the minimum and near the maximum temperatures the rate of germination is more difficult, and therefore slower, than at the intermediate or best temperatures; consequently, both in germination and in subsequent vegetation, it is necessary to recognize the fact that calculations of the sums of heat in connection with the study of the geographical distribution of plants are complicated with hypotheses and many sources of error.

Schuebeler (1862) shows that cultivated plants in northern countries have more highly colored flowers, larger and greener leaves, and larger seeds, which are more highly colored and richer in essential oils, than those of southern regions. Bonnier and Flahault (1878) have shown the same facts for uncultivated plants. Both these authors attribute this result to the prolonged action of sunlight, and the latter shows that the variations are exactly proportional to the duration of sunlight. In Flahault's more recent observations he shows that there must necessarily exist a relation between the quantity of carbonic acid decomposed and the quantity of carbonaceous matters formed by the plant, and that in general the sunlight has a very remarkable influence on vegetation since it compensates in a large measure for the deficiency of temperature.

It is, furthermore, to this influence of light that Pauchon attributes the singular fact that plants cultivated in high latitudes are endowed with a vegetating power greater than that of southern countries, so that when transported to the south their seeds ripen sooner than those of the southern plants. This subject has been especially studied by Tisserand in his memoir on vegetation in high latitudes, as cited by Grandeau in his work on nutrition of plants. According to Tisserand a plant behaves in northern latitudes as a more highly perfected machine and one that performs better than southern plants. In regions where it has neither time nor heat it gains in activity and

in the speed with which it perfects its own growth. It seems to Pauchon that we may properly interpret this phenomenon if we admit that a seed transported from the north to the south finds itself in climatic conditions more favorable to the development of the embryo which it contains and of the plant which is to follow. What the action of light loses in duration in proportion as we move toward the equator it gains in intensity. It may be that the cause of this increased activity is due to the larger size of the northern seeds or to their greater richness in the essential oils. Pauchon thinks that the embryo of such a seed should not be compared to a more perfect machine; it is rather an identical machine, but better nourished by the reserve of combustible and nutritive material in the perisperm. Possibly the abundance of essential oils contained within the seed contributes to furnish to the embryo in northern countries the materials for the oxidation that is necessary in order to maintain its temperature during germination and to struggle against the severity of the climate.

Tisserand (1876) has shown that the rye cultivated in northern Norway has not the same chemical composition as that of France and Algeria, and that in general, as we go northward, or as we rise above the level of the sea, or as the temperature lowers without diminishing the quantity of light, we see the starch in the grain increase relatively to the nitrogenous components. Wheat grown at Lynden (North Cape) has a smaller proportion of gluten than the wheat of France, and the latter less than the wheat of Africa. On the other hand, barley raised at Alten, on being sown at Vincennes on the 7th of April by Tisserand, was ripe on the 18th of June, or thirty-seven days in advance of French barley, so that in order to mature it required a sum total of heat far less than the French barley. The reverse is true when southern grains are carried north and sown in colder climates. Therefore, as Marie-Davy has remarked, plants become acclimated more or less rapidly according to their own nature and the extent of the climatic variations that are imposed upon them; the climate produces in them a functional change which corresponds to an organic change the nature of which often escapes our observation. It is therefore not necessary that each phase of vegetation should correspond to a constant sum of heat in very different climates. That which it is important for us to know is what are the limits between which this sum total can vary, for the same species of plant under different climates.

The general fact that the quantity of nitrogen contained in the seeds increases as we approach the warmer climates leads to the hypothesis that the formation of albuminous reserves within the seed takes place in proportion to the temperature, and that the formation of starch and other reserves takes place in proportion to the duration

of the light and the action of the chlorophyl of the leaves. As we pass from the pole to the equator the luminous intensity of the sunlight increases from a hundred to a thousand, but its duration diminishes during the growing season from a hundred at the poles to fifty at the equator. Among the special investigations into the action of sunlight we note that of Timiriazeff (1877), who has shown that a very intense light, after traversing a certain thickness of green-leaf cells, has no further action on the phenomena of the reduction or decomposition of carbonic-acid gas; in other words, it acts the same as darkness would do. On the other hand, Paul Bert, by exposing plants to the action of light which had been sifted through a solution of chlorophyl, invariably found that the development of the green matter of the leaf was completely arrested; inversely, he found the green matter produced to its normal amount when the plant received only light that had been filtered through a solution of iodine in bisulphide of carbon, which solution, as we know, cuts off all visible rays, but allows the red and infra-red to pass through with great freedom. This would seem to demonstrate that chlorophyl is formed by the action of the red portion of the spectrum.

As to the effect of light on the germination of seeds, Pauchon (1880) gives a critical summary of views by different authors, from which we condense the following:

Miesse (1775), from observation on the Camelina (Myagrum sativum), concludes that the seeds grow in darkness the same as in full daylight, and that light does not seem to influence this stage of vegetation.

Sénébier (1782), from observations on seeds of lettuce and beans, some of which were exposed to the full sunlight, others to sunlight after filtering through a thickness of water, others in the dark, and others in red, violet, and yellow light, respectively, reached the conclusion that light was injurious; but his results were not decisive, because of his neglect to observe exactly the temperatures under different conditions.

Ingenhousz (1787) exposed an equal number of mustard seeds in places receiving different amounts of light. He himself concluded. that the light of the sun is as injurious to vegetation at the beginning of its life as it is advantageous to vegetation in the fullness of its life. But a more careful consideration of Ingenhousz's experiments shows that the moisture and the temperature in his several localities varied so much as to prevent any serious conclusion as to the action of light itself.

Bertholon (1789), in an article on the effect of electricity, shows

that up to that time it had not been proven whether the germination of the seeds was affected by light or by humidity. His own experiments convinced him that the latter was more important.

Sénébier (1800) made additional experiments on peas and beans, sowing them in sponges, which were kept equally moist, all inclosed under glass covers, so that no evaporation could take place. Some were exposed to sunlight and some were kept in the dark, but those which were in the dark germinated much sooner than those in the light. But in such experiments as these the sources of error are numerous, and the fact that there was no renewal of the air under these covers was especially unfavorable to germination. In fact, Leclerc (1875) has shown that under the influence of mercurial vapor, as it existed in Sénébier's experiments, a large portion of seeds are killed, so that with our present knowledge we can not accept Sénébier's conclusions.

Lefébure (1800), having finally accepted the conclusions of Sénébier and Ingenhousz relative to the injurious influence of light on germination, repeated the experiments, but also observed the temperatures more carefully, and in addition sought to determine the effect of light that had passed through plates of white, green, black, red, and blue glass; but he added little to our knowledge, although he himself concluded that the seeds under white glass were retarded. Th. de Saussure (1804) endeavored to ascertain whether the influence observed by others was due to light or heat, and he concluded that nothing demonstrates that light has an injurious influence independent of the heat that accompanies it.

Keith (1816) made no observations himself, but controverted the conclusions of De Saussure.

Boitard (1829) sowed the auricula seeds in three flower pots, but the conditions as to temperature and moisture are not sufficiently known to justify us in drawing any conclusion.

A. P. de Candolle (1832) says:

I do not deny that darkness may be useful in germination, but I do deny that it is necessary to think that light has no action on germination. Analogy indicates this, theory confirms it, and experience demonstrates it.

According to De Candolle, 'light favors the decomposition of carbonic acid, but germination demands the formation of carbonic acid; therefore darkness will favor germination. This theory thus enunciated by De Candolle has been accepted by many authors without proper experimental basis.

Ch. Morren (1832) experimented upon water cresses grown under different colored glasses. He concluded that as darkness favored germination, so the individual colors of the spectrum, acting each by itself, have a special influence that favors germination in such a way that