species studied by him, the age of seeds varied from 125 to 135 years, and that a few of the old seeds germinated, notably those belonging to the pulse, water lily, and mallow families. In this connection the study made by Ewart on the Longevity of Seeds is of interest. Thus in some of the common Indian mallow or butter-print, 6 per cent of seed germinated after the lapse of 57 years, the common shoo-fly after 57 years, the

I,

Fig. 32. Seeds of some grasses. quack grass; 2, 3, orchard grass; 4, 5, meadow fescue; 6, English rye grass. (Hillman.)

seed of white sweet clover showed germination after 77 years, and 52 per cent after 44 years, chicory after IO years. In this connection the experiments conducted by Dr. W. J. Beal are also of interest.

A number of years ago he made an interesting experiment on the prolonged vitality of the seeds of some common plants, testing for vitality at different periods of 5, 10, 15, 20, and 25 years. The seeds were placed in sand in bottles slanting downward so that water could not enter, and buried in the soil, 20 inches below the surface. The seeds were all grown the year they were buried. The * indicates that the seed germinated.

The following table indicates the vitality at these periods:

Prof. L. R. Waldron has shown that in the case of

some weeds there is delayed germination. He experimented with seven different kinds of weed seeds, as follows: Great ragweed or kinghead, wild oats, green foxtail, shepherd's purse, wild mustard, wild buckwheat and French weed.

In the fall of 1889 there were planted in a seed bed out of doors at depths of 1, 2, 3, 4, 5, 7, and 10 inches the above seeds. During the same fall shepherd's purse produced a few plants at depths of 1 and 2 inches; French weed, 25 plants at 1 inch; wild mustard, many plants at I, 2, and 3 inches; wild oats, several plants at 1, 2, and 3 inches.

During 1900 the seeds continued to germinate, and it was true of them that the small weed seed did not come up through 2 inches of soil; no seeds buried below 3. inches germinated except kinghead or great ragweed and wild oats, which came up through 5 inches of soil. The writer, Miss King and Mr. Fawcett have conducted some experiments on delayed germination of weed seeds.

It was found that the seeds of different species showed great differences in germination; in general the results

of 1902 and 1903 indicated that stratification in sand and freezing were favorable to germination.

Weed

Milkweed (A. syriaca).
Western ragweed

Lamb's quarter.

Cocklebur

Per cent of germination

Before stratification After stratification

Mr. H. S. Fawcett made a study of 52 species of weed seeds.

12

18

88

25

The samples were gathered after maturity in September, October and November of 1904. The seeds were threshed out, and placed in paper envelopes. Fifty seeds of each kind were placed in November in sand in boxes kept during the winter in the greenhouse under uniform conditions. Similar plantings were made each month until May. The sand in all these germination trays was kept moist.

In addition a large number of weed samples was placed out of doors for the purpose of subjecting them to freezing and thawing. The seeds were placed in sacks in a wooden box and covered with a thin layer of sand; the box was then deposited about one foot below the surface, where it remained till spring. The general effect of freezing and thawing was to increase the percentage of germination and lessen the dormant period, especially in case of seeds with hard coats; in all cases the dormant period of hard-coated seeds seems to be greater than that of thin-coated seeds. The following examples illustrate these differences:

The dormant period of common pigweed (Amarantus retroflexus) was nine and one-third days when kept in packages in a dry room, and six and one-third days after having wintered out of doors; its germination was increased from 40 to 50 per cent. In common pigeon grass the average dormant period was lessened from eleven to seven and one-quarter days; the percentage of germina

tion increased from 341⁄2 to 38 per cent. In wild rye the dormant period was lessened from nine to five days; the germination increased from 22 to 48 per cent.

It was found that the longest dormant period was found in those seeds which have the hardest and thickest seed-coats. Some of the ragweed seeds had a dormant period of 152 days, while some seeds of barnyard grass had a dormant period of 178 days.

The highest average percentage of germination was observed in common mustard, which was 100 per cent, and for the six tests, 90.3 per cent.

It has long been known that many seeds refuse to germinate until they have passed a period of rest. Nobbe and Hänlein made a study of the seeds of 31 species of weedy plants, continuing the experiment for 1,173 days. A number of these weeds showed germination after a lapse of 1,173 days; among them were Campanula persicifolia, silvery cinquefoil (Potentilla argentea), mouse-tail (Myosurus minimus), hoary plantain (Plantago media), and field penny-cress (Thlaspi arvense).

It would seem that freezing is decidedly favorable for the germination of many weed seeds. Of 65 species planted November, 1908, four species failed to germinate; of 60 species planted in December, 48 failed; of 63 species planted in January, 44 failed; of the March planting of 62 species, 27 failed; of 59 species planted in April, 46 failed; of the seeds exposed to the weather, 64 species, 24 failed to germinate.

With the closest care and observation there was lack of constancy in the germination, resulting in marked irregularities not to be attributed to known causes, due possibly to inherent qualities.

Germinability of Weed Seeds After Having Been Kept in Compost and Manure.-There is a widespread impression that germinable weed seeds may be scattered with

manure. This is true so far as seeds which get into manure not thoroughly composted are concerned.

Mr. E. I. Oswald made a series of experiments to ascertain the vitality of seeds when placed in manure under different conditions. Three kinds of manure were usedhorse manure, cow manure, and horse and cow manure mixed in equal parts. The manure was placed in separate piles, and seeds inclosed in gauze bags were placed in the manure and allowed to remain for varying periods of time, in some cases 60 days, in others six months. This lot contained 52 different kinds of seeds. After the required length of time, the seeds were planted in the greenhouse, and it was found that they had become thoroughly rotted and their vitality destroyed. In order to cover the conditions usually followed by dairymen and gardeners, where the manure remains but a short time, another set of experiments was started in the fall. The seeds remained in the pile only one month and were then planted as in the previous experiment. The results of the experiments entirely confirmed the previous ones, except in the cases of the seeds of ribgrass, horse nettle, common plantain, large ragweed, bitter dock, and mallow, which were still firm.

Early in the summer of 1908 we placed 31 different seeds of weeds and seeds of a few cultivated plants in gauze bags and left them in horse manure for five weeks. The seeds were placed in this manure on May 10 and were removed June 19. The seeds were then germinated in the greenhouse. Only a small percentage of the following weed seeds were capable of germination.

We also conducted some experiments in feeding various weed seeds to animals, among them sour dock, quack grass, and sunflower; very little seed was capable of germination after it had passed the digestive tract of the animals. It is well-known, however, that hard-coated seeds do pass the digestive tract uninjured. Numerous