on a perforated cork floating in water in a jar so that the roots extend into the water. Their growth may be observed for several weeks. =21.= _Effect of darkness on seeds and seedlings._ A box may be placed mouth downward over a smaller box in which seedlings are growing. The empty box should rest on half-inch blocks to allow air to reach the seedlings. Note any effects on the seedlings of this cutting off of the light. Another box of seedlings not so covered may be used as a check. Lay a plank on green grass and after a week note the change that takes place beneath it. =22.= _Seedling of pine._ Plant pine seeds. Notice how they emerge. Do the cotyledons stay in the ground? How many cotyledons have they? When do the cotyledons get free from the seed-coat? What is the last part of the cotyledon to become free? Where is the growing point or plumule? How many leaves appear at once? Does the new pine cone grow on old wood or on wood formed the same spring with the cone? Can you always find partly grown cones on pine trees in winter? Are pine cones when mature on two-year-old wood? How long do cones stay on a tree after the seeds have fallen out? What is the advantage of the seeds falling before the cones? =23.= _Home experiments._ If desired, nearly all of the fore-going experiments may be tried at home. The pupil can thus make the drawings for the notebook at home. A daily record of measurements of the change in size of the various parts of the seedling should also be made. =24.= _Seed-testing._—It is important that one know before planting whether seeds are good, or able to grow. A simple seed-tester may be made of two plates, one inverted over the other (Fig. 31). The lower plate is nearly filled with clean sand, which is covered with cheese cloth or blotting paper on which the seeds are placed. Canton flannel is sometimes used in place of sand and blotting paper. The seeds are then covered with another blotter or piece of cloth, and water is applied until the sand and papers are saturated. Cover with the second plate. Set the plates where they will have about the temperature that the given seeds would require out of doors, or perhaps a slightly higher temperature. Place 100 or more grains of clover, corn, wheat, oats, rye, rice, buckwheat, or other seeds in the tester, and keep record of the number that sprout. The result will give a percentage measure of the ability of the seeds to grow. Note whether all the seeds sprout with equal vigour and rapidity. Most seeds will sprout in a week or less. Usually such a tester must have fresh sand and paper after each test, for mould fungi are likely to breed in it. If canton flannel is used, it may be boiled. If possible, the seeds should not touch one another. [Illustration: FIG. 31.—A HOME-MADE SEED-TESTER.] NOTE TO TEACHER.—With the study of germination, the pupil will need to begin dissecting.