In order to examine the structure of the pollen-grains properly, they should be examined under various circumstances--some dry, others placed in water to which a little sugar has been added, others in oil, and it will often be found useful to try the effect of different acids upon them. Fig. 20 is the pollen of the common violet, and is easily recognisable by its peculiar shape and markings. Fig. 21 is the pollen of the musk-plant, and is notable for the curious mode in which its surface is belted with wide and deep bands, running spirally round the circumference. Fig. 22 exhibits the pollen of the apple, and Fig. 23 affords a very curious example of the raised markings upon the surface of the dandelion pollen. In Fig. 24 there are also some very wonderful markings, but they are disposed after a different fashion, forming a sort of network upon the surface, and leaving several large free spaces between the meshes. The pollen of the lily is shown in Fig. 25, and is a good example of a pollen-grain covered with the minute dottings which have already been described. Figs. 26 and 27 show two varieties of compound pollen, found in two species of heath. These compound pollen-grains are not of unfrequent occurrence, and are accounted for in the following manner. The pollen is formed in certain cavities within the anthers, by means of the continual subdivision of the “parent-cells” from which it is developed. In many cases the form of the grain is clearly owing to the direction in which these cells have divided, but there is no great certainty on this subject. It will be seen, therefore, that if the process of subdivision be suddenly arrested, the grains will be found adhering to each other in groups of greater or smaller size, according to the character of the species and the amount of subdivision that has taken place. The reader must, however, bear in mind that the whole subject is as yet rather obscure, and that further discovery may throw doubt on many theories which at present are accepted as established.