THE LIVING TREE 109 shoots are referred to as lammas shoots; it is safe to say that we do not understand the underlying causes which promote the formation of these lammas shoots: high rainfall, drought, mutilation, previous defoliation and heavy manuring are all said to favour their formation. Sometimes the leaves of lammas shoots differ from those of the rest of the tree : thus in Oak their leaves may be more deeply incised while in one variety (Quercus petraea mespilifolia (Wallr.) Schwarz) in which the leaves are entire and pointed at the tip, the leaves of the lammas shoots are of the normal form for the species. It would also be of supreme interest to know what stimulus in some trees causes the leaf primordia to develop into scale leaves in a period of the year which appears to be most favourable to active growth, and then, more or less suddenly, to revert to the formation of normal leaves ready for the following year. On this matter we are ignorant. So far we have tacitly assumed that the extension growth of a shoot is continued by its terminal bud: this is sometimes true, as in Spruce, where a terminal bud persists throughout life, but in other trees like Lime and Elm the terminal bud dies and its place is taken by the topmost lateral bud. In the former case, of course, the branch is directly formed from a single bud, increasing in length year after year by the activity of the meri- stem of this bud : the branch forms a monopodium and this type of growth is spoken of as monopodial. In the latter type of growth the branch is different, for each successive increment is, in fact, a lateral outgrowth of its predecessor : it is a sympodium and growth of this type is termed sympodial. As we shall see, these two types of growth directly affect the ultimate shape of the tree. It has been found possible to prolong the life of the terminal bud by removing the upper leaves of a shoot or by removing the more distally placed buds, while abnormal rainfall has a similar effect : but sooner or later the result is always the same and we must regard this abbreviated life of the terminal bud in some trees as an inherent factor in the make-up of the tree. It is well known that all the shoots in some trees are not similar. Thus most varieties of Apples, and Pears have shoots which bear the leaves and dwarf, short or spur shoots on which the fruit is borne. Sometimes the difference between the two types is striking. In the Pines the foliage leaves are borne in clusters of two, three or five on a very short stem and these dwarf shoots arise on long shoots which themselves produce no foliage leaves, but merely scale leaves. In the Cedars, on the contrary, both the long and short shoots bear needles. A short shoot may suddenly ex- pand into a long shoot (regeneration shoot) or it may die after a