December 5, 1888.] 



Garden and Forest. 



489 



or Thorns are badly infested with bark-scale, and I have never 

 found a sure, practical cure for this pest except by cutting out 

 and burning the affected branches. This scale sometimes 

 takes such full possession of Lilacs as to compel the sacrifice 

 of the plants. In the case of Oaks, Beeches, Maples, and other 

 large trees which no longer need pruning to regulate their 

 growth, we can do little now, because the dead twigs and 

 limbs cannot be readily distinguished among the leaHess 

 branches. Cutting these away should be attended to when 

 the trees are in full leaf. Where it is necessary to remove 

 trees in order to open vistas, or for other purposes, the trees 

 should always be rooted out and not cut down ; the stumps 

 should never be left in the ground. Sometimes beautiful 

 views across the country can be seen over the tops of a 

 clump of trees. See to it that these trees are headed so low that 

 they do not intercept the view in any way. 



Pruning Shrubs. — In pruning shrubs we wish to preserve 

 symmetry of form and promote vigor, and at the same time 

 to secure a profusion of flowers. Shrubs that bloom in 

 spring on the previous year's wood should not now be pruned, 

 but rather after they have done blooming. These include 

 Daphne, Forsythia, Bush Honeysuckle, Japan Quince, Red 

 Bud, African Tamarisk, Corylopsis, Snowball and the early 

 blooming shrubby Spirseas. But such shrubs as bloom on the 

 current season's wood should be pruned now, and in some in- 

 stances quite severely back. These include Hydrangea panicu- 

 lata, Lcspediza bicolor, Alth;cas, Chinese Tamari.x, Stiiartias 

 and the like. Desinodiuiii pcnduliflorunt. Hydrangea radiata, 

 H. arbffrescens and Callicarpa purpurea seem to do best when 

 cut down close to the ground every year. 



Out-door Roses.— Roses should not be pruned at this time of 

 year except to cut in very long shoots. If pruning is deterred 

 till spring, we can cut back into the living wood, where the tips 

 have been winter-killed. Prairie Roses and other climl)iiig 

 varieties grow late in the fall, and these late-formed, sprawl- 

 ing growths should be cut away of tied up, for tidiness' sake. 

 If Rosa rugosa has outgrown its bounds, dig around it deeply 

 and remove all suckers; these make capital plants for a fresh 

 plantation. Also save the seed hips and sow them now in 

 boxes of sandy soil in a pit or greeir-house, or if you do not 

 want them yourself, exchange them with sonie neighbormg 

 florist for something that you may need. Although this Rose 

 reproduces itself freely from suckers and seed, it has never 

 become very plentiful. Moderately tender Roses, that need a 

 little protection in winter, can be bent down and covered with 

 earth. But, usually, a good mulching of littery manure over 

 the roots, or, if the shoots are bent down, over them as well, 

 will suffice. Evergreen branches laid over the bushes are 

 also a good protection. Tea Roses are safer if lifted now and 

 heeled into a frame or potted and phmged in a frame till 

 spring. Hermosa, Mrs. Degraw and some other Bourbons 

 get cut back to the snow line or near it every winter, but this 

 does not seem to injure them, and, when pruned in closely, 

 they throw up vigorous, free blooming shoots. IV. F. 



Glen Cove, New Yoi-|<. '' 



The Forest. 

 Do Forests Influence Rainfall ? 



IF I could find the place on the earth of which it was first 

 and emphatically said, " It never rains but it pours," I am 

 convinced that it would be a plain largely deficient in forest- 

 gTowth. For, if there be an influence upon moisture condi- 

 tions of the atmosphere exerted by forest areas — and such 

 areas must not only be of sufficient size, but also densely 

 enough covered to exert their proper influence upon tempera- 

 ture and moisture within and without — it consists, I believe, in 

 a more equal distribution of precipitation with reference to 

 space and time. 



In the end, what does it matter whether it is by increased 

 precipitation that the forest benefits the field, or whether the 

 same physiological effect is produced by increased relative 

 humidity in other ways, or by raising the water level and in- 

 creasing or advantageously disposing of the available water 

 supplies through favorable ground-water conditions or surface 

 channels ? 



As this question of forest influences is one which, to a 

 great extent, underlies the demand for national interest in the 

 forestry problem, it may be of advantage to review briefly the 

 methods which have been employed to solve the question. 

 Space will not here allow a critical consideration of the value 

 of each method, which may be done at some future time. 



As is natural, the first suggestion that a relation between 

 climate and forest-areas exists, came from general observation. 



History testifies that districts once surrounded by verdant 

 groves, with fertile soil and favorable climate, have become 

 inhospitable and desert wastes, with treeless mountain-sides, 

 and the conclusion follows that there is some connection 

 between the forests on one hand and fertility and genial 

 climate on the other. This method of proving the proposition, 

 which has been tlie most |)opular, and is still largely in vogue, 

 may be called the historico-statistical. Among the eminent 

 men who have used this method may be mentioned Du 

 Monceau, Reaumur, Button, Humboldt, Arndt. It is not to be 

 entirely discarded now, but its results must be adopted with 

 caution, for not only are the reports of the facts in many 

 cases dubious, but the inferences are not always reasonable. 



About the middle of this century, with the development of 

 physical, and especially meteorological science, a second 

 method was applied. "This method attempted, upon a theo- 

 retical basis, to discuss and reason out the assumed rela- 

 tion by employing the accumulated physical anti meteorologi- 

 cal data, which, scanty at first, has lately been considerably 

 increased. Among the prominent meteorologists to employ 

 this method fii'st was Becquerel. The results of this method 

 have lirought us considerably forward in the determina- 

 tion of the direction in which an influence would be possible, 

 or even probable ; and while it has not been able to either 

 prove or disprove satisfactorily the existence of this influence 

 nor advanced our knowledge of its degree and quality, it has 

 cleared the way for a more scientific consideration and inves- 

 tigation of the subject. 



The next step and method of demonstration employed was 

 the mathematical one, using numerical data which had either 

 accumulated independently of the question or were specially 

 provided for the purpose. We have here to distinguish two 

 methods, a wholesale and a retail one, if I may so express it, 

 or, more scientifically speaking, the one using large averages 

 and comparing data from extensive areas, though not specially 

 provided for this end ; the other comparing data obtained for 

 the pm'pose in limited localities by direct detail measurements 

 within and without forest areas. The latter method, which I 

 call the retail one, is the one now largely adopted by German 

 investigators. 



The first attempt to obtain, for the settlement of this ques- 

 tion, a series of exact, methodical observations, dates back to 

 the year 1864, when Dr. Ebermayer, Professorat the University 

 of Munich, constructed the necessary apparatus, and with the 

 aid of the Bavarian Government and Forest Administration 

 established in 1866 the first three double stations, where a set 

 of meteorological instruments were observed within a forest 

 area, and another set simultaneously in a field. In the follow- 

 ing year the number of the double stations was increased to 

 six. In 1869 Switzerland followed with three stations ; in 1870 

 Italy established a station, and in 1874 to 1877 Prussia entered 

 upon this field of work, having now sixteen stations in connec- 

 tion with the forest experiment stations ; and to-day quite a 

 numljer of double stations are collecting data in all p.arts of 

 the country. 



The points of observation at the Prussian stations are chosen 

 200 metres (about 664.5 feet) distant from the edge of theforest 

 within and without. An enormous amount of material has 

 accumulated, but as yet has not been summarized or turned 

 to account. It is difficult to see how anything else can be 

 demonstrated by it than what is already known — namely, that 

 the meteorological conditions within the forest are dift'erent 

 from those prevailing without. Whether the forest conditions 

 are communicated to the open field, and to what degree, if at 

 all, can certainly not be proved by the data obtained. By 

 establishing points of observation in the field at varying dis- 

 tances, it might have been possible to demonstrate the presence 

 or absence of climatic interaction between forest and field. 



In the wholesale methods, which use data obtained over 

 large areas independently of the special objects of this inves- 

 tigation, we may again discern two ways of handling them : 

 the one comparing the data found during various periods at 

 the same stations and liringing them in relation with forest 

 conditions existing at the various periods; the othercomparing 

 data obtained simultaneously from stations situated differently 

 as regards other climatic influences. The first method has 

 been employed by Mr. Gannet and Mr. Harrington. Mr. Gan- 

 net endeavors to estalilish by a combination of data that 

 neither for Ohio, which has been largely cleared, nor for New 

 England, which is said to have largely increased its forest 

 area, nor for the Prairie States, which contain more timber in 

 recent times than formerly, can a noticeable dilTerence in 

 rainfall be demonstrated. In fact, however, he only proves 

 that his method leads to no certain result for lack of adequate 

 data to work upon. Mr, Harrington's method fails to be con- 



