made the present summer in the Hudson River valley. The alteration, however, is not in the direction of the statement made by Dr. Lintner, but in the opposite direction. These observations were greatly facilitated by the condition of affairs which exists in the city of Kingston, Ulster County, N. Y. There the plan has been to plant trees of different varieties in an alternating manner. Instead of a solid row of elms, there will be a row of elms alternating with sugar maple. The trees being planted rather closely and the elms growing to much greater height and frequently reaching over the maples, afforded an opportunity for testing this matter which does not usually exist along streets in cities. My attention was first directed to the extraordinary numbers in which the insects were transforming high up on the trunk. The trees were mainly old, the bark being rougher than common, and a close green moss had grown over large sections of the trunk. Embedded in this green moss and in the crevices in the bark were thousands upon thousands of the yellow pupa. These observations were made July 10. Climbing upon a tall fence and examining the crotches of one of the largest of these trees, the pupa were found in such numbers that they could have been scraped up by the handful. Examination of the surface of the ground at the base of the tree showed perhaps one-fifth as many pupa as were to be found upon the trunk of the tree high up and in the main crotches. This, in itself, would indicate the almost total futility of any wash applied to the lower part of the trunk and upon the ground about the base of the tree. Further observations of equal interest and which make the case still stronger were soon made. The writer's attention was attracted to the yellow pupa of the elm leaf-beetle in the crevices in the bark of a maple tree standing between two elms. He ascertained without doubt that it belonged to this species, and upon search found many others.. The branches of the elms overhung the maple, and the inference was plain that the larvæ dropped from the terminal branches of the elms,. descended the maples, and transformed at the first convenient point.. Many such cases were observed. The most interesting case, however, was still to come. Upon the grounds of one of the wealthy citizens on one of the principal avenues, a low-growing horse chestnut tree, perhaps 25 years old, with the usual shaggy bark, was noticed. At almost equal distances from this horse chestnut stood three large old elms, each about 30 feet from the chestnut. The terminal branches of each of these elms interlaced and overhung the top of the horse chestnut tree at a distance of perhaps 20 feet above its tip. An examination of the trunk of this horse chestnut showed the pupa of the elm leaf-beetle by thousands and thousands, under the loose bark which afforded particularly appropriate places for pupation. Few, if any, had reached the ground. The trunk of each of the elms in question was examined and the ground around the base of each as well, with the result that upon the trunk of the horse-chestnut tree were more than twice as

many (at an estimate) of the pupa as were to be found upon any one of the elms. It must be remembered that but a small portion of the circumferential growth of the top of each elm overhung the horsechestnut. Perhaps the portions of the three tree tops so overhanging would constitute a third of the arc of a circle. It at once becomes evident that in the case of these three elms nothing like 60 per cent of the larvæ descended the trunk of the tree upon which they were raised, and, in fact, that only about one-half of this percentage was correct in this case. Fully 70 per cent must have dropped from the limbs.

I do not pretend to say that this instance was not unusual, but that it demonstrates the vast superiority of spraying over all other means of fighting the elm leaf beetle more forcibly than it has previously been shown can not be doubted.

Mr. Marlatt read the following paper:

INSECTICIDE SOAPS.

By C. L. MARLATT, Washington, D. C.

The decided insecticide value of the so-called whale-oil (more prop erly fish-oil) soaps against scale insects particularly has been fully demonstrated in the last few years in the work against the San Jose scale, and has fully substantiated Professor Comstock's early recommendation of this means of controlling scale insect pests. The merit of these soaps is not only in their effectiveness as insect destroyers, but from their being entirely without injurious effect on the treated plant. In this respect they are perfectly safe in the hands of any person, in contradistinction to all oily washes, which are very liable to be injurious in greater or less degree, although the injury may be insignificant, or perhaps not apparent immediately, or during the first season. As pointed out at a previous meeting of this association, and also in the bulletin on the San Jose scale issued by the Division of Entomology, United States Department of Agriculture, the use of soaps is attended with certain hitherto unavoidable difficulties. It seems impossible to secure with any degree of certainty soaps which may be depended upon to behave in the same manner when dissolved for application as a paint to trunks of trees, or in a spray for application to the entire plant. During the past two years we have been making a strong effort by correspondence and personal interviews with leading soap makers of Washington, Philadelphia, and New York to get a suitable soap, uniform in composition and behavior, which could be relied upon, but so far with indifferent success. The leading Washington soap maker after one trial refused to bother further with fish-oil soaps, on account of the disagreeable nature of the oil and the vile odors engendered while working with it. Some half dozen samples each have been

received for experimental purposes from William H. Pinner, of New York City, and James Good, of Philadelphia, and we have had a number of proprietary insecticide soaps, testing also some common brands of washing soaps for purposes of comparison.

Assuming that one has received a brand of fish-oil soap from an honest manufacturer, to be satisfactory for insecticide use it must when dissolved at the desired rate, say 2 pounds to the gallon of water, remain a liquid capable of being sprayed with an ordinary nozzle at an ordinary temperature. The presence of water in the soap, which is a most variable quantity, should also be approximately known to enable one to determine the amount of soap to use to make a correspondence with the standard of strength.

The experiments with soap have been in the direction of testing, therefore, first, the physical properties, viz, when dissolved if it would remain in liquid form; second, the percentage of water in the soap, and third, the constituent elements of the soap, so far as could be determined by chemical and other tests.

The first proposition, whether the soap would remain liquid when dissolved at the strength employed, may be determined by a very simple test and one which should be invariably given any soap before it is accepted for spraying operations. It consists in simply dissolving a small quantity of the soap at the desired rate and allowing it to cool. Some 18 soaps were subjected to this test, dissolving them at the rate of 2 pounds to the gallon. Nine of these stood the test satisfactorily, remaining liquid even when subjected to low temperatures in a refrig erator, and nine solidified or became gelatinous and tenacious at high summer temperatures. All but two of these soaps were especially manufactured for insecticide purposes. Among the soaps sustaining satisfactorily the solution test were the old whale-oil soap (Leggett Bros.); three of the Pinner brands, including his concentrated chemical soap; a light-colored fish oil soap made by Good, and a potash soap made by the same manufacturer, at my earnest request several times repeated. The other soaps proving satisfactory in this respect are firtree oil soap (Stott's), tobacco soap (Rose & Co.), and a soap made at the United States Department of Agriculture with potash. Some of these soaps were very hard, as the Leggett whale oil, Pinner's concentrated chemical, and the light-colored soap made by Good. The rest were not so hard, and Good's potash soap was almost a soft soap, although the water percentage in it was not very much greater than some of the hard soaps.

Among the soaps resulting in a gelatinous or nonfluid mass with the solution test were three of the lots of alleged fish oil soap submitted by Good at different dates; the Pinner chemical tobacco soap; another tobacco soap, manufacturer unknown, and two washing soaps, the "Star" and "Babbitt." One of the soaps made by Good (No. 6) was dissolved at the rate of 8 pounds to the gallon and dried down to a

thoroughly firm soap, approximately one half of the added water being lost. This product was afterwards shown to contain 48 per cent of water, and yet on diluting it at the same rate as the other soaps it solidified on cooling, showing a very strong tendency in this direction. The opposite results obtained with different products of both the Good and Pinner factories indicate the total unreliability of the insecticide soaps at present obtainable and the necessity always of a preliminary test before using, or, better, before buying a soap in quantity, as well as the very urgent need of some action which will put within our reach a reliable article.

The second examination, viz, the percentage of water contained in soaps, was determined for me for eight brands by Professor Patrick, of the Chemical Division, United States Department of Agriculture. The samples submitted to Professor Patrick included five of the soaps which remained liquid at 2 pounds to the gallon and three that became solid. Eliminating the soap referred to above which contained 48 per cent of water, although still a soap hard enough to cut and handle easily, the per cent of water in the other soaps ranged from 9.40 in the Leggett whale oil to 28.30 in the potash soap made by Good.

The percentages of water in the soaps which were liquid in the solution test are as follows:

The percentages of water in the soaps which solidified after the solution test are as follows:

Tobacco chemical (Pinner)..

Fish oil in bulk (Good)

21.80

22.90

Fish oil in bulk (Good), treated as above described with water

48.00

The last sample shows that the per cent of water in soap has very little to do with its remaining fluid when employed at the strength indicated, this sample having 48 per cent water solidified, whereas the whale-oil soap, having the least water (9.40), remained liquid. It indicates also very clearly that much money may be unwittingly spent for water in the purchase of soaps unless they be previously tested in this direction.

It is well known that potash tends to make a soft soap and soda a hard soap in combination with oleic acid, which exists in most fats and enters largely into the composition of soaps. The presence of stearin, however, will have a marked influence on the quality of the soap, making it harder with potash and very hard with soda. Soaps, therefore, made with fats containing much stearin, such as mutton and beef tallow, and also some of the vegetable fats, will be very hard. The solu bility of soaps in water, in the above combinations with potash or soda and oleates or stearates, ranges from a soap made with potash and

oleates of 1 to 4 of water to a soap made with soda and stearates of above 1 to 40 of water.

The first explanation that suggested itself of the difference in the behavior of the various insecticide soaps tested was the very natural one, under the supposition that they were fish-oil products, that it was due to the manufacture of the soaps in some instances with potash and in others with soda, the former being liquid in solution and the latter solidifying. The chemical examination of these soaps, however, failed to support this theory and showed that in every case the soaps examined had been made with soda with the single exception of the one made at my special request by Good with potash. Many of these soda soaps, including the Leggett whale oil, were thoroughly satisfactory with the solution test. The correct explanation undoubtedly is that - the solidifying of certain brands of soaps is due to the admixture with fish oil, if the latter be used at all in these instances, of more or less waste beef or mutton tallow. It may be possible that there are other reasons for this condition which have not been discovered, but the above seems to be the important one. The result of the chemical examination and physical test seems to indicate one thing at least, that whether made with soda or potash, if fish oil, which is almost wholly olein, be used the product will stand satisfactorily the solution test. If, therefore, we can have a guaranty that a soap is altogether fish oil and properly made, the kind of lye employed is not absolutely essential, although the potash product is the more desirable.

The statement made in a former communication on the subject of soaps that very little water would enter into the composition of a wellmade article is confirmed by these analyses, made by Professor Patrick, which show an average of less than 20 per cent water, and with the better brands between 10 and 15 per cent, the balance being the lye and fats.

The result of the investigation of this subject indicates that a fish oil or other oily soap should be invariably demanded and the desirability of giving a sample of the soap secured from a manufacturer a solution test before purchasing on a large scale, or preferably demanding a product which will stand such test. The determination of water constituent of soaps is difficult and can ordinarily only be made satisfactorily by chemists, and while desirable for accurate work may perhaps be ignored. if the soap is a good, firm one and stands the solution test properly. It would seem also advisable to have the station chemists of the different States empowered to inspect and certify the qualities of any soap manufactured and sold for insecticide purposes.

In discussion, Mr. Fernald referred to the statement of Mr. Marlatt that poisoned foliage seemed to be repugnant to larvæ, and stated that the experience of the gypsy moth committee did not seem to bear out this view. He called on Mr. Kirkland to give the facts as brought out in the work of the gypsy moth committee.