LABORATOliY AND GREENHOUSE STUDIES. 63 



which render the inference to be derived from these experiments not 

 at all so conclusive as it would appear. It is important to inquire 

 whether many, if not all, of the trees which eventually became 

 diseased, may not have been infected before the spraying began. 

 As has already been shown (p. 54), the infection may be present in 

 the crown of a tree and not visible externally. And, further, the 

 solution applied did not adhere evenly, but tended to run together 

 and cohere in patches. When the spray was applied to the coconuts 

 the solution would invariably run down to the under side, where it 

 would diy on in drops. These facts were not particularly prejudicial 

 to the use of the spray, for it was intended and desired to soak 

 thoroughly only the strainer and the tissues at the base of the leaves. 

 The foregoing data show that the spraying did no good in preventing 

 the spread or development of the disease, but there is no definite 

 means of determining the number of insects destroyed. From 

 examining many of the trees in the crown during the course of the 

 year it seemed evident to the writer that the insects were materially 

 reduced, as there were practically none in the crown during the fall, 

 with the exception of ants, of which there were as many as ever. 



LABORATORY AND GREENHOUSE STUDIES OF THE DISEASE. 



It has been shown in earlier paragraphs that bacterial organisms 

 are able to produce a rotted condition in the heart of the crown of the 

 coconut tree identical with the condition in the typical bud-rot 

 disease as found in the field. It has been shown, also, that certain 

 organisms apparently alike were originally isolated from a typically 

 diseased tree; several cultures of these were inoculated into other 

 coconut trees, and from the successful infections apparently the 

 same organism was reisolated; that a number of these latter cultures 

 were inoculated into other coconut plants, and after producing suc- 

 cessful infections the same organisms apparently were again reisolated. 

 In the process of isolation from the diseased tissues manj^ plates 

 were poured, and from these numerous agar-tube transfers were 

 made. Not all of the tubes from any one inoculation proved to be 

 identical, but the greater number turned out to be so. Of all the 

 cultures in the successful inoculations there were two which originally 

 came from the same naturally diseased tree and went through the 

 same process of inoculation, isolation, reinoculation, and reisolation 

 in different trees, and appeared in preliminary tests to be identical 

 in almost every case. Other cultures appeared similar in their 

 reaction in several tests, but these were not compared at any great 

 length. The two original cultures were compared with two cultures 

 from the inoculated trees and two cultures of those isolated from the 

 reinoculated trees. These six cultures were used to make rather 



228 



