13 



thoroughl}' as possible in this way. A similar experiment was made 

 with fornitiliii 1 to 1,000, and a third one with formalin 1 to 500, it 

 being thought that the formalin would be effective in penetrating 

 l)et\veen the young leaves and thus destroying the germs. Although 

 these applications were continued twice a week for a month, the 

 young growth continued to be as ])adly diseased as even-, there being 

 no apparent difference between the treated and the untreated plants. 

 A few thrips were observed working on the plants and also a few 

 aphides, but it was not believed at the time that these could possibly 

 account for the trouble, and they were looked upon simply as distrib- 

 uters of bacterial infection. 



Finally it was decided to make no further attempt to control this 

 disease, but to keep the houses a little warmer and moister, force the 

 flowers into })loom. and get as nuich out of them as possi])le. The plants 

 were therefore syringed on bright days and the air of the house kept 

 quite moist. Much to the writer's surprise, they began to grow out of 

 the trouble and produced some fine flowers. These peculiar results, 

 together with the similarity of the disease affecting the lily, as before 

 stated, led to a careful study of the trouble. 



MICROSCOPICAL INVESTIGATIONS. 



Besides the affected plants in the carnation houses at Washington, 

 other plants showing the trouble were obtained from man}' of the large 

 carnation-growing sections, including some plants that Dr. Arthur 

 kindly sent in from Mr. Fred Dorner's place at Lafayette, Ind. 

 Leaves showing various stages of the disease were killed by both the 

 chromic acid and the absolute alcohol method, and then were dehy- 

 drated with alcohol and infiltrated with paraffin in the usual way. 

 Microtome sections were then cut and mounted in series, and these 

 were stained with Ziehl's carbol f uchsin, and also, according to Gram's 

 method, with aniline water gentian violet. The ordinary aqueous solu- 

 tions of fuchsin and gentian violet were also tried along with other 

 stains. About two thousand sections, representing all phases of the 

 disease from different localities, were prepared and stained. 



The cells of the diseased spots were found to be much larger than nor- 

 mal, and thin-walled and oedematous (figs. 1, 2, and 3, and PL III, figs. 

 1, 2, 3). In the early stages of the disease the chloroplasts were unde- 

 veloped or smaller than in the healthy cells and were colorless or yellow. 

 Even after the most thorough and careful staining no parasitic or sapro- 

 phytic organisms could be detected in the tissues of these spots until 

 after the epidermal cells collapsed, when, in some cases, fungi and bac- 

 teria were readily distinguished, although usually only in small num- 

 bers. A very curious structure was always present in the earlier stages 

 of the common form of the disease. It stained slightly with carbol 



