44: 



JOURNAL OF THE ROYAL HORTICULTURAL SOCIETY. 



brane frequently stains more deeply than does the normal protoplasm. 

 Appearances which might be mistaken for altered cell-wall — in micro- 

 tome preparations — are really due to fragments of deeply staining 

 protoplasmic membrane. Nc^ hyphal organism, has been observed in 

 the diseased tissue. 



The diseased areas in tubers affected with Phytophthora infestans 

 are soon enclosed by an active wound cambium. Again, in internal 

 disease and streak-disease, the spots or blotches and streaks of 

 brown tissue are soon shut ofi from the more healthy tissue by a deve- 

 lopment of wound cork. The writer has not yet observed a formation 

 of wound cork in bruise. It seems to be rather a remarkable exception. 



A number of microtome preparations were made with the object 

 of ascertaining whether definite forms of bacteria could be detected 

 in the diseased cells. Bacteria were certainly not present in masses 

 such as they have been described as occurring in diseased plant-tissues 

 by Erwin Smith. Many minute bodies were present bearing a strong 

 resemblance to them, but upon very careful microscopic analysis of the 

 normal and diseased potato-cells, before and after fixation, it was found 

 that intracellular particles frequently simulated bacteria in form, so 

 that the result of the study was inconclusive. 



Experiments to bring about a spread of the disease from one 

 tuber to another produced a negative result. Neither was the diseased 

 area observed to increase in extent in an affected tuber, except in the 

 one case referred to above. Nevertheless, an indication was obtained 

 that the blackening did spread during the storage period, in that tubers 

 examined at progressively later times showed an increasing proportion 

 of potatos affected with the disease and an increased area of disease in 

 individual tubers. 



It occurred to the writer in 1908 to test qualitatively the diseased 

 areas of tubers affected with bruise for the alkaloid-glucoside, solanin, 

 and to compare the reaction with that obtained when similarly situated 

 but not actually diseased portions of tissue were treated with the proper 

 reagents. Both the Brandt and Mandelin reactions for solanin were 

 employed and the tests were repeated several times. In several cases 

 the reaction was distinctly stronger in the diseased area as evidenced by 

 the intensity of colour. Even in doubtful instances the balance of the 

 observations was more in favour of a stronger than a weaker reaction 

 in the discoloured tissue. 



The tests were repeated in 1912, using greater precautions to secure 

 the best possible results. These tests are now described in detail : — 



1. Mandelin' s Reaction. — This reagent was used w^hen freshly pre- 

 pared. It was made according to Mandelin 's formula by dissolving 

 one part of ammonium vanadate in 1000 parts of a mixture of 98 parts 

 of concentrated sulphuric acid with 36 parts of water. Similar pro- 

 portions were used by Wothtschall,"^ who tested the efficiency of 

 both Mandehn's and Brandt's reactions. The reagent was apphed to 



* E. WoTHTSCHALL. " Ueber die mikrocheniischen Eeaktionen des Solanin." 

 Zeitschr. f. w. Mihrosk. Bd. v. p. 19. 



