Since each tube is linked to its neighbour with clean flexible tubing, and the air bubbles through the liquid in the reservoir (C), and has to pass the cotton-wool plug (B) before reaching the leaves (F) in the air above, there can be no question of infection from outside, and the results also show that infection only occurs exactly where I place the .yores on the leaf in each case.
The spores employed were carefully tested as regards their germinating power, and, as the table shows, the results in the closed tubes fully bear out previous experience. In the aspirated tubes, however, the second pair of seedlings of B. mollis (No. 712) gave negative results, inasmuch as only flecks, and not pustules bearing spores, were developed. In the closed tubes, however see below the positive results, especially on B. velutinus and B. secalinus, were excellent, and subsequent examination showed that the spores germinated well, and were capable of infecting other seedlings.
In order to test further the behaviour in mineral solutions, I pre- pared, as Table III shows, several series in closed tubes, Nos. 713, which served as a parallel series to Nos. 711 and 712, but without aeration.
In No. 713 the sterile seedlings were raised antiseptically as before, but the roots merely penetrated cotton-wool saturated with Knop's solution, and held by the constriction over the bulb filled with the- same, no air being drawn through.
The growth was excellent, and the results very conclusive, as Table III shows.
The seedlings were allowed two days at 22 20 C. in the laboratory and then put out side by side with 711 and 712 in full sun during the middle of the day, and after two days' further growth were infected.
By the 10th day the thin leaf was well developed, and the first pustule was seen on B. mollis and B. secalinus on the eighth day after infection. The Table III shows the course of events in detail.
The growth of pustules was excellent on B. velutinux and B. secaliniis especially.
This experiment is interesting, not only as showing that plants can be grown and infected successfully in these closed water-cultures, but especially as showing the contrast between the aerated and non-aerated tubes, for, since the infected seedlings were selected in each case from the same Petri-dish cultures, we must assume that the difference in rate of development was due to the difference of ventilation, and perhaps conclude that this interferes with the success of the parasite, as measured by the somewhat longer incubation period. It is remarkable how dwarfed the continuously aerated plants are, compared with those in closed tubes, owing to the elongation of the leaves of the latter.
It is clear, therefore, that pure cultures of Uredo-spores can be obtained by this method, and it is equally clear that we can also obtain
