138 
W. J. Dowson, 
On the 1 st April 1912 the two leaves kept in the petri-dish were 
examined; and on the younger of the two leaves which was thin, small brown 
spots were visible. On cutting sections and treating in lacto-phenol, and 
Bleu-coton, these spots proved to be well developed disease areas, conidia- 
bearing hyphae being found to have emerged from the stomata in some 
cases. Below each spot mycelium was found to be present in the leaf 
tissues. Penetration had evidently taken place some days previously. The 
mode of infection was not made out, but it was thought that in some 
cases this had probably taken place by way of the stomata. In the other 
leaf which was considerably thicker only two short infection hyphae were 
found. The manner of infection was not made out. As both leaves were 
treated under exactly the same conditions, the difference in the results 
must lie in the nature of the leaves themselves. The degree of resistance 
of the younger leaf had evidently sunk faster than that of the older, and 
thicker one, as in the former the fungus had developed at a rate, and in 
such a manner as to indicate that it was growing saprophytically rather 
than parasitically, whereas with the older leaf the behaviour of the fungus 
seemed to be similar to that under natural conditions. 
On the 2 nd April all the other inoculated leaves were removed and 
the inoculated areas cut out and four of these areas after being cut each 
into two pieces and killed in chromo-acetic acid solution, were imbedded 
in paraffin of 52° C melting point in the usual way. The fifth piece was 
at once cut and examined after treatment in lacto-phenol, and Bleu-coton. 
No germinating spores or infecting mycelium could be found in any of 
the sections. 
From the above recorded infection experiments the following 
conclusions may be drawn: The mycelium of H. echinulatum in the tissues 
of Dianthus is intercellular. Infection takes place from 6 to 7 days after 
inoculation, either by hyphae which pierce the epidermis probably by way 
of the middle lamella between two epidermis cells, or through the stomata. 
Conidia are again produced by the parasite after a period of from 3—4 
weeks growth within the tissues of the host-plant. 
IV. Examination of diseased tissues of Dianthus. 
For this purpose original material from the nurseries both fresh and 
in alcohol, and artifically infected plants were used. The material to be 
imbedded in paraffin was killed in Juel’s fluid, boiling alcohol, acetic 
alcohol, dilute Flemming solution, and in chrom-acetic acid solution. The 
last proved the most satisfactory. For free-hand sections, the material 
was killed in alcohol or in hot lacto-phenol. The alcohol material stained 
best in the combination Fuchsin-lichtgrün, or gentian violet-orange G. 
That treated in lacto-phenol was invariably stained in Bleu-coton G 4 B, 
which was sometimes followed by orange G, but excellent results 
were obtained with the Bleu-coton alone and the most satisfactory 
observations were made from this material. This is a plasma stain, 
and hence is taken up by the cytoplasm of the cells of both the 
host and the parasite. The hyphae of the parasite are more deeply 
stained than are the host cells, hence it was possible to wash out 
the Bleu-coton from the host cells, and leave the hyphae stained; the 
