Pethybimdse and Lafeerty — A Disease of Flax Seedlings. 373 



but it was found that every infected seed could not with certainty be picked 

 out from a sample by naked-eye examination or even witli the help of a 

 strong lens. Thus, one hundred seeds which showed none of the above- 

 mentioned characteristics, but appeared to be absolutely normal and healthy, 

 were selected from an infected sample and sown in moist sterilized silver 

 sand. Fifteen of these gave rise to diseased seedlings, suggesting that in all 

 probability a small amount of mycelium was present in their seed-coats, but 

 not enough to make the seed appear in any way abnormal. 



As is well known, the flax seed-coat consists of several layers of cells. 

 The innermost layer, abutting on the endosperm, is the pigment layer. Next 

 to this comes a layer of compressed, empty, thin-walled cells, which is followed 

 by a layer of thick-walled sclerenchymatous fibres ; and .this is succeeded by 

 a layer, one or two cells thick, of scpiarish parenchymatous elements. The 

 outermost layer of all — the epidermis — is composed of cells, the outer walls 

 of which absorb water readily and become much swollen when wetted, giving 

 to soaked flax seed its mucilaginous character. A distinct cuticle is present 

 on the outer surface of these walls. 



Sections cut through dry infected seeds, when mounted in water, present 

 the appearance shown in fig. 4, Plate XIX., and fig. 7, Plate XX. It will be 

 observed that the mycelium is found only in the cells of the epidermis, and we 

 have not been able to discover cases of penetration into any of the other layers 

 of the seed-coat or into the endosperm or the embryo itself. Hence, although 

 infection of the seed in the case of this disease is, strictly speaking, internal, 

 it is only skin-deep. The embryo, within the seed, is free from infection. 



It was next necessary to ascertain for certain whether the mycelium 

 present in the epidermis of infected seeds was alive or not. 



For this purpose a sample known to contain infected seeds was taken, and 

 from it eight seeds were selected, each of which contained mycelium in its 

 epidermis, this being determined by preliminary microscopic examination. A 

 further lot of eight seeds was selected from the same sample, in which 

 microscopic examination failed to reveal the presence of mycelium. The 

 infected and non -infected lots of seed (as they may be called) were placed 

 on moist filter paper in Petri dishes and kept in an incubator at 20° C. In 

 three days all the seeds in both lots had germinated, and on the surface of 

 six of the infected seeds Colletotrichum was present^ and bearing conidia. 

 No fungus growth was present on the seed-coats of the eight non-infected 

 seeds. Two days later the experiment was concluded. It was then found 

 that seven of the non-infected seeds were free from fungus, while on the 

 eighth there was a trace of Penicillum but none of Colletotrichum. Of the 

 two infected seeds which had not developed Colletotrichum on the third day 



