360 JOURNAL OF THE ROYAL HORTICULTURAL SOCIETY. 



to sever one and layer it in sterilised loam to see what it would do. 

 Early in September the normal fronds followed their usual habit of dying 

 down, but those of the little cluster remained green and vigorous. The 

 layered one did the same, and in October, on examining it with a lens, 

 I found two excrescences had appeared near the tips of two of the side 

 divisions, and these eventually asserted themselves as bulbils. I then 

 layered another in the same way, and this remained perfectly green 

 through the winter, as did those left untouched, but showed no signs 

 of growth until April of this year, when the central cells began to develop, 

 and at the moment of writing (May 22) have produced spherical ex- 

 crescences which are obviously incipient bulbils. The frondlet first 

 layered has now turned black, but the bulbils have produced one frond 

 and two fronds respectively of the normal seedling type, and the plants 

 are practically established. During the winter the axis of growth whence 

 the original cluster of frondlets sprang developed several incipient 

 fronds, which, when the growing season came, grew up into quite normal 

 C. montana fronds, a remaining abnormal frondlet still persisting at their 

 base. One had been eaten by insects and the other removed for layering 

 or microscopic examination. By the latter it is seen that these frondlets 

 are only one cell thick ; and Professor Farmer, who has kindly examined 

 them, remarks that, although abnormal in this respect, the outline of the 

 cells is normal, and not of that modified type common to aposporous 

 cellular tissues. In any case, however, we have here the unique fact 

 of a normal fern producing two distinct kinds of fronds — the one of 

 normal long- stalked decomposite structure and cellular formation and 

 perfectly deciduous, growing, moreover, in succession from a creeping 

 rootstock, and the other of a very simple short-stalked type of abnormal 

 cellular thinness, persistently evergreen, produced in a circle round 

 a central axis, and, moreover, endowed with a proliferous character of 

 a very peculiar type. Whether this phenomenon was induced by the 

 check to growth caused by the disturbance incidental to the two successive 

 shifts above cited is an open question. If it were, the case becomes 

 an extremely curious example of an adoption, as it were, of modified 

 proliferous structures to increase the chance of survival. The crowns 

 of many ferns, if crushed or mutilated, are capable of developing bulbils 

 in self-preservation ; but so far as records go these are always normal 

 bulbils, induced buds, and are never arrived at indirectly as in this case, 

 through specially modified preliminary outgrowths differing in all respects 

 from the normal specific characters of form, cellular construction, and, 

 above all, deciduousness or otherwise. 



