POLYPODIUM VULGARE AND ITS VARIETIES. 



113 



full sun for some hours daily during the summer months. It has a 

 wonderful power of enduring drought. But while the others flourished 

 P V. cornubiense only existed, and its most perfect development 

 P. V. cornuhiense tricliomanoides could scarcely be kept alive. Many 

 devices were tried — differing sites, exposures, and composts — but results 

 w'ere disappointing. Some good pans were obtained under glass, but 

 out of doors little progress was made. 



At this stage it happened that the trunk of a Portugal laurel wdtli 

 a central hollow was introduced into the fernery, and an attempt to 

 establish a cornuhiense in the hollow w^as made. It seemed a desperate 

 venture, considering how the fern had treated the many and varied 

 allurements with which I had already wooed it. Nevertheless I deter- 

 mined to make the attempt as an experiment. Accordingly a compost 

 of rich leaf mould with a plentiful addition of lime rubble and sand 

 was prepared. The liking of the fern on the one hand for the top 

 of a wall, on the other for a low situation where leaves fall thick, 

 pointed to the kind of soil which it enjoyed. I selected a plant of 

 P. V. coniabiense and one of P. v. fnultifido-elegantissimum, i.e. a 

 crested curnubiense, and placed them in the hollow, keeping the compost 

 in position by means of sods of turf made firm by wire. The specimens 

 were healthy, but their fronds were small. Under no circumstances' 

 that I am aware of does P. vulgare move quickly after disturbance. I 

 was therefore not surprised at the passing of some months with nothing 

 fco note except that the ferns did not seem likely to die. When, how- 

 ever, in the following summer the next set of fronds appeared it was 

 manifest that a change for the better had taken place. Not only had 

 the root stocks spread more rapidly than before, but the new fronds 

 were bolder and better developed than any I had hitherto grown, and the 

 improvement continued. In the meantime, encouraged by the success 

 of the experiment of growing in wood, I made use of other tree stumps 

 for the same purpose. 



The fernery is sheltered by high walls on the south-east and south- 

 west ; on the north-west, the point from which our worst winds come, 

 there are thick shrubs. A large elm was cut into lengths, some of 

 which were hollow^ in the centre, and others were easily scooped out 

 to a sufficient depth, w^hile one or two were sound, and had to be 

 cut out with an axe. The hollow lengths were filled with stones to 

 within a few inches of the top. On the stones tough peat sods were laid 

 with the upper side down. The shallow hollow which remained was 

 filled with a compost of leaf mould, say four parts, and of lime rubble 

 and sand, say half a part of each. The lengths which showed decay 

 were considered capable of letting the water drain downward through 

 the wood ; all that was done was to lay crocks in the hollow with moss 

 on top of them, and over this a similar compost. The sound lengths 

 were carefully and somewhat kboriously drained by auger holes, after 

 which they were treated like those that were decaying. The stumps 

 w^ere arranged in December 1905, and the planting was done during the 

 winter (fig. 50). 



VOL. XXXVI. 1 



