155 



divided instead of being merely bipinnate. All the pinnae halt- 

 way up the frond were truncate, ending squarely, with the 

 midrib projecting as a small translucent thorn, and the 

 top of the one complete frond (that of the other was broken 

 of!) was truncate also with a similar projection. On 

 digging this plant up, we found to our intense surprise 

 that the remains of five or six fronds of the previous 

 season were still attached, and sufficiently intact to show 

 that they were absolutely of the common normal type 

 without a trace of the three abnormal characters, lucent 

 surface, tripinnation, and truncation, visible. The fronds 

 bore spores profusely, and despite the obvious inconstancy 

 we determined to sow some, in the hope that at any rate 

 a few plants of the new and very uncommon form would 

 result. We therefore severed a pinna (a truncate one for 

 preference), laid it on a glass slip, and shortly after, 

 examining it with the microscope, discovered a good batch 

 of shed spores. These we sowed in the usual way, first 

 sterilizing the soil and also, as is our custom, sowed them 

 in one of our dwelling rooms and then covered them close 

 with glass. Subsequently a minute patch of an incised 

 Asp. trichomanes from Mr. Moly's collection, was sown with 

 them. 



The first results of the sowing, when the plants declared 

 themselves, was a dense forest of Lastvea filix mas, many 

 of which showed signs of cresting, and none of which 

 struck us as normal. These grew so robustly as to give 

 no chance to the Polystichum we hoped for, and hence as 

 they presented no promising features, they were all cut out 

 with a pair of sharp scissors. It was then seen that beneath 

 them were a few plants of Asp. t. of the incised type of 

 Mr. Moley's plants, and only here and there a Polystichum 

 seedling, while one or two crested Lady Ferns and as many 

 cvispate lastveas had obviously, despite our precautions, 

 resulted from strays. These Polystichums, only some half- 

 dozen in all, instead of developing the truncate and compo- 

 site character of the presumed parental frond, all produced 



