1041 



this part being split by the least accident, causes it to be more or less 

 notched or bifid. This character is common to C. teretinscula, pani- 

 culata, paradoxa, and to my pseudo-paradoxa, but on examination it 

 will be the least distinct in paradoxa and most distinct in my pseu- 

 do-paradoxa. Carex paniculata is spoken of as having the beak of 

 its perigynium more deeply bifid ; with a broader and deeper fringed 

 margin. In this I find no difference between paniculata and its allies, 

 as they are all equally variable in this respect. I have the perigyni- 

 uras of paniculata with their margins as narrow and as indistinctly 

 fringed as any I ever saw in teretinscula; and I have often met with 

 paniculata with the beaks of its perigyniums quite entire, that is to 

 say, neither bifid nor notched. To speak of the beaks or lateral mar- 

 gins of these perigyniums being more or less white or green, would 

 only be losing time, as any person who has paid the least attention to 

 the subject will know how far we can depend on such a character, as 

 all of them will (T think) be found green when in a young state, and 

 more or less white with age. 



Having now .spoken at some length on the beaks of the perigyniums 

 of these plants, I now come to notice their lower parts : and on exa- 

 mination I find these parts (in all the four forms) to be equally varia- 

 ble, that is, in being convex on one side, and flat or concave on the 

 other, or in being more or less convex on both sides. I have seen C. 

 paniculata with its perigyniums even concave on both sides ; and at 

 other times I have met with it ha^dng its perigyniums as much inflat- 

 ed and quite as gibbous on their inner faces as ever I saw those parts 

 in C. paradoxa; and on the other hand, I have seen paradoxa with 

 the inner faces of its perigyniums flat, and in some instances concave 

 on that side. In short, I always consider the perigynium as the most 

 variable part in all Carices, the shape depending much on their being 

 more or less densely placed on the rachis. Thus, a spike or spikelet 

 (of the same species) having four, six or eight rows, would have them 

 of various shapes, and when placed very close one above the other, so 

 as to press the beaks outwards, they will be found more convex upon 

 their inner faces than when they are more distantly placed. 



My reasons for not considering the perigynium of Carices as being 

 part of their fruit, are as follows. First; because I often find them 

 with pistils which remain abortive : secondly, with stamens only: third- 

 ly, with both stamens and pistils : fourthly, proliferous, that is, with 

 the peduncle of one perigynium passing through the beak of the next 

 below, and thus rising one above the other for four or five heights ; in 

 this state they are sometimes abortive, and at other times produce per- 



