XI. CLIMATAL ZONES. 493 



by the limits of any single species ; but that they were to 

 be known partly by the presence or absence of several 

 species, partly by geographical boundaries. In truth, the 

 upper limits of all the plants mentioned are subject to 

 local variations both in regard to elevation and in rela- 

 tion to the limits of other plants ; so that it is quite 

 doubtful in some instances, whether certain species ought 

 to be considered as ceasing in one zone, or as passing 

 slightly into the zone next above. For instance, in the 

 three lists of upper limits, adduced in volume first, pages 

 23 to 27, the following sequences of names occur amongst 

 others : — 



Anthoxanthum. Calluna. Anthoxanthum. 



Calluna. Anthoxanthum. Tormentilla. 



Tormentilla. Tormentilla. Calluna. 



With only these three sequences of names to guide our 

 decision, it would be impossible to feel much confidence 

 in the assignment of the Anthoxanthum and Tormentilla 

 to the superarctic zone, that above the Calluna, as made 

 in the summary of distribution. On the contrary, it 

 would so far appear more i^roper to place the limit of the 

 Calluna between the limits of the other two plants, 

 thus: — 



Anthoxan., 12 1. Calluna, 2 13. Torment., 3 3 2. 

 But the zonal position of these species in the ' summary 

 ,of distribution' has been determined on far more ex- 

 tended observation than shown by the three lists only. 

 The Anthoxanthum has been so frequently seen much 

 above the Calluna, as to leave little or no uncertainty in 

 respect to its natural upper limit being above that of the 

 heath. The tormentil is less certain, the alternations of 

 upper limit between it and the heath being more equal, 

 and the distance between them being usually less. But 

 the former having been observed decidedly above the 



