THE ANATOMY AND AFFINITY OF DEPAR1A MOOREI, HOOK. 853 



do not reach the leaf-margin may have provided new points of origin for sori. The 

 sori thus initiated must be, from their first origin, superficial sori. The selection of 

 the upper rather than the lower leaf-surface for the origin of the superficial sori is 

 an exceptional feature, but there seems some ground for the opinion that whatever 

 determines the selection of the upper surface for encroachment by the marginal sori, 

 likewise determined the position of the truly superficial sori. It may be that the 

 superficial sori are, when the leaf is expanded, open to exposure to too strong illumi- 

 nation, but the broad and thin type of leaf probably indicates that the plant is 

 adapted to a shade habit. The situation of dense shade in which Moore's original 

 plants were gathered may lend some support to this suggestion. 



It would be difficult to prove which had the priority in initiation, the reticulate 

 venation, or the superficial position of the sori. They may have evolved together. 

 They are, however, at least interdependent in that it is the reticulum which supplies 

 the conductive tissue to the superficial sori. I am inclined to believe that in this 

 case the priority of initiation is to be given to the venation, and that the superficial 

 position of certain exceptional sori is adopted because convenient vein-endings which 

 do not reach the leaf-margin have offered the physiological opportunity for soral 

 formation even at points apart from the normal. Such sori may be held as relatively 

 late and individual developments, rather than as undermining the value of the 

 normal position of the sorus as a criterion for comparison of Ferns. 



The mixed sorus with slightly basipetal succession is certainly an advanced 

 feature, and it is not surprising that going along with this are the long-stalked 

 nature of the sporangia, the vertical annulus, and the small spore-output per 

 sporangium. 



As regards the systematic or phyletic position of Deparia, various opinions 

 have been held. The genus was recognised as distinct in the Synopsis Filicum, 

 having been founded by Hooker and Greville. In the generic description (I.e. p. 55) 

 it is stated to include " three very rare tropical species, with broad, ample leafy 

 segments, which differ from Dennstsedtia mainly by their extra-marginal sori." 

 By "extra-marginal" is here meant the stipitate state which they show. The 

 relationship with Dennstsedtia is clearly suggested. The reticulate species designated 

 Cionidium by Moore, and later styled Deparia Moorei was also included by Hooker 

 in the genus Deparia. More recently, however, J. G. Baker suggested the actual 

 inclusion of the genus Deparia in Dicksonia (Gard. Chron., Feb. 1872). This was 

 at a time when Dicksonia was held to include Dennstsedtia. This inclusive genus 

 — Dicksonia — having now been broken up, the genus Deparia, like the genus 

 Dennstsedtia, would take substantive rank, but would remain related to the 

 Dicksonioid-Davallioid series. 



On the other hand, Christ (Farnkrduter, pp. 223, 230) placed Deparia prolif era 

 in Athyrium, and Deparia Moorei in Aspidium, thus separating generically these 

 two allied species. Diels [E. and P., I, 4, p. 186) includes Deparia in Aspidium; 



TRANS. ROY. SOC. EDIN., VOL. L, PART IV (NO. 26). 122 



