or 



Campbell (V. ) state tlmt the;; ure of external origin^ as 

 Goehel iincl Bf^sgen Jiad found to "be the case in Rlarsilia. 

 According:; to my ov.ti observations on Mcirsilia quad- 

 rifolia, tlie sporocarp first makes its appoarancn v-hen t:ir: 

 yoiing leaf on v/hich it is borne, consists of aboui' six 

 or seven eegments on each side; thus lonr hefore tlin li'jnina, 

 or even the segments tliat are to develope tiiis, iiave been 

 fonned. It arises by the increase in size and bulging out, 

 of apparently eitlier^the acroscopic o:* the basiscopic 

 ultimate marginal cell, of what ir. probably the second 

 segment on its side from the base of the leaf (Fig. SV). 

 Because of the begirining of curvature of the leaf, the 

 proximity of tlie sporocarp rudiment to the axillary bud 

 of the same leaf and the resulting difficulty in orient inr 

 so as to get exactly longitudinal sections of the leaf 

 through the sporocarp rudiment, i'. v/as impossible to de- 

 cide certainl" in v/hich sofment 'he latter arises. It is 

 probably the so cond rather thcui either the first o third 

 segment, and cannot be a i^oviij^^ji' one tlian the latter in 

 any of the cases seen, and though no indication of this 

 v;as seen, it is possible tliat tiie exact point of origin 

 may be shovn, on further observation to vtiry* The sporo- 

 carf; always arises from a marginal cell of tlie iriner and 

 ventral side of tlie young petiole (Fig. 3), the tip of 

 which at thiu uh.i-j is beginning to take a position nearly 



