108 



4. The ovules and the anthers of Ginkgo are both marginal for- 

 mations of sporophylls, though the anthers of Goniferce in general have 

 hitherto heen often considered as organs developed on the under surface 

 of staminal leaves. 



5. The normal seed-stalk is the floral axis, whose apical bud is 

 usually suppressed, and which bears only two rudimentary carpels. 



6. The elongated stalklet of the ovule of Ginkgo occasionally 

 developed is not simple outgrowth, as has been considered by some 

 authors, but is the petiole of the carpellary leaf developed alternately 

 along the floral axis. 



7. The normal female flower is not an inflorescence as has been 

 considered by Strasburger and others, but a real single flower as con- 

 sidered by Eichleb, Celakovsky, and others. 



8. The disputed point in the homology of male and female flowers 

 of Goniferce is thus clearly established as far as Ginkgo is concerned, 

 as is seen when we compare a normal male flower and a female flower 

 with a number of stalked ovules. 



Remarks on the phytogeny of Ginkgo: — The view that among Goni- 

 ferce, Ginkgo comes nearest to Gycadacece has hitherto been often held 

 on account of its lobed dichotomously veined leaf and drupaceous 

 fruit. We may here briefly enumerate the points of resembrance between 

 Ginkgo and Gycadacece, and especially Ginkgo and Gycas. 



(a) The formation of ovules upon the foliage leaves of Ginkgo 

 reminds us of the carpellary leaves of Gycas, and the ovules are on both 

 genera marginal formations of the sporophyll. 



(6) Drupaceous fruits of Ginkgo and Gycadacece much resemble 

 each other. 



(c ) The neck of archegonia consists in both Ginkgo and Gycadacece 

 of two cells. 



( d ) The formation of the canal cell was formerly believed not to 

 take place in Gycadacece, while it was well known in Ginkgo, and 

 indeed in Goniferce in general ; but it has been recently proved to take 

 place in Gycas revoluta^ . 



(1) S. Ikeno. Vorlaufige Mittheilung iiber die Canalzellbildung bei Cycas revoluta. Bot. 

 Centralblatt. Bd. LXVIL No. 7, p. 193 (1896). 



S. IkeilO. Note pr^liminaire sur le Formation de la Cellule de Canal Chez le Cycas revoluta 

 (With pi. V.). The Bot. Mag. Tokyd, Vol. X, No. 115, Part II p. 63 (1896). 



