98 



The Ohio Naturalist. 



[Vol. VII, No. 5, 



Sagittaria variabilis. 



1. Monoecious or sometimes 



dioecious. 



2. Leaves sagittate, variable in 



form and size, sometimes 

 broader than long, 15-40 

 cm. long, basal lobes, ovate 

 or lanceolate, acute or 

 acuminate. 

 3 Scapes 3-6 dm. long, angled, 

 simple or branched. 



4. Filaments not dilated, gla- 



brous. 



5. Mature heads 1.5-3 cm. in 



diameter. 



6. Bracts 1.5 cm. long, gla- 



brous, acute or acuminate. 



7. Flowers 1.5-2.5 cm. broad. 



8. Achenes obovate, about 2-4 



mm. long, erect, undulate 

 winged; beak ascending or 

 recurved. 



9. Summer and fall. 



Sagittaria lancifolia. 



1. Monoecious. 



2. Leaves not variable, 4-9 



dm. long; leathery, broadly 

 linear or elliptic, acute. 



3. Scapes 6-20 dm. long, sim- 

 ple or branched. 



4. Filaments not dilated, pu- 

 bescent. 



5. Mature heads 1 cm. in diam- 

 eter. 



G. Bracts ovate or ovate-lan- 

 ceolate, 1-2.5 cm. long, 

 acute or acuminate. 



7. Flowers 1-2.5 cm. broad. 



S. Achenes cuneate or obo- 

 vate, 2-3 mm. long, winged ; 

 beak short, ascending. 



9. Spring and summer. 



With such striking external differences one would naturally 

 expect equally interesting internal differences but to my surprise 

 I found the development of the embryo-sac and embryo of 

 5. lancifolia practically the same as had been described by 

 Schaftner for 5. variabilis. The comparison with Schaffner's 

 results will be brought out in the following discussion. 



Embryo-sac. 

 The author was unable to determine the origin and develop- 

 ment of the archesporium satisfactorily but traced without 

 difficulty, the deveropment of the embryo-sac beginning with the 

 one-nucleate stage (Fig. I). In the formation of the two- 

 nucleate (Fig. 2) stage the sac elongates to twice the length of 

 the one-nucleate stage. Schaffner did not secure material for 

 these two stages but from this point on the development of the 

 embryo-sac of the two species is exactly the same, except that I 

 was inclined to believe the antipodals in 5. lanceolata not quite 

 so presistent as he found them in 5. variabilis. 



Endosperm. 

 The development of the endosperm follows exactly the same 

 course as that of 5. variabilis as described by Schaffner. Schaffner 

 did not fully understand the significance of the lower endosperm 



