1916] Fernald, — The Genus Sabatia in New England 149 



Saybrook station is unrepresented. Similarly, recent collections from 

 Guilford do not show any large-flowered Sabatia there. The plant 

 may now he extinct in Connecticut and until actual specimens can be 

 examined we cannot tell whether the records rest upon true S. dode- 

 candra (which is probable) or upon the commoner New England repre- 

 sentative of it. 



In southeastern Massachusetts no large-flowered Sabatia is known 

 to the writer in brackish marshes, although S. dodecandra is to be 

 sought in the Buzzard's Bay region. In Norfolk, Plymouth, Bristol 

 and Barnstable Counties a large Sabatia, which has been passing as 

 S. dodecandra,' is common and so invariably a plant of damp sandy or 

 peaty margins of fresh ponds or of acid sphagnous bogs, that the 

 Massachusetts botanist, supposing his plant to be S. dodecandra, is 

 naturally impressed by Dr. Stone's obvious surprise at finding on the 

 Cape May peninsula S. dodecandra " in fresh marshes over a mile from 

 the coast." The Massachusetts plant is rarely if ever found on the 

 actual coast; at least all the herbarium-labels indicate fresh habitats 

 and a majority of the stations are from five to ten miles from the near- 

 est salt water. In many morphological characters the plant of the 

 fresh pond-shores of Massachusetts (and likewise of southern Rhode 

 Island) departs from the somewhat halophytic S. dodecandra. The 

 Massachusetts plant is freely stoloniferous, even at the beginning of 

 the flowering season carefully collected specimens exhibiting elongate 

 stolons with well-developed rosettes of acuminate leaves; the plant 

 is taller (2.5-8 dm. high); the basal leaves are oblanceolate and 

 acuminate, distinctly longer than the firm lance-acuminate subulate- 

 tipped cauline ones; the calyx-lobes are firm and linear-subulate, not 

 foliaceous, with slightly hyaline margin, very obscurely 1-3 nerved, 

 5-15 mm. long, 0.5-1.5 mm. broad, and the calyx-tube nerveless; the 

 corolla-lobes are cuneate-obovate, rounded or emarginate at summit, 

 1.5-3 cm. long, 6-15 mm. broad, with the margins commonly meeting 

 or overlapping, so that the expanded flower suggests the head of a 

 single Dahlia or Cosmos. The yellow spot at the base of each lobe is 

 much broader than in S. dodecandra (2.5-5 mm. broad) and commonly 

 has 3 long pointed lobes, so that the complete brown-bordered yellow 

 central star of the flower has 21-36 rays. This plant of the Massa- 

 chusetts pond margins begins flowering in June — some weeks earlier 

 than S. dodecandra, apparently, — and is in its prime through July, 

 although belated or small secondary flowers may be found through 



