290 MEMOIRS OF THE CARNEGIE MUSEUM. 



L. H. D. 



Size: (Males) 1. Erie, Cat. No. 61.4858 89 mm. 49 mm. 31mm. 



2. do. Cat. No. 61.4020... 76 " 44 " 27 " 



3. do. Cat. No. 61.4020 51 " 29 " 18 " 



(Females) 4. Erie, Cat. No. 61.4020 (gravid) 81 " 54 " 35 " 



5. do. Cat. No. 61.4020 (gravid) 64 " 41 " 28 " 



6. do. Cat. No. 61.4860 52 " 34 " 20 " 



Soft parts identical with those of the main species, glochidia also identical. 



Breeding season: Gravid females are on record for May 21, 1909; May 22, 

 1909; May 24, 1909; July 7, 1910; July 12, 1910. All these specimens had 

 glochidia, and on July 7 and 12 a number were observed discharging the glochidia. 

 The variety is certainly bradytictic, like the main species. 



Remarks: The original description of rosaceus DeKay is founded upon speci- 

 mens with special features: rosy nacre and pallid brown epidermis. But Simpson 

 (1900, p. 535, footnote 1) pointed out that this probably is only a local effect, 

 and that other specimens have white nacre and differently colored (darker) epi- 

 dermis.'^^ I find, that there is in Lake Erie great uniformity in the color of the 

 nacre, which is always white, while there is great variability in the color of the 

 epidermis. Specimens from the surf-beaten shores of the open lake are remarkably 

 light, and the ground color of the epidennis is light yellow or light greenish (grayish- 

 green), while those from protected localities in Presque Isle Bay, and from beach- 

 pools, are more greenish-olive, lighter or darker, and often incline toward a reddish 

 brown, which is most intense and brilliant towards the beaks. In the growth- 

 lines there is great variability, but the tendency is to have them rather closely set, 

 rather distinct and regular. This is most evident in specimens from deeper water, 

 least so in those from the beach-pools. 



The characters of this variety are very poorly marked. It is true that speci- 

 mens from Lake Erie generally are recognizable by size, color, and distinct growth- 

 lines; but in specimens from other localities these characters are more or less 

 unstable, and transitional conditions toward true luteola are frequent. I have 

 already mentioned that in L. luteola from Lake Chautauqua, which incline toward 

 rosacea in size, other characters are normal. Specimens from Winona Lake look 

 much like the Lake Erie form, having the reddish epidermis and very distinct 

 growth-lines, but are larger. Specimens from Moose River are small and pale in 

 color, but the growth-lines are indistinct. 



Specimens collected by O. E. Jennings on the North shore of Lake Superior 

 much resemble the Lake Erie form, being pale in color, without rays, but they differ 



'*' Utterback (1916, p. 185) relying only on the color of the nacre, erroneously called the red-nacred 

 form of Missouri (also found in Arkansas) by the name of L. luteola rosaceus. 



