608 THE PALEONTOLOGY OF MINNESOTA. 



[Ly rodesmldSB . 



to the beaks, placed obliquely and so that they converge inwardly, the direction of 

 the anterior series being nearly at right angles to that of the posterior series. 

 Clavicle strong, nearly straight, almost vertical, sharply defining the somewhat semi- 

 circular and large anterior muscular scar and leaving a strong furrow in casts of the 

 interior just in advance of each beak. The furrow extends beyond the middle of 

 the distance to the basal margin. Posterior scar faint, smaller than the anterior, 

 occupying a central position on the post-cardinal slope. Several small umbonal 

 scars may be observed on good casts, and obscure rays are occasionally visible on 

 their sides. 



Hall's C. fabula, described from Cincinnati specimens less than 2 mm. in length, 

 seem to me to be nothing more than a dwarfed variety of this species. 



Formation and locality. ID the so-called "Nucula Beds" of the Maquokcta (Hudson River) shales at 

 several localities in Lafayette county, Wisconsin; Jo Daviess county, Illinois, and near Dubuque and Graff 

 in Iowa. It is to be found, 1 think, in the equivalent beds in Fillmore county, Minnesota. 



Mug. Reg. No. 7336. 



Family LYRODESMID^E, Ulrich. 



A reconsideration of the genera included in this family, on page 486 of this work, 

 has convinced me fully that they are improperly associated and that the family must 

 for the present rest solely on the typical genus. Dr. S. A. Miller was, I now believe, 

 right in proposing a new family for his genus Technophorus (N. A. Geol. and Pal., p. 458, 

 1889), but he should have included the closely related Ischyrina, Billings, a genus 

 doubtfully referred by him to the Trigoniidu 1 . The new genus Allodesma proves to 

 be related to Cyclochoncha, Miller, rather than to Lyrodesma and should therefore be 

 removed to the provisional family Cydoconchidce. 



The proper arrangement of these three families in a scheme of classification is 

 a point upon which it is very difficult to come to a satisfactory determination. Con- 

 siderable agreement in structure is to be traced between them, and at times 1 might 

 go so far as to say that they should be regarded as closely related. Still, in view of 

 the fact that each in one way or another resembles types classed in such widely 

 distinguished families as the Trigoniidce, Crassatellid>, Cyrenidw and Myidce more 

 closely than they do known Devonian and Carboniferous forms, it would obviously 

 be an expression of opinion quite insufficiently supported by facts. 



Genus LYRODESMA, Conrad. 



Lyrodesma, CONBAD, 1841. Ann. Geol. Rep. N. Y., p. 51; HALL, 1847, Pal. N. Y., vol. 1, p. 302. 

 Actinodonta, PHILLIPS. 1848. Mem. Geol. Sur. Great Britain, ii. 



Shell moderately convex, larger than high, ovate to subquadrate, rounded in 

 front, usually obliquely truncate behind and more or less angular post-basal ly. 



