470 LESLIE B. AREY 



The statement has often been made that glochidia swim by- 

 clapping together the shells, after the manner of Pecten. This 

 is, however, wholly erroneous as regards American species.^ 

 From the time of discharge they lie on the bottom, with gaping 

 valves directed upward, incapable of locomotion, although 

 subject to passive dispersal by water currents.^ It is true that 

 glochidia, more particularly those of the hooked type, show 

 spontaneous contractions which may continue irregularly for 

 long periods;^ but these pulsating movements at most cause 

 merely a toppling over of the larva in case the closure is complete. 



III. EXPERIMENTAL 



A. Tactile excitation 



The only significant reference to the sensitivity of glochidia to 

 mechanical stimulation is in the report of Lefevre and Curtis 

 ('10, '12). These workers found that the hooked Symphynota 

 glochidia respond very readily to the touch of a needle point or 

 paper edge; they continue to clasp such objects until death 

 supervenes. This tactile response was believed to be the chief 

 factor responsible for the attachment of hooked glochidia to the 

 fins or other soft external parts of fishes. As might be expected, 

 those ventral fins which brush the bottom tend to become most 

 heavily infected. Hookless glochidia, however, were said to 

 differ markedly from the hooked type, inasmuch as they ''respond 



3 Dr. R. E. Coker writes that he is reliably informed that glochidia of Japan- 

 ese mussels do swim Pecten-fashion; such activity would off er many interesting 

 points for study. 



* Thus, a variety of glochidia may appear regularly in surface tows of river 

 water (H. W. Clark, unpublished observations), or in water samples from greater 

 depths (Kofoid, '08). This condition possibly explains the natural propagation 

 of such a mussel as the 'nigger head,' Quadrula ebena. Its glochidia have a 

 restricted parasitism on the blue herring, Pomobolus chrysochloris, which is not 

 a bottom feeder, but predaceous; since, however, both mussel and fish are char- 

 acteristic of swiftly flowing water, the chances of infection are still good. 



^ In several species of the Lampsilis group this was most common directly 

 after removal from the maruspium. Hooked forms may continue spontaneous 

 contractions, at irregular intervals, for days. In Anodonta corpulenta the 

 amplitude is great; the opened valves of the larva lie in the same plane, and 

 explosive, winking contraction nearly closes them. 



