372 ANNUAL KECOKD OF SCIENCE AND INDUSTRY. 



pared material. A study of living forms would have shown, 

 however, that the cushion, or gelatinous stipes, by which the 

 frustules of the Isthmia are attached, and which constitutes in 

 another form the stalks of Gomphonemew and Cocconemece, is 

 in fact a closed gelatinous tube, out of which the diatom is 

 constantly slipping. When the tube is not closed we have 

 the fronds of ISchizonemece and Collet onemece. All the dia- 

 toms secrete a more or less gelatinous sheath, over the whole 

 surface of the frustule; it is not exuded at one end. "What 

 Mr. Cox calls the earlier view, viz., that the sporangial frus- 

 tule becomes " the parent frustule of a new and vigorous gen- 

 eration, by recommencing the cycle of self-division," is not 

 the earlier view, but is the later, accepted by all who have 

 studied this subject in recent times, excepting perhaps Dr. 

 Wallich, whose article, alluded to by Senator Cox, has many 

 misjudgments and wrong statements. Thwaites himself did 

 not express an opinion clearly. "W. Smith, his intimate friend 

 and the well-known British authority, in his classical work, 

 supposes the cell-contents of the sporangium to break up and 

 form broods of small individuals. Kabenhorst has even fig- 

 ured something of this kind in Melosira. and this was Kutz- 

 ing's view in a general way. There is not a shadow of doubt 

 that the opinion expressed by Senator Cox is the correct 

 one. It has been advocated by Mr. Carter and others, and 

 especially by Professor H.L.Smith (see the Lens,vo\. i., p. 73), 

 who lias not merely observed the conjugation in some fifty 

 genera, but has kept the sporangial frustules living until the 

 self-division has been completed. 



TISSUE -STAINING, BLOOD, ETC. 

 Structure of the Colored Blood -Corpuscles. 



Dr. II. D. Schmidt, of New Orleans, states that the colored 

 blood -corpuscles of the Amphiuma represent organic cells 

 consisting of homogeneous protoplasm surrounding a nucleus, 

 that part of the protoplasm forming the outer surface being 

 of greater density than the rest, and so a membranous layer 

 is formed, and may be seen in the form of a clear, narrow 

 border of a greenish tint, and by the action of reagents it as- 

 sumes the appearance of a distinct cell-wall. The coloring 

 matter (luemoglobiii) of the blood-corpuscle does not seem 



