THE AMERICAN NATURALIST. [Vol. XXXVI II. 



to have any decided effect. This question, however, can be 

 satisfactorily settled only by the study of fresh tissue. 



Between the areas of parenchyma are sharply defined, more or 

 less cylindrical masses of tissue surrounding the vertical tracheal 

 stems and their branches. It is about these cylinders and their 

 tracheae that the greatest interest is centered. The cylinders are 

 from 23 /a to 68 /A apart, and average about 30 /x in diameter. 

 Their appearance differs greatly with the fixation. In material 

 fixed in alcohol or Gilson's fluid, and stained with hematoxylin 

 and eosin they appear as areas less granular, and consequently 

 less deeply stained, than the intervening parenchyma. They 

 contain a large number of small nuclei, especially abundant near 

 the tracheae. After fixation in Hermann's fluid and staining 

 with safranin the cell outlines appear very distinct. (Figs. 10 and 

 II.) Preparations fixed in Flemming's solution and stained 

 with safranin also show cellular structure, though not so defi- 

 nitely as the preceding. Less indication of the structure of the 

 cylinders is shown in the osmic acid material. 



If fresh material, placed for fifteen minutes in .5 % osmic 

 acid and then treated for a few minutes with weak caustic 

 potash, is viewed from the ventral surface, the cylinders appear 

 as very distinct rings. In all preparations, both temporary and 

 permanent, the boundaries of the cylinders are sharply defined. 



Within each cylinder is a main- tracheal stem which gives rise 

 to numerous branches in the characteristic aborcscent manner. 

 There is no change in the structure of the tracheae until near 

 the periphery of the cylinder, where each fine tracheal twig 

 breaks up into tracheoles. The number of tracheoles arising 

 from one tracheal twig seems to vary somewhat. Ordinarily 

 there appear to be only two, but three or four are not uncom- 

 mon. Emery gives the number of tracheoles in Luciola as 

 being uniformly two. In Lampyris, Wielowiejski found the 

 number variable, as many as six sometimes occurring. 



The tracheoles are fine tracheal branches and are characterized 

 by having no spiral thickening of the intima. Their chitmous 

 structure is plainly shown by the fact that they persist m 

 material treated for some time with caustic potash. 



Max Schultze, Targioni-Tozzetti ('70) and Kmer\ were all ot 



