304 H. MORLAND ON DIATOM STRUCTURE. 



that the normal structure of diatom valves consists of an inner 

 layer, more or less perforated, strengthened by an outer frame of 

 tubes more or less hexagonal, and that this upper framework of 

 tubes may be found in various stages, either very fully developed 

 (with even an exterior layer somewhat similar to the interior) or 

 slightly wanting, or reduced to spines only, or even entirely 

 wanting. If I judge Dr. Van Heurck rightly, he takes a highly 

 developed structure as his type. 



On the other hand, I would prefer taking a very simple structure 

 as my type, and working upwards, and this I would do because I 

 come across structures which cannot be explained upon the theory 

 of working downwards. 



In my view the simplest structure would be well represented by 

 taking a piece of perforated sheet zinc ; this I consider to be the 

 structure of the Pleurosigmce, with a sigmoid cleft down the centre, 

 strengthened with a nodule at each end, and a larger one in the 

 centre, this cleft having thickened sides and being called the "raphe." 



A little higher structure is that of the Stictodisci and some of the 

 Triceratice, where the perforations have thickened borders, some- 

 times further strengthened with minute veins, like those of a leaf, 

 running from hole to hole as can be seen in the valves of 

 Stictodiscus Jeremianus. I have a slide of what I believe to be 

 Triceratium venulosum, in which I have mounted a frustule on 

 edge ; by careful focussing perforations can be seen running 

 through the substance of the valves. I have also what may be 

 called a skeleton valve of some species of Auliscus, selected from a 

 deposit which I had to subject to very severe treatment with caustic 

 potash in order to disintegrate it ; it appears as if the thinner 

 portions of the valve had been dissolved by the potash, but that the 

 thick veins or ribs, having more substance in them, had been able 

 to resist the action of the potash to a certain extent. 



I will next take the case of what I consider to be a highly de- 

 veloped structure, viz., that of AuJacodiscus margaritaceus. This 

 was one of the forms of which I did manage to cut some sections, 

 but not to my satisfaction ; however, I mounted a section with the 

 cut edge in full view, and examined it under my microscope. I 

 could see that the section had thin lines running across, and at first 

 I thought it was a case of simple structure where the valve was 

 strengthened by means of extra depth or thickness. However, 

 more careful examination of perfect valves and fragments of 



