98 THE AMERICAN MONTHLY [May, 



times as many as in the case of the desmids. That alone is enough to 

 frighten the beginner, especially if he has already attacked the desmids 

 and discovered that their classification was quite a study. But a glance 

 through the list shows that a number of the genera are fossil and others 

 marine. He rules out then, we will say, the fossil forms, and at once 

 cuts the number of genera down to about 90, and then he throws out 

 all those that are exclusively marine, or marine and fossil, and again 

 the number goes down rapidly — to about 50. Here, then, are 50 

 or more genera which the student who searches the fresh-water 

 ponds and ditches need not bother his head about, for he will not find 

 one of them. And yet I venture to say that almost every beginner 

 goes asti'ay on this point. Why ? Because he has a slide or two of 

 diatoms of Isthmia^ or Arachnoidisctis or Triccratluin. and he starts 

 ofi' for his pool expecting to light on some of these. Yet he may 

 search all his life without finding one of them in fresh water. The 

 first thing then for him to learn is this : Which are the fresh water 

 forms ? 



Well, let us look at the list and familiarize ourselves with their names. 

 Efiithemia^ Etifiotia^ Himaiitiditim^ Meridion^ Diatoma, Aster i- 

 onella^ Fragilaria^ Dctitictila^ Odontidium^ CyclotcUa^ ]\lelosira^ 

 Stephanodiscris^ Canipylodiscus^ Nitzschia^ Cy?natople?ira^ Suri- 

 rella^ Syncdra^ Amphiplcura^ Cocconeis^ Achna?ithcs, Cocconeina^ 

 Gomphoncnia^ Navicula^ Plcurosignia^ PiJinularia^ Stauroneis^ 

 Amphora^ Petracyclus^ Tabellaria^ Tci-psincc^ Mastogloia^ ScJiizo- 

 neffia., Encyo7icina^ Frustulia. 



There are 9 others, but we may omit them as being unusual or synony- 

 mous, namely: Oncosphcnia^ SpJiviiosira^ Diadesniis^ Perizonium^ 

 Sphcnella { = Gompho?ici/ia) , Ach/iauthidium { = Ach?ianihes) , 

 Cymhella { =Cocconcma)^ Staurosigtna { = Stauroneis) ^ Colleto- 

 nerna ( = Schizonc7ua) . And we have as the result but 34 genera to 

 distinguish ; quite a practicable mmiber as compared with over 100. 



And further, four of these may be recognized by the gelatinous frond 

 in which they are imbedded, namely: Alastogloia^ Frristulla^ Schizo- 

 nctna^ and Encyonema. If the frond has little cushions for the frus- 

 tules it is the first ; if it is shapeless, the second ; tubular, the third or 

 fourth, and if the valve is bowed or bent it is the latter. 



Again, of the 30 genera left, which have no frond, 3 more may be 

 quickly disposed of, namely, Tetracychis^ Pabellaria^ and Perpsinoe ; 

 for they all have heavy dark lines on the connected frustules. In 

 Terpsince the lines have heads resembling notes in music. In both 

 Tetracychis and Pabcllaria the lines run lengthwise of the filament ; 

 but Petracyclus has the frustules connected all along the sides, Tabel- 

 laria usually only at the corners, forming a zigzag chain. It is true this 

 is the description of thejO-ow/ view, but that is what the student will be 

 likely to see, as they usually form a series. 



Thus we have but 27 genera to study, which are divided into two 

 main classes, distinguished by the presence or absence of a nodule or 

 knob in the centre of the raphe or middle line of the valve. Let us take, 

 first, the division with the nodule, as it is the smaller. Here are 9 

 genera. Look at your diatom, therefore, and if it has the nodule it is 

 one of these : Coccoiieis^ Achnanthes^ Cocconema^ Goniphone7na ^ 

 Navicular Pleurosig?na^ Plnnularia^ Stauroneis, AmpJiora. And 



