520 Brebner. — Ok the Anatomy of 
protostelic to be used as previously. When distinctly present, 
and developed from the procambium-strand along with the 
other elements, the ‘ pericycle 1 would be included in this or 
any other type of stele 1 . It is difficult to find a suitable 
word to express the idea of scattered vascular strands as 
typically found in Monocotyledons. As indicating irre- 
gular arrangement, atactostele may serve for this type, 
whether the individual bundles lack or possess a cambium. 
An atactostele therefore consists of a number of more or less 
irregularly arranged vascular bundles together with the ground 
tissue in which they are imbedded. It is now known that 
a large number of Dicotyledonous plants have scattered 
vascular bundles, and, as already indicated, they would be 
said to be atactostelic. In certain of them the vascular 
bundles are all of one type, and the stele in that case would 
be said to be homodesmic . In other cases, however, there are 
‘ medullary 5 strands which consist of phloem only (e. g. in 
Gentianaceae and Melastomaceae), and then the steles might 
be called heterodesmic. Cucurbita may be taken as an 
example of atactostely among the Dicotyledons and Habenaria 
bifida of eustely among Monocotyledons. The term solenostele 
as defined by Gwynne- Vaughan 2 will be retained for the 
continuous amphiphloic tube with widely separated leaf-gaps 3 . 
Another and closely allied type of { siphonostele 4 ,’ in which 
a tubular network of vascular tissue results from the over- 
lapping of the leaf-gaps, seems well described by the term 
dictyostele. This is a common condition found in Ferns, and 
well illustrated by the vascular system of Aspidium , with 
1 It would perhaps be better to fix the phloem as the outermost limit of the 
stele on account of the want of constancy in origin of the ‘ pericycle,’ but in the 
meantime, at any rate, the other definition is adopted. 
2 Gwynne- Vaughan, Observations on the Anatomy of Solenostelic Ferns, 
Annals of Botany, xv, p. 73, 1901. 
3 Neglecting the internal endodermis as a morphological criterion, it might be 
well to include it and the rest of the central parenchyma in the definition of the 
solenostele , but at present Van Tieghem’s term, as revived and defined by Gwynne- 
Vaughan, will be retained. 
4 Jeffrey applies this term to the eustele of Dicotyledons and the solenostele of 
certain Vascular Cryptogams and also to the network-like stele of others. 
