June 1, 1868.] 



HARDWICKE'S SCIENCE-GOSSIP. 



131 



in thickness, the glutinous ones rather more. It 

 has been generally stated that each of these threads 

 is composed of a number of minute filaments, 

 depending upon the number of tubules in the 

 mammulae from which they were spun. I cannot 

 help thinking, however, that such is not the case, 

 and it certainly appears to me to be more natural 

 and more in accordance with the facts, as I have 

 observed them, to suppose that the secretion from 

 the numerous tubules coalesces on the surface of 

 the mammulae to form a single drop of the viscus 

 fluid, from which drop a homogeneous, and not a 

 compound, thread is spun. 



Fig. 124. Silk glands and tubules. 



It seems to me, if the thread was so complex as 

 has been supposed, that some observer would have 

 succeeded in isolating the various filaments ; but 

 as far. as I am aware this has never been done. 

 Their existence, therefore, is certainly not proven, 

 and is, I think, an unnecessary assumption. 



A single web often contains more than three 

 hundred longitudinal and as many transverse 

 threads : so elaborate is the seemingly simple net 

 a spider sets to catcli its prey ! 



T. Graham Ponton, E.Z.S. 



"PERLEY'S MEADOW" DEPOSIT. 



rpHE following forms of Diatoms have been 

 -*- detected in the deposit from Perley's Meadow, 

 South Bridgton, Maine, which through the kind- 

 ness of the Rev. E. C. Bolles, of Portland, has been 

 sent to the Editor of this journal for distribution 

 amongst its microscopical readers. It contains, as 

 will be observed, some very interesting species. 



Navicula rostellum (Smith) ; Navicula apiculata 

 (Gregory). Valve elliptic oblong, apices produced, 

 nipple-like, transverse stria} fine, reaching median 

 line, curved striae crossing in both directions towards 

 the median line, coarser than the transverse (fig 125). 



The arrangement of striae seems peculiar to this 

 species. I have never seen any other form of the 

 naviculoid group showing this very peculiar stria- 

 tum ; in some of the discord forms a similar striatum 

 may be detected. Dr. Lewis refers this form to 

 Navicula placenta of Ehrenberg ; I prefer retain- 

 ing Professor Smith's specific name, as it is very 

 doubtful whether Ehrenberg's form is really the 

 above species, the markings being far beyond the 

 resolving power of his instrument, and his outline 

 would suit many other species. 



Fig. 125. 

 Navicula rustellum x 6c0. 



Fig. 126. 

 Navicula gust rum x 400. 



Navicula gastrum (Ehr. ?) Valve elliptic, apices 

 slightly produced, median blank space small but 

 distinct, striae distinct, slightly radiant (fig. 12G). I 

 have followed Dr. Lewis in adopting Ehrenberg's 

 specific name, but have much hesitation in doing so; 

 it resembles N. firma in the median blank space 

 and intra-marginal lines, but differs from that species 

 in the slightly radiant striation. 



Navicula Americana (Ehr.) Valve linear, ends 

 rounded slightly, constricted towards the centre, 

 median liue and central nodule conspicuous, stria; 

 distinct, about 40 in "001, not reaching the median 

 line (fig. 127, x 400). This form is very rare in the 

 Perley's Meadow deposit, but is not uncommon in a 

 deposit from West Point, New York, in which it 

 occurs mixed with Navicula Bacillum, a species it 

 much resembles ; it may possibly be only a sporangial 

 form of it. Ehrenberg's figure in the Microgeologie 

 is very good, and leaves no doubt of the identity 

 of the Perley's Meadow form and the species he 

 figures. 



Navicula scut divides (?). Valve minute nearly 

 orbicular, median line more or less excentric, central 

 nodule conspicuous, striae faint, about 50 in '001 (fig. 

 12S). I refer this species to Navicula scutelloides 

 with much hesitation, but think it better to do so 

 until specimens from other localities enable us to 

 decide whether it is an effete state of Navicula 

 scutelloides or a new species. It is not uncommon 

 in the lightest portions of the deposit. 



Navicula cocconeiformis (Gregory) ~? N. Ca- 

 rassius (Ehr.) Valve elliptical, apices slightly 

 produced, striae obscure, median line and central 

 nodule distinct (fig. 129, x 400). 



Pinnularia cardinalis will be found in the heavy 

 densities, and is generally very fine ; the costae are 

 somewhat closer than those on the species found 



