152 SILVA OF NORTH AMERICA. CONIFERS. 
divided by shallow fissures into broad flat ridges, which separate on the surface into long thin closely 
appressed fibrous scales. The branchlets are slender, light green when they first appear, hght red- 
brown and rather lustrous during their first winter, and darker during the following year, when the 
thin bark separates irregularly into fibrous scales. The deciduous lateral branchlets are three or four 
inches in length, and spread at right angles to the branch, or in the form with acicular leaves they are 
pendulous or erect, and often six or seven inches long.’ The leaves on the distichously spreading 
branchlets are linear-lanceolate, apiculate, from one half to three quarters of an inch in length, 
about one twelfth of an inch in width, and light bright yellow-green on both surfaces, or, especially on 
trees growing toward the southwestern limits of the range of the species, silvery white below; and on 
the form with pendulous or erect branchlets they are compressed, long-pointed, keeled and stomatiferous 
below, concave above, more or less spreading at the free apex, and about half an inch in length; in the 
autumn the branchlets with their leaves turn dull orange-brown before falling. The panicles of 
staminate flowers are four or five inches long and from an inch and a half to two inches wide, with 
slender light red-brown stems; the flower-buds are obovate, nearly an eighth of an inch in length, pale 
silvery gray during the winter, and purple when the flowers expand in the spring. The cones, which 
are usually produced in pairs at the extremity of the branch, or are irregularly scattered along it for 
several inches, are nearly globose, or obovate, rugose, usually about an inch in diameter, and generally 
destitute of mucros. The seeds with their wings are about a quarter of an inch long, and nearly an 
eighth of an inch wide, but vary considerably in size. 
this downward and upward growth several times within a distance 
of six or eight feet ; and at each point where the root approaches 
By the 
second method knees are produced by the more active growth of the 
the surface of the soil a knee grows from its upper side. 
upper surface of old roots of submerged trees at certain definite 
points. When the ascending and descending parts of the roots are 
close together they become, with increased diameter, united in the 
formation of the knees ; several knees which began their growth 
near together often become consolidated into one; and the root 
between the tree and the knee is smaller than it is beyond the knee, 
at the base of which a cluster of roots is frequently developed, later 
becoming consolidated with it. 
The most usually accepted belief with regard to the functions of 
these root-developments is that they serve to aerate the submerged 
roots, which, without their aid, would be entirely deprived of air. 
(See Berkeley, Gard. Chron. 1857, 549, f.— Shaler, Mem. Mus. 
Comp. Zoil. xvi. No. 1 [Notes on Taxodium distichum].— W. P. 
Wilson, Proc. Phil. Acad. 1889, 67; Forest Leaves, ii. 110, f. — 
Lotsy, Studies Biol. Lab. Johns Hopkins Univ. v. 269, t. 17, 18.) 
The hypothesis has also been advanced that the function of the 
knees is mechanical, aiding the roots to anchor the tree in the soft 
muddy ground in which it grows. (See Lamborn, Garden and For- 
est, iii. 21, f. 4. —Shaler, Garden and Forest, iii. 57.) 
Cypress knees, which are frequently hollow in old age, consist of 
soft spongy pale fibres, covered with thin red-brown scaly bark, 
and are exceedingly light, the average of four determinations, two 
from the top of a large knee and two from the base of the interior 
of the same knee, giving a specific gravity of only 0.2303. They 
have been occasionally manufactured into razor strops, which, how- 
ever, are soon ruined unless preserved from dust, as this adheres 
to the soft wood and becomes imbedded in the grain. (See Gar- 
den and Forest, i. 480.) 
1 No one unfamiliar with the fact that branches of the two forms 
occasionally appear on the same individual would imagine that the 
Cypress-trees with erect or pendulous thread-like branchlets and 
closely appressed acerose leaves belong to the same species as those 
with spreading distichous branchlets and flat leaves. The acerose 
form, which, so far as I have been able to observe, is not uncom- 
mon in South Carolina, in northern Florida, and in the neighbor- 
hood of Mobile, Alabama, but does attain a large size, appears to 
have been first noticed by Nuttall ; it has long been an inhabitant 
of the gardens of the eastern United States and Europe, and is 
generally cultivated as Glyptostrobus pendulus, and believed to be a 
native of China. The synonymy of this form is :— 
Taxodium distichum, var. imbricarium. 
Cupressus disticha, B imbricaria, Nuttall, Gen. ii. 224 (1818) ; 
Trans. Am. Phil. Soc. u. ser. v. 163. — Elliott, Sk. ii. 643. — 
Croom, Am. Jour. Sci. xxviii. 166. 
Taxodium microphyllum, Brongniart, Ann. Sci. Nat. xxx. 182 
(1833). — Endlicher, Syn. Conif. 68. 
Taxodium ascendens, Brongniart, /. c. (1833). — Endlicher, J. c. 
69. 
Taxodium distichum Sinense pendulum, Loudon, Arb. Brit. iv. 
2482 (1838). 
Taxodium Sinense, y pendulum, Forbes, Pinetum Woburn. 180 
(1839). 
Schubertia disticha, B, Spach, Hist. Vég. xi. 349 (1842). 
Schubertia disticha, y, Spach, l. c. 350 (1842). 
Glyptostrobus pendulus, Endlicher, J. c. 71 (1847). — Lindley & 
Gordon, Jour. Hort. Soc. Lond. v. 208. — Knight, Syn. Conif. 21.'— 
Carriére, Traité Conif. 152. — Bot. Mag. xcii. t. 5603. — Hoopes, 
Evergreens, 369, f. 59, 60. 
Taxodium Sinense, Gordon, Pinetum, 309 (1858). 
Taxodium distichum pendulum, Carriere, 1. c. ed. 2, 182 
(1867). — Veitch, Man. Conif. 215. — Beissner, Handb. Nadelh. 
152. — Hansen, Jour. R. Hort. Soc. xiv. 304 (Pinetum Danicum). 
Elliott (/. c.) describes this as a small tree growing in Pine-bar- 
ren ponds, and producing more numerous knees than individuals of 
the more common form. He noticed that the branchlets on upper 
branches were often distichous. 
It is possible that it was this tree, which was described by Aiton 
(Hort. Kew. iii. 372 [1789]) and figured by Loudon (Arb. Brit. iv. 
