Feb. 1. 1894.J THE TROPICAL 
AGRICULTURIST, 
A)i(fh/sis of Plant Ashf-i. 
per cent. 
Moisture ... ... ... fi'lO 
Carbonaceous matter ... .. •52 
Oxides of iron and alumina ... 4'03 
*Lime ... ... ... 24-50 
Magnesia ... ... ... 4-18 
t Potash .. ... ... 18-70 
Soda ... .. ... •SS 
Pliosplioric acid ... ... trace 
Sulpliuric acid .-. . ... 2-42 
Chlorine ... ... ... .3-40 
Carbonic acid &c. ... ... 32-76 
Insoluble siliceous matter ... 3-04 
100-00 
... 43-75 
... 29-01 
remarkal)le 
* Equal to calcium carbonate 
Equal to potassium carbonate 
Tiie following is tiie analysis of 
saniide of woihI aslies, which exhibits [)otasli at 
about its minimum, and lime at its maximum. 
It is the ash of the tree known by the Sinhalese as 
the kumbuk, which is one of the genus Termiiialia. 
On siftinj.; the .sample through a sieve having 
900 meshes to the square iucli, 1-17 per cent, of 
quartz sand was separated. The ash, freed from 
the small quantity of qiuirt/, had the following 
composition : — 
Anrih/iis 0/ Wood Ashes of the Kiiiiitntk Tree. 
per cent_ 
Moisture ... ... ... -38 
-; In.sohible siliceous matter ... 2-35 
Silica soluble in acid ... ... "26 
Oxide of iron and alumina ... -75 
*Lime ... .. 56-00 
Magnesia ... ... ... '37 
Potash ... ... ... -54 
, Phosjthoric acid ... ... trace 
Carhonic acid iiud other constituents 39-35 
Equal to calcium carbonate 
10000 
100-00 
A portion of the lime was in the caustic state, 
wliicli accounts f(n- the fact that it contains as 
much lime as is present in chemically pure car- 
bonate of lime. 
i\ venerable specimen of this tree grows at 
Mutwal, {'olombo, which the late Mr. W. Fer- 
guson measured in the year 1850, and found the 
gii'lh close to the earth to be 45 feet, and 21 feet 
at 12 feet above the ground Mr. Ferguson states 
that lie had his lirst view of the tree fiom the 
s(M n(iar Negomb(>, about twenty miles distant. 
The calcareous nature of the ashes of the kumbuk 
tree is well-known to the natives of Ceylon. 
AsJies xtf Marine Plants. 
i (,)u seii, coasts,, whurii iiea weed abounds, it ha« 
bnan largely used tor manuring the land bordering 
ou the .sea -. but the large jicreciilage of moisture 
in mnriiie plants prevents it< use 011 land, at any 
grt-at dislance from the sea. 1 am unalile to 
qui'ti- the analyses of any cut or <lrift weeds found 
till- I'eylon I'lmsts ; Imt \-ery elaborate analyses 
of sea weeds, and es])ecially of the ash of sea 
weeds, kiiown as kelp in Scotland and Ireland, 
and varec in France, have been made in connection 
with the iodine industry. 'J'hese may be found 
in Kichardson an<l Watts' (!hemical technology. 
Viewed as a potash manure, it will be sufticient to 
([note the analyses of the late Profes.sor Anderson 
of Glasgow, of the mixed cut and drift weeds in 
the state in which they are actually used, i.e., 
w ith small shells and marine animals adhering 
to the plants, which increase the amount oi 
]>liosplioric acid and nitrogen. 
Avcrae/r Conipositioa 0/ Miined Sea Weed. 
(Anderson.) 
per cent. 
Water ... ... ... 80-44 
* Albuminous compounds... ... 2*85 
Fibre &c. ... ... ... 6-40 
tAsh ... .. ... 10-31 
* Containing nitrogen 
t Containing ])otash 
Phosphoric acid 
Fidl Aiudijsis of the Ash. 
Peroxide of iron 
Lime 
Magnesia 
Potash 
Potassium chloride 
Potassium iodide 
Sodium chloride 
Phosphoric acid 
Sulphuric acid 
Carbonic aciil 
Silicic aci(I ... 
100-00 
•45 
1-95 
•47 
2 - .3,5 
18-15 
648 
1277 
910 
1-68 
22-08 
4-59 
6-22 
13-58 
3- 00 
100 00 
The following are Kicliard.son's analj'ses of 
the sulphates of potash of commerce, prepared from 
solutions of kelp : — 
Analyses of Commerckd Stdphatc of Potash. 
(PaCH.VRD.SON. ) 
Irish. 
Sulphate of potash 
77-43 
75-28 
Sulphate of soda 
21-31 
20-89 
Sulphate of lime 
_ 
•80 
Sulphate of iron 
trace 
( 'hloride of sodium 
•76 
•54 
Insoluble matter 
trace 
104 
Moisture 
•59 
1-55 
ItJOtX) 
100-00 
Scotch 
83-00 
14-89 
-67 
1-44 
100-00 
I might hej-e give itlie anajys^is <>f a SHUtpl«^ of 
volcanic ashes or deposit brought to .me from 
.Mount Vesuvius. For a natiiral pnxluot it i> 
n-iiiarkably ri<-li in |M>tash, and if it could be 
