Skip to main content

Full text of "The School of Mines Quarterly: A Journal of Applied Science"

See other formats


This is a digital copy of a book that was preserved for generations on library shelves before it was carefully scanned by Google as part of a project 

to make the world's books discoverable online. 

It has survived long enough for the copyright to expire and the book to enter the public domain. A public domain book is one that was never subject 

to copyright or whose legal copyright term has expired. Whether a book is in the public domain may vary country to country. Public domain books 

are our gateways to the past, representing a wealth of history, culture and knowledge that's often difficult to discover. 

Marks, notations and other maiginalia present in the original volume will appear in this file - a reminder of this book's long journey from the 

publisher to a library and finally to you. 

Usage guidelines 

Google is proud to partner with libraries to digitize public domain materials and make them widely accessible. Public domain books belong to the 
public and we are merely their custodians. Nevertheless, this work is expensive, so in order to keep providing tliis resource, we liave taken steps to 
prevent abuse by commercial parties, including placing technical restrictions on automated querying. 
We also ask that you: 

+ Make non-commercial use of the files We designed Google Book Search for use by individuals, and we request that you use these files for 
personal, non-commercial purposes. 

+ Refrain fivm automated querying Do not send automated queries of any sort to Google's system: If you are conducting research on machine 
translation, optical character recognition or other areas where access to a large amount of text is helpful, please contact us. We encourage the 
use of public domain materials for these purposes and may be able to help. 

+ Maintain attributionTht GoogXt "watermark" you see on each file is essential for in forming people about this project and helping them find 
additional materials through Google Book Search. Please do not remove it. 

+ Keep it legal Whatever your use, remember that you are responsible for ensuring that what you are doing is legal. Do not assume that just 
because we believe a book is in the public domain for users in the United States, that the work is also in the public domain for users in other 
countries. Whether a book is still in copyright varies from country to country, and we can't offer guidance on whether any specific use of 
any specific book is allowed. Please do not assume that a book's appearance in Google Book Search means it can be used in any manner 
anywhere in the world. Copyright infringement liabili^ can be quite severe. 

About Google Book Search 

Google's mission is to organize the world's information and to make it universally accessible and useful. Google Book Search helps readers 
discover the world's books while helping authors and publishers reach new audiences. You can search through the full text of this book on the web 

at |http: //books .google .com/I 





Vol. XXII. 

November, 1900, to July, 1901 



r i \ 

/6^ 2L , / / ^ 

, ■*. 





A. J. MOSES, Prof, of Mineralogy. 
J. P. KEMP, Prof, of Geology. 
R. PEELE, Adj. Prof, of Mining. 

Adj. Prof, of Architecture. 
R. E. MAYER, Instructor in Drawing. 
E. WALLER, Analytical Chemist. 


Lecturer in Metallurgy. 
I. H. WOOLSON, Instructor in 

Mechanical Engineering. 
S. O. MILLER. Assistant in 

Mechanical Engineering. 

Manaffinff Editor, R. E. MAYER. 
Business Manager, S. O. MILLER. 

i^Rtss or 





November, 1900, to July, zgoz. 

All Abstracts ar« Printed in Italics, snd thoss rslsting to Chsmicsl Anslysis appear 

only under the heading Analytical Chemistry. 

Alternating Current Laboratory Experiments and Commercial Tests. By Fitz- 

hugh Townsend i6 

Alternating Currents, As3rmmetrical, by Means of Electrolytic Polarization, On 

the Production of. Pt. II. By W. L. Hildburgh I 

Alumni and University News. 

Architecture 1 15, 388 

Astronomy and Geodesy 116 

Chemistry io8, 384 

Civil Engineering 114, 388 

Geodesy and Astronomy 116 

Geology 117 

Metallurgy 115 

Mining 113 

Analysis of Slags and Cinders. By C. H. Jotiet 64, 140 

Analytical Chemistry, Abstracts. By Elwyn Waller. 

Ag loi 

Alkaline Carbonates 92 

Alumina and Ferric Oxide in Natural Phosphates 93 

Analysis 95 

" of Uranium and Vanadium Ores 453 

Arsenic in Alloys . , . 98 

•* in Insecticides 98 

** Marsh Test 454 

it t* it £ffcts 0/ Selenium and Tellurium 454 

As loi 

Au Id 

Basic Slag 105 

Bismuth Ccbalticyanide 100 

** Titration 100 

Boric Acid in Boracite 104 

Bromide in Presence of Chloride and Iodide 103 

Calcium Carbide 106 

** Oxalate 'Converiion to Sulphate 92 

Carbon in Iron and Steel Re-use of Copper Solution 103 


Chlorine in Bleaching Powder 103 

Chromium in Steel, 93 

Cobalt Estimation 96 

Cobalticyanides 91 

Colorimetric for Vanadium 102 

Commercial Analysis of Copper loo 

•* Calcium Carbide 106 

Coppers loi 

Separation of Platinum and Gold 99 

Copper and Chilled Slag 102 

" in Cyanide Solutions lOO 

Cu and Ag loo 

Delicate Reaction for Manganese 453 

Detection of Nitrogen, Jacobsen^ s Reaction 105 

Determination of Alumina 93 

** of Zinc 94 

Determining Thallium as Sulphate 102 

Dissolved Oxygen in Water 455 

Electrolytic Separation of Manganese 94 

Errors in Volumetric for Boric Acid. 104 

Estimating Lead by Electrolysis 99 

Estimation of Manganese in Steels 94 

of Mercurous Mercury ♦. 455 

of Zinc 94 

Examination of Coals for Gas Manufacture 107 

Fe loi 

Ferrocyanide Determination 103 

Ferro- Silicon and Silico Spiegel. 452 

Free Lime in Portland Cement 92 

Iron in Puddle Slag 93 

Kjeldahl Method Applied to Nitrates 106 

** Process 105 

Lead in Potable Waters 454 

Lime in Presence of Iron and Alumina 92 





Manganese. 452 

as Phosphate 454 

in Manufactured Irons 94 

or Cobalt as Phosphate 454 

Mercury Cyanide 99 

Metallic Sodium as a Blowpipe Reagent 452 

Method. 93 

Methods of Estimating Fluorine 104 

Molybdenum in Alloys 97 

in Iron and Steely etc 97 

in Irons 453 


Naphthalene in Coal Gas 107 

New Coal Calorimeter IC7 

** Indicator 91 

** ** Luteol 91 

Niand Co loi 

Nickel in Nickel Steel 95 

Nitrates f Commercial 106 

Nitric Acid. 105 

** «' and Mixed Acid 106 

** ** in Combination with Alkalies 455 

Nitrite Determination 105 

Oxygen in Commercial Copper 455 


J^S loi 

Pettmkofct^ s Test for Carbon Dioxide 103 

Phosphoric Acid. 104 

Phosphorus in Steel 104 

Potassium Estimation 92 

** Xanthate 91 

Preparing Normal^ Semi-normal ^ etc, , Acid Solutions 452 

Qualitative for Boric Acid, 104 

** for Gold in Ores 98 

** for Nickel in Presence of Cobalt 454 

** Tests for Alkaline Earths 92 

Red Lead. 99 

Residual Cu lOI 

S lOl 

Sb loi 

Separation of Bismuth from Lead 99 

of Nickel and Iron 96 

of Tungsten from Molybdenum 97 

•* of Tungstic Acid, 453 

Silicon 452 

Small Quantities of Cobalt in Presence of Nickel.,,, 96 

Standarditing Acidimetric Solution 91 

Titrating Persulphates 105 

Tungsten in Ores^ etc 98 

" in Steels 452 

Volutnetric Determination of Hydrogen 102 

for Cerium I02 

for Gold and Platinum 99 

Water which Attacks Lead, 455 

Zinc- Electrolytic 95 

Andrews and Murdoch. Current Mining Decisions, Abstracts 456 

Antip3rrin and its Derivatives. By D. C. Ecclcs 259 

Assay of the Zincy Precipitates Obtained in the Cyanide Process, Notes on the. 

By C. H. Fulton and C. H. Crawford 153 

Asymmetrical Alternating Currents, see Alternating Currents. 

Bergen, Charles Hill 382 

Book Reviews : 

Chemical Analysis of Iron. By Andrew Alex. Blair 476 

Chemistry — Its Evolution and Achievements. By Ferdinand G. Wiech- 

mann 255 

Coal and Metal Miners* Pocketbook of Principles, Rules, Formulas and 

Tables 254 

Kurzes Lehrbuch der Analytischen Chemie. By Dr. F. P. Treadwell.... 390 

Metallurgy of Gold. By M. Eissler 253 

Qualitative Chemical Analysis. By Albert B. Prescott and Otis C. Johnson. 477 

School Chemistry. By John Waddell 256 

Text book of Important Minerals and Rocks. By S. K. Tillman 389 

Topographic Surveying. By Herbert M. Wilson 257 

Tunneling. A Practical Treadse. By Charles Prelini, with Additions 

by Charles S. Hill 475 

Cadmium, The Quantitative Determination of. By E. H. Miller and R. W. Page. 391 
Calcite, A List of the Crystal Forms of, with their Interfacial Angles. By A. Y, 

Rogers 429 

Cathcart, William Ledyard. The Powering of Ships 163 

Chemical Analyses of Rocks, The Re- calculation of. By J. F. Kemp 75 

Cinders and Slags, Analysis of. By C. H. JoUet 64, 140 

Gassifier, Laboratory. By H. S. Munroe 303, 449 

Conunercial "Dry Cells," see **Dry Cells," Commercial. 


Commercial Tests and Alternating Current Laboratory Experiments. By Fitz- 

hugh Townsend i6 

Correction. — Thomas Egleston Memorial. By A. J. Moses 451 

Crawford, Charles H. and Fulton, C. H. Notes on the Assay of the Zincy 

Precipitates Obtained in the Cyanide Process 153 

Crocker, Francis B. Electrochemistry and Electrometallurgy. Pt. 1 1 19 

Crystal Forms of Calcite with their Interfacial Angles, A List of the. By A. F. 

Rogers 429 

Current Mining Decisions^ Abstracts. By Andrews and Murdoch 45 6 

Currents, see Alternating Currents. 

Cyanide Process, Notes on the Assay of the Zincy Precipitates Obtained in the. 

By C. H. Fulton and C. H. Crawford 153 

Cyclic Analysis of Heat Engines, A Method of. By C. E. Lucke 223, 329, 411 

Derleth, Charles, Jr. Herman Andreas Loos 89 

Discussion of Paper of Mr. Scherr on Redaction -Roasting. Its Value for Ar- 
senic Expulsion from Copper Ores and Mattes. By H. M. Howe 381 

"Dry Cells," Commercial, for Secondary Standards of E. M. F., Experiments 

with. By H. St. J. Hyde 366 

ECCLES, David C. Antipyrin and its Derivatives 259 

Egleston, Thomas, Memorial. A Correction. By A. J. Moses 451 

Electrochemistry and Electrometallurgy. Pt. L By F. B. Crocker 119 

Electrolytic Polarization, On the Production of Asymmetrical Alternating Cur- 
rents by means of. Pt. II. By W. L. Hildbui^h I 

Electrometallurgy and Electrochemistry. Pt. I. By F. B. Crocker 119 

Experiments with Commercial ** Dry Cells" for Secondary Standards of E. M. 

F. By H. St. J. Hyde 366 

Fauna, Siluric, near Batesville, Arkansas. Pt. I. By Gilbert Van Ingen 318 

Fulton, Charles H. and Crawford, C. H. Notes on the Assay of the 

Zincy Precipitates Obtained in the Cyanide Process 153 

Heat Engines, A Method of Cyclic Analysis of. By C. E. Lucke 223, 329, 411 

Hildburgh, Walter Leo. On the Production of Asymmetrical Alternating 

Currents by Means of Electrolytic Polarization . Pt. II. I 

Howe, H. M. Discussion of Paper of Mr. Scherr on Reduction- Roasting. Its 

Value for Arsenic Expulsion from Copper Ores and Mattes 38 1 

Hyde, Henry St. John. Experiments with Commercial ^^ Dry Cells'^ for 

Secondary Standards of E, M. F. 3^ 

JouET, C. H. The Analysis of Slags and Cinders 64, 140 

Kemp, J. F. The Re- calculation of the Chemical Analyses of Rocks 75 

Laboratory Classifier. By H. S. Munroe 303, 449 

** Slime Table. By H. S. Munroe 306 

List of the Crystal Forms of Calcite with their Interfacial Angles. By A. F. Rogers. 429 

Loos, Herman Andreas.* By Charles Derleth, Jr , 89 

LucKE, Charles E. A Method of Cyclic Analysis of Heat Engines 223, 329, 411 

Method of Cyclic Analysis of Heat Engines. By C. E. Lucke 223, 329, 411 

Miller, Edmund H., and Page, R. W. The Quantitative Determination 

of Cadmium 391 

Mining Decisions^ Current^ Abstracts. By Andrews and Murdoch 456 

Moses, Alfred J. A Correction. — Thomas Egleston Memorial. 451 

Munroe, Henry S. A Laboratory Classifier.... 303, 449 

** ** «' Slime Table 306 

Murdoch and Andrews. See Andrews and Murdoch 

Newi^ND, D. H. The Serpentines of Manhatten Island and Jlcinity and 

their Accompanying Minerals 307, 399 

Notes on the Assay of the Zincy Precipitates Obtained in the Cyanide Process. 

ByC. H. Fulton and C. H. Crawford 153 


Bergen, Charles Hill 382 

Loos, Herman Andreas. By Charles Derleth, Jr 89 


On the Production of Asymmetrical Alternating Currents by Means of Electro- 
lytic Polarization. Pt. II. By W. L. HUdbui^gh I 

Page, Robert W. and Miller, E. H. The Qttantitatvve Determination of 

Cadmium 391 

Polarization, see Electrolytic Polarization. 

Powering of Ships. By W. L. Cathcart 163 

QuantitatiTe Determination of Cadmium. By E. H. Miller and R. W. Page 391 

Re-calculation of the Chemical Analyses of Rocks. By J. F. Kemp 75 

Reduction-Roasting, Discussion of Paper of Mr. Scherr on. By H. M. Howe... 381 

Rocks, The Re- calculation of the Chemical Analyses of.. ..By J. F. Kemp 75 

Rogers, Austin F. A List of the Crystal Forms of Calcite with their Inter- 
facial Angles 429 

Scherr, Mr., on Reduction- Roasting, Discussion of Paper of... .By H. M. Howe. 381 
Secondary Standards of E. M. F., Experiments with Commercial **Dry Cells" 

for. By H. St. J. Hyde 366 

Serpentines of Manhattan Island and Vicinity and their Accompanying Minerals. 

By D. H. Newland 307, 399 

Ships, Powering of. By W. L. Cathcart 163 

Siluric Fauna near Batesville, Arkansas. Pt. I. By Gilbert Van Ingen 318 

Slags and Cinders, Analysis of. By C. H. Joflet 64, 140 

Slime TaWe, Laboratory. By H. S. Munroe 306 

TOWNSEND, FlTZHUGH. Alternating Current Laboratory Experiments and 

Commercial Tests 16 

Van Ingen, Gilbert. The Siluric Fauna near Batesville, Arkansas. Pt. I... 318 

Waller, Elwyn. Analytical Chemistry, Abstracts 91, 452 

Zincy Precipitates Obtained in the Cyanide Process, Notes on the Assay of. By 

C. H. Fulton and C. H. Crawford 153 

Vol. XXII. No. i. 

NOVEMBER. 1900. 








A.J. MOSES, Prof, of Mineralogy. 
J. F. KKMP, Prof, of Geology. 
R. PBBi^E, Adj. Prof. Mining. 

Adj. Prof, of Architecture. 
R. B. MAYBR, Instructor in Drawing. 
£. V^ALLBR, Analytical Chemist. 


Lecturer in Metallurgy. 
I. H. WOOLSON, Instructor in 

Mechanical Engineering. 
8. O. MILLER. Assistant in 

Mechanical Engineering. 

Manarinff Editor, R. E. MAYER. 
Business Managrer, S. O. MILLER. 


On the Production of Asyminetrical Alternating Currents by Means 
of Electrolytic Polarisation. By Walter Leo Hildburgh» E.E., 

A.M I 

Alternating Current Laboratory Experiments and Commercial Tests. 

By Fitzhugh Townsend i6 

The Analysis of Slags and Cinders. By Cavalier H. Joiiet 64 

The Re-calculation of the Chemical Analyses of Rocks. By J. F. 

Kemp 75 

Herman Andreas Loos 89 

Abstracts in Analytical Chemistry. By Elwyn Waller 91 

Alun^ns and University News 108 



at the New York Post Office as Second Class Matter. 

All Reailttancas should be made payable to Order of *«The School of Mines Quarterly. 




MuDfactonn at r'tttSCCBSlT CTRCI For EtKik Drill. 
Uia ««ll-kni>ini WlVuOwd^ I O 1 CCLi BOd MUloc Work. 

Tool and Die Steel of the Beat QuaJity, SbeM Steel, 
SpringaadUachinetySted. Bicel for Bdge-Toola and 
Hardware of Bveiy Deacription. Polished Drill Rod! 
and Fine Wire. Steel Forgings, Coiled Springe, etc- 

CBICA.aO, TLTj. new YORK, N. V. 

JenkiDs Standard '96 Packing 

Wttigha 33 i per cent, less, 
does not rot, burn or blow 
out, and will last l9nger than 
other joint packings; there- 
fore the cheapest and best 
for all purposes. 
Gennlne Stamped Like Cut. Good Bnjfrs Compare Weights 

14,000,000 Barrels 



No other Cement Company can Mbow aacb a record. 

THE Lit RENCE CEDENT CO, Siles Office, Ho. 1 Bnulf ar, HEf TOEK. 



Portland Cement 


THE UWRENCE CEMENT CO. OF PENNt., Siegfried, Penna. 



Vol. XXII. NOVEMBER, 1900. No. i 


(Continued from page 382, Vol. XXI.) 

II. Single-Cell Rectifiers. 

Description. — A single- cell rectifier is a device containing essen- 
tially only one electrolytic element, which will asymmetrically 
aflfect an alternating current. If two electrodes differing in com- 
position be immersed in an electrolyte one electrode may be con- 
sidered as inert, since, even if it be ordinarily soluble in the elec- 
trolyte the presence of the second, and more soluble, electrode in 
the circuit will prevent its solution. 

Certain metals are ordinarily insoluble in certain acids, as copper 
in dilute H,SO^, their solution pressures in solutions of their own 
salts being less than that of hydrogen in a solution of its corre- 
sponding salt. If such a metal be used as an anode, with a plati- 
num cathode, it will dissolve freely and liberate a quantity of en- 
ergy which will be taken up in the production of hydrogen bubbles 
at the inert electrode. There will be a counter E.M.F. due to the 
excess energy necessary to liberate the hydrogen, and equal to the 
algebraic sum of the potential differences at the two electrodes. 
The more nearly equal those potential differences the lower the 
counter E.M.F. 

When the soluble metal acts as cathode the hydrogen previously 
deposited on the platinum will be taken up, and metal be deposited 



on the soluble electrode. When the hydrogen is all removed from 
the inert plate oxygen will be deposited, and the counter E.M.F. 
of the cell be equal to the sum of. the potential differences of the 
oxygen in H^SO^ and of the soluble metal in a solution of its sul- 
phate. Until the impressed E.M.F. is high enough to overcome the 
counter E.M.F. representing that sum no continuous current can 
flow, but only enough electrical energy to charge the inert plate 
sufficiently to produce a counter E.M.F. equal to the impressed. 
The cell gives in one direction a counter E.M.F. equal to the differ- 
ence between the potential differences of the soluble electrode 
and a substance deposited on the inert electrode, and in the oppo- 
site direction an E.M.F. equal to the sum of the potential differ- 
ences of the soluble electrode and of some second substance de- 
posited on thie inert electrode. 

The soluble electrode gives an E.M.F. fixed in value and direc- 
tion, and independent of any quantity of current passed into or 
out of it : its capacity may be therefore regarded as infinite. The 
potential difference at the inert electrode depends on the direction 
and value of the impressed E.M.F., and its capacity is limited. 
The theory of a rectifier of this kind is consequently the same as 
that of a two-cell rectifier in which one of the plates is of infinite 
area, and all deductions for the latter hold true for the former. 
The potential difference at the soluble electrode is the equivalent 
of the active E.M.F. in the two-cell rectifier, that at the inert elec- 
trode in the •' useless " direction the equivalent of the passive 
E.M.F. In the two-cell rectifier the active and passive E.M.F.'s 
are, for the best results, equal. In the single-cell rectifier there is 
no definite ratio between them, and the passive may be any num- 
ber of times greater than the active E.M.F.; the active E.M.F. 
may even be zero, as in some of the two-solution rectifiers. An 
electrolyte when used in the decomposition cell of a two-cell 
rectifier permits the economical use of an impressed E.M.F. 
just double that permissible when it is used in a single-cell rec- 
tifier, the active RM.F.'s in both cases being balanced; in the 
latter the sum of the active E.M.F. and the passive E.M.F. in the 
*' useless " direction is the decomposition value of the electrolyte, 
half the maximum counter E.M.F. obtainable in the former. 

To exactly balance the active E.M.F. and the passive E.M.F. 
in the " useful " direction alloys may be used for the soluble elec- 
trode, or the pressure at the inert electrode varied. The best 


method of balancing is by the use of an external source of E.M.F. 
to aid or oppose the active E.M.F. , according to whether the latter 
is too high or too low. When the potential difference at the solu- 
ble electrode is greater than that at the inert electrode in the 
" useful " direction the cell is an ordinary primary cell giving a 
continuous current ; when it is less the cell is " underbalanced." 

Construction, — As an electrolyte dilute sulphuric acid seems most 
suitable, with hydrogen, copper, silver, lead, mercury, etc., as sol- 
uble electrodes. With lead* and mercury f secondary actions 
occur which interfere seriously with their use in cells having a 
small inert electrode, though making it possible to obtain a very 
slight asymmetrical effect when the inert electrode is large. A 
cell consisting of two large plates, lead and platinum, in dilute 
HjSO^ gave a curve which showed a difference in the two leads (first 
and second half-waves), but practically no other indication of 
asymmetry. With a small platinum plate the A.E. was appar- 
ently about 6 per cent., but the curve as taken was very irregular 
and entirely untrustworthy, due to the continual changes going on 
at the surface of the lead. 

The following table shows the effect of some typical cells. Sev- 
eral of the curves are illustrated in Fig. 10. 

Table IV. 

Conditions in Single-Cell Recfifiers. 


.1465 sq. 
.0507 «♦ 
.1465 •* 

Arrangement Elzt. Res. « loo Ohmi . 

Imp. E.M.F. 






mm. R and 

<i II it 

large Pb electrode in HjSO^ 
.1 cu ** •* " *• 
II Cu ** ** " " 







.1465 " " ** •* 
.0507 ** " " •* 
.050^ *« «* ** •' 
.0507 " " •* *«• 
N X electrode in H,SO 

II pe ** ** " ** 
II zn ** ** ** '* 
1. H " '« " ** 
II H ** *' " ** 
4, N electrode in HNO3 





.998 (?) 


N " 

II 11 II 

II H 11 

large platinum in HNO, 
D electrode in HI 






" ZnSO^, D *♦ ** " 





and large rotating platinum in H^SO^ 



In the cell Cu — H,SO^ — Pt, equivalent to the cell Cu — CuOS^ 
- H,SO^ — Pt, the active E.M.F. is too small. The impressed 

* Sheldon and Waterman, Phy^ Rev,, Vol. IV., p. 324, Jan., 1897. 
tLowric-Hallc, Lond, Elec.^ Vol XXL, p. 157, June 8, 1888. 
t N = 6.41 sq. mm. Pt. 
g D = . 1465 tq. mm. Pt. 




Fio. lO. Current curves of single-cell rectifiers. 


E.M.F. used for testing it was slightly too high (Curve 20). In the 
cell H, — H,SO^ — Pt the E.M.F/s are always exactly balanced 
(Curves 24 and 25). In the cell Fe — H,SO^ — Pt, or Fe — FeSO^ 
— H,SO^ — Pt, the active E.M.F. is slightly too high, and a small 
continuous current flows. The current cell Zn — ZnSO^ — 
H,SO^ — Pt illustrates excellently (Curve 23) the effect of an 
active E.M.F. much too high, the apparent A.E. being .998 ; con- 
siderable continuous current flows and only the part of the curve 
above the true zero-line is affected, the curve corresponding to cal- 
culated curve C II, 

Aluminum differs from most other metals because of the pecu- 
liar resistance effects at its surface, but can be used as a soluble 
metal in a heated solution, or if the hydrate be continually re- 
moved ; the potential difference at the active electrode, is consid- 
erably higher than that at the passive, and the action consequently 
uneconomical. When the resistant-film effect is utilized the cur- 
rent is rectified in the opposite sense. 

If the salt solution formed about the active electrode be not 
removed the cell becomes a two-fluid cell, to be described later. 
A disadvantage of cells of which Cu — H,SO^ — Pt is a type is 
the tendency of the active metal to travel toward, and cover, the 
passive electrode, so that it is no longer inert ; this may be nearly 
overcome by the use of a large volume of liquid, or of a dia- 
phragm between the electrodes, the bulk of the electrolyte sur- 
rounding the active plate. 

The ideal cell is one in which the electrolyte is a solution oi a 
compound of the active metal, the substance (in the •' useful *' 
direction) then being the same at both electrodes. In such cases 
the soluble material must be a gas, discharging freely, or the pas- 
sive electrode will be quickly covered with active metal. Hydrogen, 
oxygen or chlorine, when occluded by platinum black, will act as 
gaseous electrodes, hydrogen in H,SO^ giving by far the best 
results. Oxygen can also be used in acid solutions in the form of 
peroxide of lead or manganese, or in alkaline solutions as oxide 
of copper, the current then passing freely from the inert plate as 

Cell //, — H^SO^ — PL — A long series of experiments was tried 
with the cell H, — H,SO^ — Pt,* the hydrogen electrode consisting 
of platinized platinum partly in the gas, partly in the electrolyte. 

* See also Part I. for table of current curves of this cell. 


The direct impressed E.M.F., as in other preliminary experiments, 
was furnished by gravity cells, the alternating by commutating the 
direct (see description of apparatus), and the low resistance coil 
of voltmeter served as an ammeter. Before the hydrogen was 
allohved to enter the platinum the asymmetrical effect was almost 
nil ; immediately after entrance it rose to a high figure. The read- 
ing dropped slowly when the electrode was entirely immersed. 
Under certain circumstances the ammeter reading from the "alter- 
nating " current was higher than the final reading produced by the 
un-commutated direct current, though lower than the direct cur- 
rent reading immediately after stopping^ the commutator. A large 
bottle provided with electrodes and inlet and outlet tubes was used 
for trying the effect of hydrogen in H,SO^, oxygen and air in acid 
and alkaline solutions, chlorine in hydrochloric acid and chlorine 
water, and illuminating gas in H,SO^. The hydrogen alone gave 
satisfactory results. With oxygen the effects were slight and 

Several sealed hydrogen cells * were made with small bottles 
2 cm. to 5 cm. in diameter, from which a number of curves were 
plotted. They were " continuous-acting *' as long as the impressed 
E.M.F. was not too high, and changed very slightly with use. 
The area of the larger electrode considerably affected the results ; 
after running a few minutes the reading was invariably lower than at 
starting, due to the too rapid exhaustion of the active electrode, or 
possibly to a layer of highly resistant pure water at its surface, and, 
it is probable, to an imperfect action of the small electrode. After 
a cell has run undisturbed for a considerable time the electrolyte at 
the bottom is denser than that above. The E.M.F. it is neces- 
sary to add to balance a hydrogeii-cell in operation is not known. 
It is assumed to be approximately .085 volt, as that E.M.F. is 
just insufficient to send a current through the cell, while it reduces 
the lead of the current to zero, and the curves taken with it show 
the best agreement with the pure theory. It has been found, how- 
ever, that after passing an alternating current through a " bal- 
anced" cell for a short time, and then stopping, a small " useful " 
current will flow for awhile. When .085 volt is used to balance 
there is an evolution of hydrogen at the small plate for several 
minutes after stopping. Even with .01 volt an afterflow of cur- 
rent may be noted. 

* Sec " Continnoos- Acting Rectifiers." 



Curve 24 shows the effect of the E.M.F. containing the third 
and fifth harmonics ;.it is of the same general form as correspond- 
ing curves in Part I. 

As a support for the platinizing very fine platinum gauze was 
tried, but the i^esults was no better than with foil. Palladium 
black on the gauze did not seem to improve the effects. A fairly 
satisfactory electrode was formed of small glass rod^ rolled in gold- 
leaf and heated, bound about a central gilded rod, and platinized. 
The best results were obtained Mrith palladium coated platinum 
foil, the ammeter readings, on long-continued currents, falling very 
slightly from their first values ; exactly similar cells having platin- 
ized electrodes were comparatively unsatisfactory. 

Numerous experiments were tried with charcoal as the hydrogen 
absorbent. Gas-carbon, plain or charcoal-coated, drawing char- 
coal, and fuel charcoal, were used, but the results appeared to show 
no effect due to the hydrogen atmosphere. Asymmetrical effects 
were produced, but it was concluded that they were due principally 
to the anodic actions noted at the end of the next division of the 
paper; it was found that the higher the impressed E.M.F. the 
greater the difference between the + and — readings, and that 
diluting the electrolyte lowered the higher reading and greatly 
raised the lower, the impressed E.M.F. being 4 volts. By heat- 
ing charcoal and plunging it into water or dilute acid excellent 
electrodes were obtained, but their effect was found to be mostly, 
if not entirely, due to hydrogen generated within the pores by the 
action of the red-hot carbon on water. Curves with the carbon 
electrodes were plotted, but showed no particular features of in- 
terest; the A.E. was negligibly low in all but one, in which 
quenched charcoal formed the electrode. 

Proportionaltty of Readings, — A series of experiments to determine 
the proportionality of the ammeter readings to the alternating 
voltage was made, the cell being just balanced by direct E.MF. 
from a slide-wire so that no deflection could be observed. Over- 
balancing while slight did not seem to affect the results, but when 
greater impaired them. Readings were taken at low and high 
frequencies, up to 750 p.p.s.,the voltage being taken from a gradu- 
ated slide-wire. Better results were obtained with a small electrode 
than a large. When properly balanced very closely proportional 
readings, with a small electrode, were noted between zero and .a 
point on the ammeter equivalent to about .4 volt (^ of scale) across 


its terminals, ten readings being taken. Above this point the 
divisions became too small ; in unbalanced cells the middle divi- 
sions were larger than those below or above. Although the read- 
ings of a set in balanced cells were proportional the divisions were 
not always the same length, depending on whether the readings were 
made with increasing or decreasing voltage, and other similar cir- 

At low frequencies a vibration of the needle, due to the giving 
off of gas bubbles, is noticeable; at high frequencies it is very 
slight. That the bubbles have no effect on the rectification is evi- 
denced by the smoothness of the curves; were the bubbling irreg- 
ular, with any effect, it would be impossible to determine the wave- 
forms. The bubbles seem to be produced on small areas which 
are always covered with gas. At low frequencies, also, the periods 
of the needle and of the current approximate each other, giving a 
tendency to swing. 

" Continuous-acHng" Rectifiers. — Single-cell CA.'s adapt them- 
selves particularly well to a "continuous-acting" construction, 
avoiding the wasting of the active cell in the 
two- cell rectifier, or of the active electrode in 
the ordinary single-cell form. Continuous- 
acting cells are made self-regenerative by 
causing the substance discharged from the 
passive electrode to return, by gravity or 
otherwise, to the active electrode whence it 
came. A typical form that is simple and very 
satisfactory is shown in Fig. ii, consisting of 
a sealed Jar of dilute H,SO, containing a 
small platinum electrode and a large plati- 
nized electrode whose upper portion is in a hydrogen atmosphere. 
The "useful" current carries hydrogen from the large to the 
small electrode, at which it is given off in bubbles which return to 
the hydrogen atmosphere above, thence to be reabsorbed by the 
large electrode. In the " useless " direction oxygen is concen- 
trated at the small electrode, but not given off unless the impressed 
E,M.F., the frequency, and the external resistance are such as to 
raise the counter E.M.F. of the cell to 1.67 volts. As long, 
therefore, as the impressed E.M.F. is kept sufficiently low the 
hydrogen atmosphere will remain pure, and the counter E.M.F. 
of the cell be a few hundredths of a volt in one direction, and 


1.67 volts in the opposite direction. By reducing the gas pressure 
in the cell the excess E.M.F. necessary to the formation of hy- 
drogen bubbles may be largely reduced, while but slightly lowering 
the active E.M.F. Increasing the pressure has the reverse effect, 
but the advantage of decreasing the necessary size of the large 
electrode. The large electrode should be of sufficient area to ab- 
sorb the hydrogen as rapidly as it is produced at the small elec- 
trode ; its size bears no relation to the size of the small electrode 
or to the efficiency and depends only on the size of the " direct *' 
current. The electrodes can be brought as near together as desir- 
able, leaving room for the escape of bubbles, though the tendency 
is then for only the part of the large electrode directly opposite 
the small to act. This disadvantage may be avoided by dividing 
the inert electrode into a number of parts, thus reducing the re- 
sistance and facilitating the discharge of bubbles, and by the tend- 
ency of the hydrogen to flow from the charge to the uncharged 
part of the plate. 

Gee and Holden * have investigated resistance effects at a small 
anode which are absent when the cathode only is small. Frommef 
has found a higher polarization possible with a small anode and 
large cathode than when the reverse is the case. The difference 
in the polarization attainable with smooth and platinized electrodes 
has been shown by Glaser % to be almost entirely due to the anode 
actions. These three effects act in the same direction as the main 
effect, increasing the A.E. Continuous-acting two-cell rectifiers 
can be constructed by using a U-tube as the decomposition cell, 
the gas above the electrolyte in each leg being in communication 
with the gas about one of the electrodes of a hydrogen- oxygen 
gas-battery. The electrolytes in the two cells must be insulated 
from one another, as by paraffining the interior of the tubes con- 
necting the atmospheres. The active E.M.F. is 1.07 volts, the 
passive 1.67. Water is continually carried from the decomposition 
cell to the gas battery ; it may be replaced by gravity, without 
breaking the seal, by tipping the rectifier. The arrangement is 
clumsy and has many disadvantages. 

*Proc, Lond. Phys, Soc, VoL IX., 1888; or Phil Mag,, Vol XXV., p. 276, and 
Vol. XXVI , p. 126. 

t Wied. Ann,, VoL XXXIIL, p. 80, 1888; also VoL XXXVIII., p. 362, and VoL 
XXXIX., p. 187. 

\ Ziit, Jur EUkirochimU, Vol. IV., p. 376, Feb. 20, 1898. 


Two-solution Rectifiers Having Differing Electrodes, — Cells having 
two electrodes differing in composition dipping into two different 
electrolytes in electrolytic communication may be divided into 
three classes : The active electrode may be in a solution of one of 
its own salts, the passive in a solution of a salt of some other 
metal ; the active electrode may be in a solution of a salt of some 
other metal, the passive in a solution of a salt of the active metdl ; 
or both electrodes may be in solutions other than that of the active 
metal. The first of these classes is of great importance, the others 
though interesting, of but little. To the first class belong the two- 
fluid primary cells, in which the active is greater than the passive 
E.M.F. By lowering the active E.M.F. sufficiently a rectifier 
is produced which gives no current, whose theory is the pure 
theory of a two-cell rectifier having a finite and an infinite elec- 
trode, and which is equivalent to a single-liquid, single-cell rectifier 
in which the salt formed about the soluble electrode is allowed to 
remain. The advantage of this type of cell is that it is possible to 
use a solid active electrode, and still have a gas as the substance 
deposited at the passive electrode. The cells having copper, iron 
and zinc electrodes, described for simplicity under one heading 
with single-solution cells, really belong to the class under present 
consideration. Balancing may be done as previously described. 
Electrodes of two soluble metals in solutions of their salts will not 
form a rectifier ; if the potential differences are unequal a continu- 
ous current will flow, if equal none, in either case the alternating 
current passing unchanged because both electrodes are perfectly 
reversible and of practically infinite capacity. 

Iwo'Solution Rectifiers Having Similar Electrodes, — Rectifiers may 
be constructed both of whose electrodes are inert, but dip into 
electrolytes differing in composition or concentration. In the cell 
Pt — H,SO^ — cone. HNOj — Pt the " useless " current deposits 
H in H,SO, and O in ^HNO,, and the " useful " O in H,SO, 
and nothing in HNO3, the H being oxydized by the HNO,. The 
counter E.M.F. in the first instance is consequently considerably 
greater than in the second. The same general principles govern 
the size of the electrodes as in rectifiers having active and passive 
plates. The HNO, forms an irreversible electrolyte, it being pos- 
sible to decompose, but not to recompose it electrolytically, and 
its action on the cathode is the equivalent of the giving off of hy- 
drogen bubbles without the production of counter E.M.F. The 



action of the cell in one direction is that of a condenser having 
both plates of finite size, and in the opposite direction that of a 
cell having one plate infinitely large ; that is, one half. wave and a 
part of the succeeding one are choked down as if there were two 
electrostatic condensers in series, while the remainder of the sec- 
ond half. wave passes as if only one condenser were left in circuit. 
There is a resistance effect due to the formation of bubbles which 
aids the E.M.F. effect, though small as compared with it Dilute 
HNO, can be used in the place of the H,SO^ without greatly affect- 
ing the results. 

A direct E.M.F. of i.i volts applied to a cell having small elec- 
trodes diflfering slightly in size give o ammeter readings in both 
directions. 2.2 volts gave readings of o, 50 and 25 when corn- 
mutated to about ten periods per second, the larger electrode being 
in the H,SO^. With the larger in HNO, the readings were o, 65 
and 35. With 3.3 volts, larger electrode in H,SO^ readings were 
I35» 50 and 25 ; for larger in HNO, readings were 135, 50 and 50. 
Curves (26) and (27) were from this type of cell ; the current fol- 
lows the E.M.F. quite closely. 

Very curious effects are produced by platinized electrodes in 
concentrated or fuming nitric acids * Four equal-sized small elec- 
trodes, two smooth (5), two platinized {P), were tried in pairs in 
dilute H,SO^ and concentrated HNO, in electrolytic communica- 
tion, and in each alone, the impressed E.M.F. being 3.3 volts di. 
rect. The following readings were taken after the current had 
reached a steady value, each being an average of several. 




^ — HNO, Cells 


— H,SO, — S 

, Smooth 


and Platinized Electro< 



S— H.SO. 



S— H.SO, 

— H,SO, — P 

225 , 


P— H^, 

— H,SO, — P 

— HNO, — S 



S — HNO, 

350 to 375 

350 to 375 


— HNO, — P 

425 1 

170 and 425 

P— HNO, 

— HNO, — P 

210 and 450 1 

210 and 450 

S — HNO, 

— HjSO^ — S 



P — HNO, 

— HjSO, — S 

80 and 210 


S — HNO, 

— H-SO4— P 



P — HNO, 

— H,SO, — P 

185 and 340 I 


* Care must bs taken that the platiniziDg be not removed by the rapid evolution of 




The double value of the current when P is in HNO, seems to 
be produced by a counter E.M.F. efifect possibly due to a double 
or multiple decomposition value of HNO, ; the resistance is com- 
paratively little changed. The lower reading was given when the 
external resistance on closing the circuit was small, the higher 
when the current density at starting was low and was gradually 
brought to the final value by removing resistance. The higher 
reading could also be obtained if the circuit were broken and im- 
mediately restored, or by the introduction and removal of resist- 
ance. The rise from the lower to the higher value was quite rapid 
when once started ; it seemed as if an unstable equilibrium were 
formed which could be thrown into stable equilibrium in the manner 
described. It should be observed that when P is in HNO, and S 
in H,SO^ asymmetrical conduction can take place in either direc- 
tion, depending on the external resistance at starting. 

A similar table follows. S' and P' are smaller than S and P, and 
L is an electrode i,ooo sq. mm. in area. " Imm." current values 
are those read immediately after closing the circuit, *' Final " those 
after the current had ceased falling. The periodicity of the •' alter- 
nating " current was about lo. 

Table VI. 

H.SO4 — HNO5 Cells, Smooth and Platinized Electrodes. 







Direct Current. 


Imm. I Final. Imm. Final. 

S/ — HNO 



S^ — HNO,' — 




p/ _- HNO, — 


HNO3 — 
HNO3 — 
HNO" — 









































The main flow of current in all cases being toward the HNO,, 
any sulphate solution may be used in the place of H^SO^ without 
interfering with the action by covering the sulphate electrode with 
a layer of active metal ; certain advantages accrue from the use 
of solutions having a higher decomposition value than H^SO^. 
ZnSO^, pure or with a little H,SO^ added to improve its conduc- 
tivity, in place of the H,SO^ gave almost the same effect as the 
H,SO^. Chromic acid, binoxide of hydrogen, potassium perman- 


ganate, and potassium chlorate may be used instead of the HNO,, 
but with less success. Fuming nitric gives results little, if any, 
better than the pure acid, and is clumsier to handle. These oxi- 
dizing electrolytes give the plate immersed in them an infinite 
capacity for neutralizing the charges of hydrogen ions, and none 
for those of oxygen. 

Cells in which irreversability is secured by the discharge of the 
gases from their electrodes, instead of by means of an irreversible 
oxydizing electrolyte, may be constructed. The two essential 
requisites of the liquids are that their decomposition values be 
different, and they be of such composition that the " useful " cur- 
rent deposits a gas on both electrodes. A typical cell is Pt — 
ZnSO^ — H,SO^ — Pt, the •* useful '* current flowing toward the 
H,SO^ ; with 2.2 volts impressed no asymmetrical effect was ob- 
served; with 3.3 volts readings of 15 were obtained with the al- 
ternating current. Using smooth electrodes differing in size, or a 
platinized electrode in the H,SO^, changed the direct current read- 
ings, though affecting the alternating but slightly. Zinc is gradu- 
ally carried into the H,SO^, while the ZnSO^ becomes mixed with 
H,SO^, and water is carried off in the form of oxygen and hydro- 
gen gases. 

Cells can be formed in which the differences in counter E.M.F. 
are produced by the non-metallic ions, instead of by the me- 
tallic as in the foregoing cells. A typical cell consists of the 
combination Pt— H,SO^ — HI — Pt, the counter E.M.F. in one 
direction being produced by H in H,SO^ and I in HI ; in the other 
by O in H^SO^ and H in HI. Curve 28 shows the effect of such a 
cell. No direct current records were kept for either that cell or for 
thecell Pt — ZnSO^ — HI — Pt, in which the action is produced by 
Zn and I in one direction, and by O and H in the other. The 
efficiency in the ZnSO^ cell is the lower, due to the higher polari- 
zation attainable by the plate in ZnSO^ and to the lower con- 
ductivity of ZnSO. 

It should be possible (though the writer has not attempted it) to 
construct cells having two inert electrodes in two electrolytes of 
such composition or concentration that the electrodes will polarize 
and no current flow except on the application of a properly di- 
rected, or an alternating, E.M.F. It should also be possible to 
construct cells on the principle of oxidation and reduction cells 
will have similar properties. 


Conclusions. — It may be remarked, in conclusion, that the actions 
in the foregoing rectifiers are undoubtedly due to the counter 
E. M. F., effects by which they have been explained. Resistance 
produced by gas bubbles is shown, by the smoothness of the plot- 
ted curves, to be negligible. Resistance produced by a dense film 
at the anode is not the cause, for Gee and Holden state that no such 
film is formed in HNO,, and the formation of such a film in the 
H,SO^ would retard, not aid, the action which goes on, and it has 
been shown that dilute HNO, can be used to replace the H^SO^. 
Finally, the direct current produced is not due to the action of a 
liquid (concentration) cell, for in the cell Pt — cone. HNO, — dil. 
HNO, — Pt, the alternating current drives the hydrogen towards 
the concentrated acid, while the natural tendency of the hydrogen 
(without the application of an external E.M.F.) is to go toward the 
dilute acid. In general the E.M.F.'s due to the equalization ol 
liquid concentrations are too evanescent to produce the long.con- 
tinued effects observable in the curves plotted. 

In all cases in which two electrodes dip in two electrolytes in 
electrolytic communication, arranged to present some condition of 
irreversibility, the main flow of a partially rectified current will be 
toward the electrolyte having the lower decomposition value. 

In general, when an alternating current is applied to a large and 
a small plate in two electrolytes in electrolytic communication the 
nature of the electrolyte has little effect on the efficiency (except 
indirectly, by its resistance), providing the main flow of current be 
toward the smaller plate. If the main flow of current be toward 
the larger plate the action may quickly stop, due to that plate be- 
coming covered with a non-diffusing deposited material. 

Large and Small Plate in Electrolyte. — Cells having simply a large 
and a small plate in 25 per cent. H,SO^ were tried. The current 
density was kept too low for resistance effects due to steam,* and 
the electrolyte used was probably too dilute for those investigated 
by Gee and Holden ; it seems possible that the results were pro- 
duced by differences in the polarizations at the small plate when 
used as anode or cathode, as noted by Fromme. Mengarini f pro- 
duced asymmetry in alternating currents by using a large and 

* Wchnelt Interrupter Effect ; Eichbcrg and Kallir, Zeii.fur EUk.f/L^x, 16, 1899 ; 
ab8. in Dig, EUc. World, Vol. XXXIII., May 13, 1899, p. 627. 

\EUc. World, Vol XVIII., Aug. 8, 15 and 22, 1891 ; Und. EUc.,Vo\, XXVII., 
July 17 and 24, 1891. 


small plate in dilute H^SO^, obtaining, apparently, an effect at low 
voltages due to polarization, and at higher E.M.F.'s to an action 
like that of a Wehnelt interrupter run on an alternating current. 
Mr. W. H. Freedman and the writer have found a "direct" cur- 
rent to pass in an alternating current Wehnelt interrupter circuit, 
at low voltage in one direction, at higher voltages in the opposite, 
and at still higher voltages in the original direction. At low 
voltages the current as shown by a CuSO^ voltmeter with clean 
platinum electrodes, passes the more easily through the electrolyte 
toward the smaller plates, but only for a short time. Curves taken 
with the large plate stationary showed practically no asymmetrical 
effect. By rotating the large plate (a i cm. x 10 cm. strip of 
platinum wound about a cork) an efficiency of about 6 per cent, 
was obtained. A small platinum and large platinized electrode in 
dilute acid permitted direct current to pass the more easily to the 
small plate as cathode,* but no effect could be noted with the al- 
ternating current. 

Principles Utilizable in Asymmetrical Conductors. 

In a rectifier in which the action depends on electrolytic polari- 
zation the accumulation of a certain deposited substance should 
be permitted at one electrode during an instant, and prevented 
at the other electrode at the next instant, when the current 
reverses. The following methods may be used to accomplish 
this; they include several methods which, though incapable of 
giving an asymmetrical effect alone, can be used to increase the 
asymmetry of a current partially rectified by other means. It 
should be noted that there will be a continual bringing of a sub- 
stance toward the electrode on which its accummulation is pre- 
vented, in excess of that toward the other electrode, necessitating 
some means of discharging the accumulation and making the de- 
vice irreversible. 
Production of Asymmetry, — I. Physical. 

1. The driving of gas from small electrodes by its expansion 
and electrification. 

2. The use, in the same liquid, of electrodes whose conditions 
of surface differ. 

3. The use, in the same liquid, of electrodes whose areas of 
surface differ. 

* Discovered by PoggeDdotff; Pogg. Ann., Vol. LXX., p. 177, 1847. 


II. Mechanical. 

4. Moving of one electrode. 

5. Moving of the liquid about one electrode. 

6. Moving of the liquid about electrodes of different size. 

7. Heating one electrode (to drive gas off). 

8. Heating or boiling the solution, with different sized electrodes, 

9. Removing the electrode periodically from the liquid. 

III. Chemical. 

10. Electrodes of different materials in the same liquid. 

11. Electrodes of the same material in different liquids. 

12. Electrodes of different materials in different liquids. 
Securing of Irreversibility, 

1. The production and giving off of gas bubbles. 

2. The solution of the substance deposited in an irreversible 

3. The use of an irreversible active electrode. 

4. The removal of the deposited substance by scraping the pas- 
sive electrode, or moving that part of its surface which has become 
covered into a liquid which dissolves the deposited material and 
then back into the original solution. 

5. The removal of the salt produced by. the action of the elec- 
trolyte on the active electrode from the neighborhood of that elec- 
trode by a rapid motion of the liquid. 


Tutor in Electrical Enginetring^ Columbia University, 


§ I. The subject matter of the following pages consists in a 
description of a series of experiments which make up the alternat- 
ing current laboratory course prescribed for the fourth-year stu- 
dents in the electrical engineering department of Columbia Uni- 
versity. In addition to most of the usual commercial tests, a 
number of experiments are given which are designed solely to 


point out and explain the characteristic properties of the apparatus 

One of the objects in view is to bring out the main diflferences 
between alternating and direct current machines of the same class. 
The reader is assumed to be familiar with the methods of direct 
current testing, and with the fundamental principles of the theory 
of alternating currents. 

Although no pretence is made of presenting a complete manual 
of alternating current testing, it is hoped that a pamphlet contain- 
ing the principal commercial tests, together with short discussions 
upon the theory of the machines, may be of some value to elec- 
trical engineers and students as well. 

Table of Symbols. 

§ 2. E Voltage, E.M.F. potential difference. 
/ Current. 
P Power. 
R Resistance. 
L Inductance. 
C Capacity. 
X Reactance. 
Z Impedance, 
f Total magnetic flux. 
B Magnetic density, induction. 
H Magnetomotive force per unit length. 
F Magnetomotive force.- 
A* Permeability. 
R Reluctance. 

/ Number of pairs of poles. 
/ frequency. 

N Number of revolutions per minute. 
H. P. Horse-power. 
Eff. Commercial efficiency. 





(z) D. C. Voltmeter. 

VOL. XXII.— a. 


"S^ A. C. Voltmeter. 


. C. Ammeter. 

Yy -^* ^* Ammeter. 



Variable Inductance. 


1 a. 



Variable Resistance. 


Synchronous Motor. 

m5 D. C. Motor. 

3-Phase Induction Motor. 

^\LJt A. C. Generator. 

*S. ^ 

. C. Generator. 

Motor Driven. 
A. C. Generator. 

[t CI W Rotary Converter. 


§3. Useful Equations. 

/ and E denote the effective values of current and E.M.F., 

Effective value of sine i 
Maximum value of sine " ^~2 

Average value of sine 2 
Maximum value of sine " t: ' 

In general^ if both inductance and capacity are in circuit, 

X « wL , 


where o) ■■ 2;r/". 
P= EI cos d^ PR, 


tan a ^-^f sm ^ » -^ , cos ^y. 

Resistance drop =» //?, 
Reactance " ■■ AY, 
Impedance " « /Z, 

where s » number of magnetizing turns, and 

/ » length of magnetic circuit. 

Z = * — J—, where ^= cross section of the coil, 

J/= mutual inductance of two coils = r^— . 

The maximum value of the induction in a transformer is given 
by the equation, 


E X 10® 



where E^ open circuit secondary E.M.F. 

This equation is only true when ^ is a sine function. 
At any instant, 


^ s ^, where F = the res ultant M.M.F. 

Watts = H.P. X 746, 

ft. lbs. per sec 

H»P, ^ . 


The torque exerted by a machine is 

^ H,P, X 33,000 , 
r= j^ , where 

7 is expressed in lbs. pull at i ft. radius. 

In a circuit containing inductance and capacity resonance occurs 
when the frequency is such that the impedance is a minimum, or 

oiL = — • 

The period of the impressed E,M.F. is then equal to that of the 
circuit The period of a circuit of negligible resistance is 

7^= 2Tzy/LC, 

§ 4. Schedule of Experiments. 
Elementary Properties of Alternating Current Circuits. 

1. Measurement of an inductance by the method of impedance 

2. Measurement of a condensance. 

3. Measurement of power with inductance load, {a) Three- 
voltmeter method, [S) three-ammeter method, {c) wattmeter 

4. Measurement of the mutual inductance of two coils, using 
successively both coils as the secondary. 

The Alternating Current Generator. 

5. Field compounding. 

6. Full load saturation curve, or magnetization curve at full load 


7. No load saturation curve, or magnetization curve at no load. 

8. (^z) Short circuit core loss, {b) Open circuit core loss. 
9* Determination of synchronous impedance. 

ID. External characteristic. Non-inductive load. 

Alternators in Parallel. 
{a) Generator and Motor. 

1 1. External characteristic of an alternator at constant power 
factor, {a) Current leading, {b) Current lagging. 

12. The synchronous motor. Characteristic curves. 

With normal field current, determine the curve of commercial 

1 3. At no load, determine the curves of current and power fac- 
tor in relation to the motor field current. 

14. With constant motor excitation and variable torque determine 
the curves of current and power factor. 

(/9) Two Alternators in Parallel Supplying a Constant Potential 

Distributing Circutt, 

15. Adjustment of the generator excitations for maximum econ- 
omy ; the power of the two driving motors being constant. 

Tests on Transformers. 

16. Determination of the core losses. 

17. Determination of the copper losses. 

18. Determination of the efficiency curve. 

19. Heat test by the motor dynamo method. 

Curve Tracing. 

Determination of the Voltage a?ui Current Cufves of a Transformer 

under Various Conditions. 

20. With the secondary circuit open determine the following 

(d) Primary current 
\b) Primary E.M.F. 

(c) Secondary E.M.F. 

21. With full load on secondary determine the following curves: 

(d) Primary E.M.F. 
(^) Primary current. 
{c) Secondary E.M.F. 
[d) Secondary current. 


22. Determination of the curves of 19, the secondary circuit be- 
ing connected to the primary of a step-down transformer. 

23. Determination of the curves of 19, the step-down trans- 
former on the secondary circuit carrying full load. 

Resonant Rise of Potential. 

24. Connect a condenser in series with an inductance. Deter- 
mine a curve with condenser voltage as ordinate and frequency as 

25. Resonant rise of potential obtained by means of a super* 
excited synchronous motor. Determine two curves. 

{cC) Voltage at motor terminals as ordinate and frequency as 
abscissa, with constant motor excitation. 

(p) Voltage at motor terminals as ordinate and field currents 
as abscissae, with constant frequency. 

Polyphase Circuits and Transformations. 

26. Balanced two-phase system. 

27. Transformation from two-phase to three-phase by means of 
two transformers. 

28. Three-phase J with two transformers. 

29. Three-phase Fwith two transformers. 

Test of a Three-Phase Induction Motor. 

30. Determine the following curves : 
{a) Commercial efficiency. 

(p) Apparent efficiency. 

{c) Current per phase. 

(d) Speed in per cent, of synchronism. 

{e) Torque. 

(/) Power factor. 

\g) C'R of fields. 


I. Running from the A,C. End. 

31. Various methods for synchronizing polyphase converters. 
Determine the efficiency and external characteristic curves at the 


{a) Using a two-phase converter. 

{b) Using the same machine running single-phase. 


32 (fl) At fill! load determine the relation between the D.C. 
volts and the field current 

{U) Insert an equal inductance in each phase, and determine the 
same curve. 

11. Running from the D.C, End. 

33. At full load, non-inductive, determine the curves of speed 
and A.C. voltage in relation to the field current 

34. With lagging current, vary the output, while keeping the 
power factor approximately constant Determine a curve with 
speed as ordinate, and current output as abscissa. 

§ 5. Elementary Properties of A.C. Circuits. 

^Experiment (i). Measurement of an inductance by the method 

of impedance. 
In order to determine the inductance of the coil Z, the drop of 
potential in the coil is measured at a known current and frequency. 
This drop is due both to resistance and in- 
ductance, hence in order to find the induc- 
tance it is necessary to have already de- >-h /^ ^ — -^ 

termined the resistance. The drop across ' ^j^^ — \i/- 

the coil is 

Fig. I. 
Method. — Make connections as shown and 

note values of current, speed and voltage across the coil. The value 

of L may be obtained from the equation 



in which E is the drop across the coil. If R is not known, its value 
must be determined by some direct-current method. 

Report. — Evaluate L from several sets of simultaneous readings 
at diflferent values of to and /; give the values of the diflferent 
quantities in the equation. 

Experiment (2). Measurement of a Condensance. 
The drop across a condensance is 

F — T ^ — 
^^^ mC^ R^ mC 

where E^ is the drop across the known resistance R^ which is in 



series with the capacity C. R must be sufficiently great to {m-o- 
duce a convenient deflection of the voltmeter. 

Method, — ^With the connections as in the diagram read the volt- 
ages E^ and E^ across the condensance and resistance respectively, 

Note the frequency, and determine the value 

T li — , of R if not already known. Take several 

sets of readings under different conditions. 
Determine C by the equation 

C = 


Fig. 2. 

Report. — ^Tabulate the readings obtained. 
Give the calculated values of C, and from them obtain the average 

Experiment (3). Measurement of Power with Inductive Load, 

(^^) Three -voltmeter method. 

{b) Three-ammeter method. 

{c) Wattmeter method. 

[a) and [c)\ — If the non-inductive resistance R is known, it is pos- 
sible to find the power developed in the coil L by determining the 
drop across i?, then across Z, and finally across 
both R and L. Owing to the difference in phase 
between these E.M.F.'s, the drop across R and 
L in series will not be equal to the sum of the 
drops across R and L individually. 

The RM.F's, may be represented vectoriably as in Fig. 3. 

The watts developed in the coil may be deduced from this dia- 

E E 

W= EJ cos 9 = — 77-^ cos 9 

* K 

and therefore 


cos =s -pr 

-^3 ^2 -^l 



F^ ^ F^ ^ 772 
n 3 2 1 

COS ^ = -^ 






Method. — Note the three indicated voltmeter readings, and de- 
duce the value of W. Note the reading of the wattmeter in order 
to compare it with the value obtained by calculation. Take several 
sets of readings of the voltmeters under different conditions. If 
three voltmeters are not available, all the readings may be obtained 
with one voltmeter by changing the terminals from one place to 

another, provided the speed and im- 
pressed voltage are kept constant. 





Fio. 5. 

R should be chosen so as to make the three drops as nearly 
equal as possible. That is the condition of greatest accuracy. 

Report, — ^Tabulate the readings obtained, deduce the correspond- 
ing values of W^ and calculate the average value. 

{p) The watts W developed in the coil L may be found from the 
readings of three ammeters connected as in 
Fig. 5. Owing to differences of phase the 
effective value of the current /, is not equal to 
the arithmetical sum, but to the vectorial sum 

or resultant of /j and /,. 

Fig. 6. 

The readings of the three ammeters may be 
represented vectorially as in Fig. 6. From this construction the 
value of W may be deduced. 

W^ I^R X /, cos d, 

cos ^Y* 


// - // + 2bl, + I,\ 



ff'-lC^s'-W + A')]- 

Method. — Connect as shown in Fig. 5. The three ammeter 


readings should be taken simultaneously. The known ohmic re- 
sistance R should have such a value as to make the three ammeter 
readings as nearly equal as possible. Note the reading of the 
wattmeter, in order to compare it with the calculated value of the 
watts. Take several sets of readings , under different conditions. 
Report. — ^Tabulate the readings obtained, calculate the values of 
W^ and find the average value. 

Experiment (4). Measurement of the mutual inductance of two coils, 
using successively both coils as the secondary. 

When an alternating current passes through a coil, an alternat- 
ing magnetic field is produced in the surrounding space. An E.M.F. 
will be set up by these lines of force in any neighboring coil. 
This E.M.F. depends on the primary current, on the frequency, and 
on the mutual inductance of the coils. It is given by the equation 


^depends only on the number of turns in the two coils ; their di- 
mensions and their relative position. 

Method. — Fig. 7 shows the necessary connections. Note the 
frequency and the instrument readings. After this, interchange 
the connections so as to make L^ the secondary, and Z, the 
primary, and take a second set of readings. The resulting value 

of M should be the same in both 
cases. This will only be true if 
the relative position of the coils 
has not been altered when chang 
ing the connections. 

Fia 7. 
Report. — Tabulate the readings and deduce the values of M. 

§ 6. The Alternating Current Generator. 

Discussion of its Distinctive Features. 

The action of the machine is the same as that of a separately 
excited D. C. dynamo, except for certain peculiarities due to the 
difference in the nature of the armature current. The fact that 
this current is alternating affects both the armature reaction and 
the internal drop in voltage. 

{a) Armature Reaction. 

It is pulsating, not constant. It depends on the phase as well as 
the amplitude of the current 



Fig. 8. 

In Fig. 8 the current in the coil a will be a maximum when in the 
position shown, provided that it is in phase with the E.M.F. This 

is because the E.M.F. is always a maximum where —j- is greatest, in- 
dependently of the phase of the current. Neglecting field distor- 
tion, this will always be when the coil is in the position a. 

With non-inductive load, therefore, both current and E.M.F. are a 
maximum when the coil is at a. The effect on the field will be to 
produce distortion, but no direct demag- 
netization. There are no back ampere 
turns because there is no commutator.'*' 

If the current lags behind the E.M.F. it 
will not reach a maximum until the coil 
has reached some position a', depending 
on the angle of lag. The current will 

then have a direct demagnetizing action on the field in addition to 
the distorting effect which is now less than at non-inductive load. 

In a similar way it will be seen that a leading current tends to 
strengthen the field due to the magnet windings. 

{b) Armature Drop. 

In a D. C machine, this is due only to armature reaction and 
resistance, but in an A. C. dynamo or motor the reactance of the 
winding must also be considered. 

Leaving armature reaction out of consideration, the terminal 
volts are equal to the resultant of the internal drop and the in- 
duced E.M.F. 

Figs. 9 and lo show how it comes about that this internal drop, 

Fig 9. Inductive load. 

Fig. 10. Non-inductive load. 

for any given values of induced volts, current and frequency, is 
greater with an inductive than with a non-inductive load. 

In these diagrams, OE and OE represent the terminal volts and 

* See «^ Alternating Current Phenomena,'' by Steinmetz, Chap. XVI. 



induced volts respectively. Oe is the E.M.F. consumed by the in- 
ternal impedance of the armature. Whatever the angle B, OE^ 
must be the resultant of Oe and OE, Therefore, as B increases 
the terminal volts decrease more and more as the line Oe comes 
into phase with WE. 

This action occurs not only in alternators, but also in many 
other cases, as in transformers and transmission lines. 

It is apparent from the above that it is difficult to separate the 
effect of internal reactance from that of armature reaction. The 
drop due to both these causes increases with the amplitude, and 
also with the angle of lag of the current. 

Experiment {^\ Field Compounding. 

As the load increases, the terminal volts of an A. C. dynamo 

are diminished on account of the actions described above. The 

object of this test is to determine at various loads what increase of 

excitation is necessary to maintain the no-load terminal voltage. 

Method, — Fig. 1 1 shows the connections for this test. Run the 

A. C. generator at constant fre- 
quency and constant terminal 
voltage, varying the load from 
zero to 50% over-load. Note 
the field and armature currents. 
Take six readings. The load 
^^^- "• must be non-inductive. 

N. B. — Keep frequency and terminal volts constant. 
Report, — Plot a curve with field current as ordinate, and arma- 
ture current as abscissa. 

Experiment (6). Full-load saturation cupve, or magnetization curve 

at full load current. 

In this test it is designed to show the relation between the ex- 
citation and the resulting terminal volts, when the armature is de- 
livering its rated current output. 

Method. — ^The connections are the same as those of Fig. 1 1 . The 
resistance r^,, however, must be such as to carry full load current 
throughout a wide range of resistance. The current must be kept 
constant by adjusting r, as the terminal volts increase. Make at 
least ten observations of the terminal volts and field current, 



carrying the latter as high as is allowable. Care must be taken 
that all readings are made with ascepding values of field current 
N. B. — Keep the frequency and the armature current constant 
Ripoft, — Construct a curve between terminal volts as ordinates, 
and field amperes as abscissae. 

Experiment (7). No4oad saturation curve^ or magnetization cutve 

at no load. 

The voltmeter readings of experiment (6) will be lower than 
the terminal volts at no load owing to impedance drop and arma- 
ture reaction. It will be possible to determine the exact amount 
of this reduction by determining a magnetization curve with open 
armature circuit, and comparing it with the curve of experiment (6). 

Method, — Keep the connections the same as in Fig. 1 1 except 
that the armature circuit must be open. Read the terminal volts 
at various ascending values of the field current. Take ten obser- 

N. B. — Keep the frequency constant 

Report. — Plot a curve with terminal volts as ordinates and field 
amperes as abscissae. This curve is to be drawn on the same 
sheet, and to the same scale as the full load saturation curve of ex- 
periment (6). 

Experiment (8). {a) Short-Circuit Core Loss, {b) Open-Circuit Core 


{a) The core loss in an alternator is made up of eddy current 
and hysteresis losses. These losses vary considerably with the 
load. At short circuit 
they are quite different 
from their values at no 
load. This is due to ^-^ 

the fact that at short ^^^ vliJ- 

circuit the field is greatly iio\Uk 
distorted by armature re- ^^ 
action, and the magnetic 
densities in the armature 
and pole pieces are com- 
pletely altered. 

Method, — ^The only practical way by which the core losses may 
be determined is by noting the difference in the power taken by 
the driving motor when the alternator field is open and closed. 





Fig. 12. 


The required connections and instruments are shown in Fig. 12. 
The armature of the alternator G is short-circuited through an 

To determine the short-circuit core loss for any value of the 
generator field current /,, make r, equal to zero. Adjust r^ until the 
driving motor runs at llie proper speed. This should be such as 
to run the alternator at its rated frequency. Note the readings of 
all the instruments. Let these readings be denoted W^ 7^, 7^, 7,. 

Open the field and armature circuits of the alternator. Keeping 
7^ constant, adjust r, until the motor M runs at the same speed as 
before. Note the readings W and 7',. The short-circuit core 
loss corresponding to the field current 7, will be given by the 

v,= \_W- {I*R^ + ///?„)] -{W'- /,''RJ. 

In this expression R^ and Rg are the resistances of the armature 
windings of the driving motor and of the alternator. 

Proceed in the same manner in order to obtain the core loss at 
other excitations. 

In this experiment, begin with small values of the alternator field 
current in order to avoid burning out the machine. 

N. B. — The frequency must be the same for every reading. 
Take six sets of observations. 

Repoft. — Plot a curve with core loss in watts as ordinate, and 
alternator field current as abscissa. 

{V) The open-circuit core loss may be determined by the method 
just described. It is interesting as showing by comparison with 
the curve of (a) how important is the effect of the armature ampere 
turns upon the field distribution and on the core losses. 

Method, — Leave all connections the same as in {d) except that 
the armature circuit of the alternator is left open. The method of 
finding the core loss is the same as in (a), except that the term 

fR in the equation for w is absent. 

N. B. — ^The frequency must be constant throughout. Take six 
sets of readings. 

Report. — Plot a curve with core loss in watts as ordinate, and 
alternator field current as abscissa. Draw this curve on the same 
sheet of cross-section paper, and to the same scale, as the corre- 
sponding curve of (a). 


Experiment (9). Deierminaiion of Synckronotis Impedance. 

The synchronous impedance of an alternator is equal to the in- 
duced Volts at a certain field strength divided by the short-circuit 
armature current at the same field strength. This ratio will be 
found to be the same at all excitations. 

The synchronous impedance includes not only the true resist- 
ance and reactance of the armature, but also an apparant impe- 
dance due to the armature reaction. The synchronous impedance 
therefore, really depends to a certain extent on the armature cur- 
rent. It will not be the same at all loads. 

Method. — ^The same connections are to be used as in Fig. 12. 

Run the alternator at its rated frequency and note the short-cir- 
cuit current at various excitations. Take at least three readings. 
The no-load volts may be obtained from the curves of experiment 
(7). In order to determine the synchronous impedance at short 
circuit, divide the no-load volts obtained from this curve by the 
amperes at short circuit at the same excitation. 

Experiment (10.) External Characteristic Curve. 

This is the relation between the terminal volts and the armature 
current with non-inductive load. 

Method. — Run the alternator at its rated speed and no-load vol- 
tage. Keeping the field current constant, close the armature cir- 
cuit, and note the terminal volts at increasing values of the arma- 
ture current Carry the test up to 50 per cent, overload. 

N. B. — Keep the field current and frequency constant. 

Report. — Plot a curve with terminal volts as ordinates and arma- 
ture current as abscissa. This curve should be drawn on the same 
sheet, and to the same scale, as the curves {a) and {b) of experi- 
ment (II). All three curves should start from the same no-load 


§8. Alternators in Parallel. 

Discussion of the necessafy conditions. 

In synchronizing alternators, there are three electrical condi- 
tions which the E.M.F. of the machines must fulfill. They are 

1. Equality of amplitude. 

2. ** " phase. 

3. Coincidence of wave shape. 

If anyone of these conditions is unfulfilled, a large wattless cur- 
rent flows from one machine to the other. Th^y then oscillate in 
speed with respect to each other, and probably drop out of step. 


In Fig. 13 at OE^ and OE^ represent the induced E.MF.'s of 

two alternators connected in multiple. Since 
there is a difference in phase 0^ the resultant 
E.M.F. acting through the circuit formed by the 

^ armatures of the two machines is £,-£. The 

current produced may be represented by the line 

EJ, E^a is then the ohmic drop, and Efi that due to inductance.* 
As the machines rotate synchronously without any rigid mechan- 
ical connection they have a tendency to oscillate in space with 
respect to each other. The angle B will vary periodically. 

The natural period of this oscillation depends on the mass of 
the armature, on the field excitation, and upon those factors in the 
design of the machine which affect the synchronizing power. It 
may probably be approximately represented by the following ex- 

'^'^^^ SF' in which 

J/s moment of inertia of armature. 
F 5 moment of synchronizing force. 
' The machines are liable to get out of step when some periodic 
irregularity in the power of the driving motors accentuates the 
natural period or one of its harmonics. 

This may occur either by resonance or interference. The latter 
is usually the case when the amplitude of the oscillations increases 
and decreases periodically. 

In discussing the subject of alternators in parallel there are two 
distinct cases to be considered : 

(a) Motor and generator. 

(i^) Generators in multiple furnishing power to a distributing 

These two cases had best be considered separately. 

§9. (a) Generator and Motor. 

I. Variable excitation^ constant motor torque. (Speed and im- 
pressed E.M.F. constant.) 

In this combination of machines there are, for every value of the 
current, two distinct conditions. This is shown in Fig. 14. In 

* See H. Georges 00 Purallel Roimiiig of Altenmton, EUetrettchnisthi Ziihchrift^ 
Vol XXL, p. 188. 












Fio. 14. 

one case the current leads the impressed OE^ and in the other it 
lags behind it. The leading cur- 
rent corresponds to the larger 

value of OE^* 

OE^ in this diagram must fall 

upon the line AB perpendicular 

to 07, because the protection of 

E^ on 01, which represents the 
output of the motor, is a con- 

Increasing the motor excitation, therefore, evidently produces a 
leading currentf 

II. Constant excitation, varying torque, (Speed and impressed 
E.M.F. constant.) 

Applying load to the motor will change not only all the phase 

relations, but also the 
amplitude of the current. 
Fig. 15 is designed to 
show the result of load- 
ing the motor when its 
^°* '^* excitation is such as to 

produce a lagging current at no 

The energy transformed by the 
motor into mechanical work is : 

W^ = lE^ cos 9^ 

As the load comes on the arma- ^ 

ture of the motor is held back so as 

Fio. 16. 

to increase the lag of OE^, while 

02", s= ZJ increases until the above equation for W^ is fulfilled. 

It will be seen that increasing the load brings the current into 
current into phase with the impressed E.M.F. 

A similar action takes place if the motor excitation is such as to 
produce a leading current at no load. This is shown in Fig. 16. 

* See " Alternating Current Machinery," by D. C. Jackson, pp. 575-586. 
See also Alternating Current Phenomena, by C. P. Steinmetz, Chap. XVIII. 

VOL. XXJI. — 3. 



Experiment {ii). External characteristic of an alternator at constant 
power factor, [a) Leading current, (^) Lagging current. 

In this experiment the object is to show the effect of the phase 
of the current upon the armature reaction. The leading and lag- 
ging currents are obtained by varying the excitation and the load 
of a synchronous motor. 

(a) Method, — Make the connections as shown in Fig. 17. Close 
switch a to the left, and start" the machines, leaving switch b open. 







Fig. 17. 

The synchronizing lamp will begin to flicker as the frequencies 
of the two alternators approach equality. When the voltages of 
G and M are so adjusted in amplitude and phase that the lamp 
remains black for a number of seconds, the switch 6 maybe closed, 
and the machines will run in synchronism. The switch 6 should 
be closed when the lamp is black, and as nearly as possible at 
the middle of the period of blackness. 

After the machines have been thrown together by closing switch 
6, the double throw switch a is to be thrown to the right. 

G will now run J/ as a motor, and M will in turn drive M^ as a 
generator furnishing current to t^. 

In order to determine the external characteristic of G with lead- 
ing current, superexcite the field of M so as to produce a power 
factor of about .7. Apply load to Mhy drawing current from M^ 
through r,. As the load increases it will be necessary to increase 
the excitation of M in order to keep the power factor constant. 


{b) Method, — Proceed throughout in the same manner as in {a) 
except that the current must be lagging and not leading. 

It is possible to distinguish a leading from a lagging current in 
the following way. Vary the field of J/ slightly, and observe the 
effect on the reading of the ammeter /. If the current is leading, 
increasing the field strength of M will increase the current /, but 
if the current is lagging it will decrease it. In this connection see 
Fig. 14. 

N. B. — Keep the frequency, the impressed voltage and the 
power factor constant. Take five readings. 

Report. — Plot two characteristic curves, {a) with leading current, 
{b) with lagging current, taking terminal volts of G^ as ordinates, 
and the current values, /, as abscissse. 

Plot these curves on the same sheet, and to the same scale as the 
curve of experiment (10). All three curves should start from the 
same no-load voltage. 

§ 10. The Synchronous Motor. 

Characteristic curves. 

Experiment ( 1 2). With normal field current determine the commercial 

efficiency curve. 

Method, — Use the connections and method of loading the motor 
M shown in Fig. 17. Run the machine, and note the input at 
various loads, carrying the test to about 50% overload. The load 
may be determined by measuring the output of M^ and adding 
to it the losses of this machine, which must have been previously 

The field current of J/j, the motor under test, must be kept con- 
stant at that value which makes the armature current a minimum 
at no load. In calculating the values of efficiency, the PR loss in 
the field of the motor must be included in the input. The volts 
across the field terminals must therefore be noted. 

N. B. — Keep the frequency, motor field current and impressed 
volts constant. 

Report. — Plot the values of efficiency in per cent, as ordinates, 
and the horse-power output as abscissae. 

Experiment (13). At no load determine the curves of current and 
power factor in relation to the motor field current. 

The relation between the armature and field currents of a syn- 


chronous motor is commonly called its phase characteristic. It 
has been explained in § 9, Fig. 14, that the power factor de- 
pends on the field excitation. Since the energy transformed into 
work by the motor is practically constant at no load, the current 
will vary with the excitation, because it depends on the power 

Method, — Bring the motor up to speed, and synchronize by the 
method of Experiment (11). Keep the impressed E.M.F. con- 
stant, and note the ammeter and wattmeter readings at successive 
values of the field current. Vary the field through as wide a range 
as possible without throwing the machines out of step. 

N. B. — Keep the impressed volts and the frequency constant. 
Take ten readings. 

Report, — Calculate the values of the power factors from the 
instrument readings. Plot the results in the form of a curve with 
field amperes as abscissae. On the same sheet, with the same ab- 
scissae, plot the curve of current. 

Experiment {14). With Constant Motor Excitation and Variabk 
Torque ^ Determine the Curves of Current and Power Factor, 

In § 9 it has been shown that with both impressed and counter 
E.M.F.'s constant, the phase relations vary with the load. The ob- 
ject of this experiment is to show the nature of this variation. 

Method, — Use the same connections and method of synchroniz- 
ing as described in connection with experiment ( 1 1). Note the 
current and watts taken by the motor M at various loads, keeping 
the motor excitation constant. 

Take two series of readings : 

(a) With under-excited motor field. The conditions will here 
correspond to Fig. 15. 

{p) With super-excited motor field. The conditions will be those 
shown in Fig. 16. 

In both (a) and {b), vary the load as far as possible without 
throwing the motor out of step. 

N. B. — Keep the impressed E.M.F.'s, the motor field and the 
frequency constant. Take eight readings. 

Report, — Plot the two sets of curves of {a) and [b\ each set on 
a separate sheet. Plot the curves of current and power factor as 
ordinates, with torque in foot lbs. as abscissa. 


§ 1 1 (/9) Two Alternators in Parallel Supplying a Constant 

Potential Distributing Circuit. 

The most important factor in connection with A.C. generators 
running in parallel is that the division of the load is determined 
wholly by the relative torque of the driving motors, and not by 
the field excitations. This is because, when running in synchro- 
nisms, the speed of both machines must always be the same. This* 
of course, is true without regard to the nature of the driving 

In Fig. 1 8, OA represents the constant line voltage supplied at 

the bus bars, while OB and AB represent respectively the ohmic 
and inductive drop on the line.* 

If the whole load were carried by one alternator alone, the in- 
duced E.M.F. of this machine would be given by the line UE^ and 

the internal inductive drop by the line AE, The resistance of the 
armature is assumed to be negligible for the sake of the simplicity 

of the diagram. AE is then the prolongation of AB, 

The length AE will be proportional to the current because 
AE=^ IX. The phase of the current is given by the line -^/par- 
allel to UK 

If the load is taken by two machines instead of one alone, their 

E.M.F.'s will be represented by OE^ and OE^. These vectors 
must be so disposed in phase and amplitude that AE^ and AE^ will 
complete the parallelogram of which AE is the diagonal. 

AE^ and AE^ represent the drop over the armature inductances 

♦ See Eclair agi kltcirique. Vol. XXII., p. 401. 


of the two machines, and consequently are proportional to their 

The power developed in the distributing circuit by each of 
the two alternators is given by the projection of E^ and E^ upon a 
perpendicular to the line OA drawn from the point E^ 

We have seen that AE^ is proportional to the current developed 
by the alternator whose E.M.F. is E^, The direction of this cur- 
rent is along the line 01^ at right angles to AE^, The power de- 
veloped in the distributing circuit by the current 01^ is 

W.^AE. OA cosO,. 

Since OA is a constant, 

AE^. OA cos ^jOC AE^ sin ^j. 
Hence, W^^^P^Q: 

As explained above, the power developed by each machine is a 
constant depending on the torque of the driving motor. If, there- 
fore, the excitations of the alternators are varied, the points E^ and 
-fi", will move along lines parallel to OA through the points P^ and P,. 

The I^R losses of the two machines will be a minimum when 
the sum of AE^ and AE^ is a minimum. This is the case when 
E^ and E^ both lie on the line AE 

The machine whose E.M.F. is given by the line OE^ has the 
greatest lead, furnishes the most current, and is doing the most 
work. Its driving motor is the one exerting the greater torque. 

If the length OE^ is increased by raising the excitation of the 
machine in question this E.M.F. will come more into phase with 
that of the line, and cause the point E^ to approach the line AE 
which is the position of maximum economy. A similar result is 
obtained by diminishing 0£^. 

§ 12. 
Expetitnent (15). Adjustment of the generator excitations for maxi- 
mum economy f the power of the driving motors being constant. 

The conclusions that follow from the discussion of section 1 1 
are that although each alternator must always bear the same share 
of the load, its phase relation to the other alternators running in 
parallel with it can be changed by varying the field current. The 
machines are always performing the same number of revolutions 



per minute yet their relative angular position in space depends on 
their field exdtations. The problem is to adjust the field excita- 
tions so that the internal I^R loss of the alternators will be a 

Method. — Fig. 19 shows the arrangement of the apparatus. 

Bring one of the alternators as G^ up to its rated speed and 
voltage, and connect it to the dis- 
tributing circuit by the switch M, 
Next, throw G^ into synchronism 
with G^ and the line. Adjust the driv- 
ing motor of G^ so that it will take 
some share of the load. 


*— * — n- 

In order to adjust the excitations 
for greatest economy increase that of 
the machine carrying the greatest cur- pio. 19. 

rent and diminish that of the other 

until the sum of the readings of the ammeters /, and /, is a mini- 
mum. Repeat this process with different relative values of driv- 
ing motor power. 

N. B. — During the adjustment of the alternator fields keep the 
A.C. line voltage and the field currents of the driving motors 

Report. — In addition to the usual description of the experiment, 
give the value of the driving motor field currents, and the corre- 
sponding values of the alternator field currents at the condition of 
maximum economy. Give also the readings of the ammeters /j 
and /,, the A.C. line voltage and the frequency. 

§ 13. Tests on Transformers. 

Experiment (16). Determination of tfte Core Losses. 

These losses are due to hysteresis and eddy currents owing to 

the oscillations of the magnetic flux. In a 
well designed constant potential transformer 
they are practically independent of the load. 

Method. — To determine the core losses of a 
transformer, connect the low voltage side to the 
alternator, as shown in Fig. 20. Adjust the 
voltage and frequency to the rated values. 
The magnetization will then be that for which the transformer was 
designed, and the corresponding core losses will be present. The 

Fig. to. 


copper losses however will be negligible. This is because the cur. 
rent flowiag is only a small fraction of the full load current, while 
the copper losses are proportional to the square of the current. 
The wattmeter will consequently read the core losses. 

The energy lost in the pressure coil of the wattmeter should, 
however, be subtracted from this reading. 

It will be found that a slight alteration of the frequency will 
produce no appreciable change in the core losses if the voltage is 
kept constant. This is because 

and at constant voltage, the induction is inversely proportional to 
the frequency. 

If there were heavy leakage in the transformer, the core losses 
might be different depending on which coil was connected to the 
alternator. This, however, will not occur in any modern constant 
potential transformer. 

Expttimtnt (17). Determination of the Copper Losses. 
In order to find the copper losses above, it would be sufficient 
to determine the resistance of the transformer windings, and the 
current in tach of them. There are, however, additional losses 
due to local hysteresis and eddy currents, caused by the currents in 
the windings. These are known as ■' load losses," and they are 
measured together with the copper losses by the method of this 
experiment. It is probable that these load losses arc greater at 
short circuit than under normal conditions. This method of 
measurement will therefore make the losses appear a trifle greater 
than they are under working conditions. 

Method, — Connect the high voltage coil of the transformer across 
the terminals of the alternator as shown in the diagram. Short 
circuit the low voltage coil, making the con- 
nection of very low resistance. The reading 
of the ammeter may be varied by means of 
the alternator field rheostat. Take a series of 
readings of the wattmeter at different values 
of the current in the low voltage coil. 
Fm. 31. These wattmeter readings will give the copper 

losses together with the load losses. The 
iron losses will be practically absent because the magnetization is 
very low. 


N. B. — Run the transformer at its rated frequency. Take eight 
Tcadings at current values ranging up to 50 % overload. 

Report. — Plot the results in the form of a curve with watts as or- 
dinates and amperes as abscissae. 

Experimenl {18). Deferminaiion of the Efficiency Curve. 

This may be done by the following methods : 

(a) Noting imput and output (2-wattmetcr method). 

{b) Noting output and knowing losses from data of experiments 
(16) and {17). 

Method {a\-~:T\i\'i method, as shown in the diagram, involves 
the use of two wattmeters. Because of the high difference of po- 
tential on one side of the transformer, some form of voltage trans- 
former must be used in connection with the pressure coil of the 
wattmeter W, in the high tension circuit. The transformer T' in 
Fig. 22, in which the ratio of transformation is known, steps the 
voltage down to a value suited 
to the capacity of the pressure 
coil of W,- In order to deter- 
mine the true output, the read- 
ing of the instrument must be 
multiplied l^ the ratio of trans- 
formation of T'. Yvi. aa. 

A second me^od of accom- 
plishing the same thing consists in putting a multiplier, or very 
high resistance, in series with the wattmeter. The instrument 
must be calibrated under these conditions. 

In any case care must be taken to avoid the disturbing effect of 
static charges in the instrument. There must be no great differ- 
ence of potential between the two coils of the wattmeter, or be- 
tween the coils and the case. This can be accomplished by ar- 
ranging suitable metallic connection between the current and 
voltage coils. It may be necessary to ground the case if it is me- 

It is necessary for the best accuracy and also advisable as a 
check, to get some idea of the energy lost in the potential coil of 
the wattmeter and the auxiliary transformer T', in order to take it 
into account if it is large enough to make this worth while. This 
lost energy may be approximately determined by noting the read- 
ing of W^ on open secondary circuit and comparing it with the 


reading when T^ is connected in ; there being, of course, no cur- 
rent in r, while this observation is being made. 

A further correction consists in subtracting from the input the 
losses in the pressure coil of W^, 

Method, — ^The actual carrying out of this test is similar to that 
of every efficiency determination. Keeping the impressed voltage 
and frequency constant, and note the two wattmeter readings at in- 
creasing loads. The load must be non-inductive, as this is the stand- 
ard condition usually prescribed in determinations of efficiency. 
Note the readings of the ammeter / in order to be able to calculate 
the power factor. Carry the test up to 50% overload and take eight 

In performing this experiment great care must be exercised with 
regard to the high voltage wires. These must be kept distinct and 
apart from the low voltage connections. 

N. B. — Keep impressed volts and frequency constant. 

Report.^'YXcA two curves: (i) Efficiency in per cent.; (2) Power 
factor. Draw both curves on the same sheet, making horse-power 
output the abscissa. 

Give the values of the various corrections described above, and 
take then into account in determining the curves of efficiency and 
power factor wherever the corrections are greater than the errors 
of observation. 

Method[b), — The efficiency curve by this method which is known 
as the stray power method, may be obtained directly from the data 
of method {a), in connection with that of experiments (16) and 


T-/Y- . output 

Efficiency = — : — — -, 

output + losses 

Plot the efficiency curve thus derived on the same sheet and to 
the same scale as the efficiency curve of {a). 

Experiment (19). Heat test by the motor dynamo method. 

The advantage of this method lies in the fact that the conditions 
of full load are practically reproduced as far as heating is con- 
cerned, without any considerable waste of energy. It requires a 
second transformer, having the same voltage, output and ratio ol 
transformation as the one under test. If this second transformer 
cannot be had, the heat test may be performed by running the 
transformer under load in the ordinary way. 



IT ^5> 

The measurement of the temperature of the coils is done either 
by measuring their resistance with a wheatstone bridge, or by the 
fall of potential method, using direct current. The latter is the 
quicker method, and the most accurate for low resistances. 

Method. — ^The low voltage coils are connected in parallel, and 
the high voltage coils are connected 
so that their E.M.F.'s oppose each 

Close switch m and adjust the 
voltage to the rated value of the low 
potential coils. Full magnetization 
and the corresponding core losses will 
then be present. The difference of 
potential between the two outside 

wires on the high voltage side will, however, be practically zero, 
because the two E.M.F.'s are opposed to each other. 

Close switch m' and adjust the current / to its full load value. 
Full load current will also flow in the low voltage coils. The work 
done by the current / will be the full load copper losses. The 
heating of the transformer will be substantially the same as under 
practical working conditions, because the same losses are present. 

Insert a thermometer in the transformer, and determine the 
temperature of the interior. The temperature of the coils is best 
found by resistance measurements. The relation between tempera- 
ture and resistance is expressed in the following way : 

Fio. 23. 

t^t, + 


In this expression 

i?j = res. at beginning of test. 

/j = temp, of coil at beginning of test, in degrees Centigrade. 
R = res. at any subsequent temp. t. 
In order to express the result in degrees Fahrenheit, 

Degrees Fahrenheit = Degrees Centigrade x f + 32. 
In order that the initial temperature may be known, the trans- 
former should be left over night in the testing room which should 
be kept at constant temperature. The temperature of the coils 
will then be the same as that of the air. 

Carry on the test until the resistances have ceased to increase 
and the temperature has become stationary. 


During the test, measure the resistance of the primary and 
secondary coils and read the thermometer at suitable intervals of 
time, depending on the size of the transformer. A large trans- 
former will, naturally, take longer to reach its stationary tempera- 
ture than a small one. Take six sets of observations of tempera- 
ture and resistance. 

N. B. — Keep E^ /, and the frequency constant. 

Report. — Plot three curves on the same sheet : 

1. Temperature of primary winding. 

2. Temperature of secondary winding. 

3. Temperature of interior. 

State whether oil is used in the transformer or not. 
In plotting the curves, make temperature the ordinate and time 
the abscissa. 

§ 14. Curve Tracing. 

Determination of the Voltage and Current Curves of a Iransformer 

under Various Conditions, 

Experiment (20). With tfie Secondary Circuit Open^ Determine the Fol- 
lowing Curves : 

{a) Primary E.M.F. 

(^) Primary current. 

{c) Secondary E.M.F. 

In a transformer, the phase relations and wave shape of the ex- 
citing current are of considerable interest, as they present pecu- 
liarities which are not yet thoroughly understood. The angle of 
lag is independent of the frequency, and varies with the induction 
and the nature of the iron. If the E.M.F. is a sine wave, the cur- 
rent will be distorted owing to the presence of upper harmonics, 
but considerable changes in the frequency will make no very 
marked changes in its shape. 

The form of the primary current wave depends, in general, upon 
the nature of the secondary current. The primary current is the 
result of the complex harmonic exciting current, and of a compo- 
nent which depends on the secondary current output. With a 
heavy lamp load, the primary current will be almost in phase with 
the E.M.F. and similar in form. This is because the exciting cur- 
rent is then so small compared with the total current that its effect 
is negligible. 

The shape of the E.M.F. wave depends on the design of the 



armature of the alternator and on the field distribution. A smooth 
core armature will, in general, give a near approach to a sine wave. 
It is desirable in these experiments that the E.M.F. curve should 
be sinusoidal, as that is the standard form of wave. 

The primary and secondary terminal volts will always be found 
to differ in phase by exactly i8o°, except for the drop in the 

§ 15. Very many methods of curve tracing have been proposed. 
That which is. described in this experiment embodies the funda- 
mental features which appear in most of the others. The principle 
involved is that the required curve may be obtained by balancing 
the instantaneous value of the alternating E.M.F. against a known 
direct E.M.F. 

Metfiod. — In Fig. 24 the connections are shown for determining 
the curves of this experiment. R^ is a non-inductive resistance 


I ill 




Fig. 24. 

sufficiently large to give a drop of about 7 volts when traversed 
by the exciting current of the transformer 7. The drop across R^ 
will be proportional to the current at any instant, and the curve of 
this drop will be the same as the curve of the current. 

R^ and R^ are adjustable non-inductive resistances placed in series 
with the lamp boards /, and /,. By taking the drop across -^^ and R^ 
it is possible to obtain a fraction of the primary and secondary 
voltages, respectively. 

D is a, drum rheostat filled with a sliding contact m. By mov- 
ing m along the drum it is possible to obtain a gradually varying 
D. C. voltage, which may be read on the voltmeter E^. This vol- 


tage is used to balance against the drops at /, E^ and E^ in deter- 
mining the curves. 

/, E^ and E^ are receptacles into which the spring jack b may 
be inserted. / is a telephone receiver. 

a represents a disc mounted on the shaft of the alternator. The 
brush C is mounted so that its angular position may be varied. 
Its use is to pick out instantaneous values of the E.M.F.'s /, E^ and 
-£j. Contact will be made at n at the same point in each successive 
period ; thus giving the corresponding instantaneous value of the 
E.M.F. Successive instantaneous values may be obtained by 
shifting the angular position of the brush C. 

Insert the spring jack b in any one of the receptacles as /. 
A note will be heard in the telephone receiver unless the voltage 
E^ exactly balances the drop across R^ at the instant when the 
brush C makes contact at n. 

Move the point m along the drum until silence is obtained in the 
receiver. The reading of the voltmeter E^ will then be equal to 
the instantaneous values of the drop RJ corresponding to the 
angular position of the brush C, 

By taking a series of readings in this way at successive positions 
of C, the desired curve may be obtained. 

Proceed in this manner in determining all the required curves. 
In order to work more quickly, take a reading of lE^ and E^ for 
each setting of the brush C. 

Since only the form of the curves, and not their magnitude is 
required, it will not be necessary to know the value of R^, r^ and r,- 

N. B. — Keep the impressed voltage of the transformer, and the 
frequency constant. Take readings every io° for one half of a 

Report, — Plot all the curves on the same sheet. 

The curves must be plotted so that time is counted from left 
to right. Whether the readings run in this way or not will depend 
on the direction of rotation of the contact disc relatively to the 
rotation of the machine. This is an important point, as it involves 
the phase relations of the curves. Care must also be taken not to 
plot the curves so that some appear upside down, which will be the 
case if negative values are mistaken for positive. 

Experiment (21). With Full Load on the SecoTidary Determine the 

Following Curves, 
[a) Primary EM, F. 

{b) «* Current. 



{c) Secondary E.M.F. 

(d) " Current. 

These curves as well as those of the next two experiments are 
to be determined by the method described below, in order to save 
time. It will be found that by loading the transformer with an 
inductive load in which the inductance contains no iron, the prim- 
ary current is correspondingly increased, and becomes practically 
of the same form as the E.M.F. The phase of the secondary 
E.M.F. will lag a little more than i8o° behind the primary E.M.F. 
owing to the drop in the coil. It must be remembered that the 
terminal volts are what we measure and not the induced volts. 

The shape of the E.M.F. wave will probably be altered owing to 
armature reaction. 

It should be borne in mind that primarily the shape of the 
E.M.F. depends wholly on 

the design of the machine. P I ? | ^JS" 1 

The sine wave is only to be 
obtained from a smooth 
core armature. 

Method. — In Fig. 25 G? is 
the source of alternating 
potential, and T'is the trans- 
former under test. 5 is an 
inductive load, t^ and /, 
are small transformers with 
open magnetic circuits.* 

E^ is a direct current volt- 
meter, connections being 
made directly to the mov- 
able coil. J9 is a disc 
mounted on the shaft of 
the alternator. This disc 
is so constructed that the circuit through E^ is completed by means 
of the brush Cfor one-half of every period. Cis arranged as in 
the method of experiment (20), so that its angular position may 
be varied at will. 

As an illustration of the method let it be required to obtain the 
curve of the impressed E.M.F. E^. 

* See F. Townsend on «« A Ne«ir Methol of Tracing A.C Cunrei.*' T>ans, A. I.E.R.^ 
Vol XVII., p. 75. 

Fig. 25. 


Connect the terminals 00 to the binding posts of j2,i and throvr 
the switch n to the left. Then, owing to the fact that R^ is a 
high non-inductive resistance, the current in the coil Q^ will be of 
the form of the impressed E.M.F. E^, 

The mutual inductance of the primary and secondary coils of / 
being small, the current produced in Q^ will have no appreciable 
effect on the current in Qy 

Owing to the resistance R^ the current through Q^ will be pro- 
portional to its E.M.F. during the time when it is not interrupted 
by the contact maker D. It will therefore be proportional to the 
rate of change of the current in Q^ or to the rate of change of E^, 

Since the current passing through the voltmeter E^ is therefore 

proportional to —-i, the deflection produced will be 




The curve of E^ may therefore be obtained directly from the 
readings of E^ corresponding to successive angular positions of the 
brush C. 

To obtain the curve of the secondary E.M.F. E^, throw the 
switch n over to the right, and proceed as in the case of E^ 

To determine the primary current curve, connect the terminals 
00 to the binding posts of P,, close switch b, and throw m over to 
the left, leaving switch a open. As before, the current in /\ will 

be /j, and the current in P^ will h^K^ -J. The deflection of E^ 

will therefore be 

To obtain the curve of the secondary current 7^. close a, throw 
m over to the right, open ^,and proceed as in the previous cases. 

Report, — Plot all four curves on the same sheet, being careful to 
plot the results correctly, both with regard to their sign, and to the 
direction in which the abscissa time is counted, as explained in 
connection with the last experiment. 

Experiment (22). Determination of the cuwes of experiment (21), the 

secondary circuit being connected to the primary 

of a step-down transformer. 

Since the secondary of the step-down transformer is on open 
circuit, the secondary current curve of the transformer under test 


will be a complex harmonic, because it is an exciting current. The 
primary current curve will be practically the resultant of the excit- 
ing currents of the two transformers, and all the complexities of 
the secondary current will appear in the primary. 

The connections in this experiment are the same as those of 
Fig. 25, except that the primary of the step-down transformer 
takes the place of S. 

Experiment (23). Determination of t/ie cutves of experiment (22), the 
secondary of the step-dmvn transfotmer carrying full load current. 

The load should be similar to iT in Fig. 25. 

Both the primary and secondary currents of the transformer 
under test will be increased, and will approach the form of the 
E.M.F. curve. 

Each of them is the resultant of its value as in experiment (22), 
and of a component corresponding to the current drawn from the 
secondary of the step-down transformer. 

The connections in this experiment are the same as in experi- 
ment (22), except that the secondary of the step-down transformer 
is fully loaded. 

§ 16. Resonant Rise of Potential. 

Experiment (24). Connect a Condenser in Series with an Inductance 

to an A.C. Generator. Determine a Curve with Condenser 

Voltage as Ordinate and Frequency as Abscissa. 

The current in the circuit is 



where E is the impressed E.M.F. It is plain that for any values of 
L and C there is a certain value of the frequency which will make 
the impedance a minimum, and 

The period of the impressed E.M.F. will then be equal to the 
natural period of the circuit, and the so-called condition of reso- 
nance prevails. When the resistance of the circuit is ver>' small, 
this period becomes 

7; » 2n^/LC 

TOL. XXII. — 4. 


Under these circumstances the current becomes very great, as 
it is only opposed by the resistance of the circuit. 

The counter E.M.F.'s due to the inductance and condensance 
neutralize each other as they are equal in amplitude and exactly 
opposite in phase ; the former lagging behind the current by 90*^, 
and the latter leading it by a like amount. 

The E.M.F.'s across the inductance and condensance are equal to 

<f)LI and — ^ respectively. As the current is very great, each of 

these E.M.F.'s will attain a high value, even though o}L and --^ are 

themselves small. 

Method, — Connect an electrostatic voltmeter across the con- 
denser. Run the alternator at three-quarters speed and low volt- 
age. Keeping the speed constant, adjust the inductance and 
capacity until resonance is reached. Raise the voltage of the 
alternator by varying its excitation so that the point of maximum 
resonant voltage gives a reading well up on the scale of the elec- 
trostatic voltmeter. The required curve may then be determined 
by reading the condenser voltage at various frequencies. The 
voltage of the alternator must be kept constant by regulating the 
field current. Take ten readings. Vary the speed throughout a 
sufficiently wide range to include the entire rise and fall of the 

N. B. — Keep the inductance, condensance, generator volts, and 
frequency constant. 

Report, — Plot a curve with condenser voltage as ordinate and 
frequency as abscissa. 

Experiment (25). Resonant Rise of Potential Obtained by Means of a 

Superexcited Synchronous Motor, 

The curves to be determined are: 

{a) Voltage at motor terminals as ordinate and frequency as 
abscissa, with constant motor excitation. 

{p) Voltage at motor terminals as ordinate and field current as 
abscissa, with constant frequency. 

In an alternating current transmission system it is possible to 
correct for the drop on the line by connecting a synchronous motor 
with superexcited field in parallel with the load. This effect is due 
to the fact that the leading current taken by the motor neutralizes 
the lagging current on the transmission line. 


If the distributing circuit is open, and the generator is feeding 
the motor alone, the effect is much more marked. 

The conditions may be understood by a consideration of Fig. 26. 

OE^ 5 Voltage at motor terminals. 

OE = " •' generator terminals. 

OE^ 5 •• due to impedance of line. 1^ 

Ol a current in the circuit. \\ A ^^^ 

The motor terminal volts must be the re- 
sultant of the impressed voltage and the *' pj^ ^5 
drop on the line. The diagram shows the 

conditions when the motor excitation has been raised until the 
current leads the voltage at the motor terminals. OE^ is greater 
than OE, 

The condition is one of resonance in so far as there is a periodic 
interchange of energy between the motor and the line. There are, 
however, important differences between the action of a superex- 
cited synchronous motor and a condenser. The counter E.M.F. 
of the motor leads the current, it is true, but it is not proportional 
to it It is, moreover, directly proportional to the frequency. The 
circuit has therefore no natural period corresponding to any given 
values of line inductance and motor excitation. 

The resonant rise of potential depends on the angle d^ and on 
OE . The latter depends on the current. When the motor excita- 
tion is such that OE^ is in phase with the current, 01 lags behind 
OE, and OE is greater than OE^, As the motor field is increased, 
keeping the frequency and the generator volts constant, the current 
diminishes until it comes into phase with OE, and OE^ increases. 
If the motor field is still further strengthened, 01 begins to lead OE, 
and to increase in amplitude. At the same time, both OE^ and 
OE^ are increased. 

This process does not go on indefinitely, for the motor excita- 
tion reaches a point where, owing to increased losses, the angle 6 
becomes smaller. The result is a decrease in OE . 


If the motor field and the generator volts are kept constant 
while the frequency is increased, OE^ will increase because both d^ 
and OE^ are greater. 

Method.— {a) Connect the synchronous motor in series with a 
generator through an inductance of suitable current-carrying 
capacity. Run the machine at half speed and adjust the motor 
field so that with normal generator field the current has a pro- 


nounced lead. Keeping the excitation of the motor and the gener- 
ator volts unchanged, note the voltage across the motor terminals 
at different values of the frequency up to full speed. Take ten 

N. B. — Keep generator voltage and motor field current constant. 
The generator field will have to be adjusted to give the right ter- 
minal volts for each value of the frequency. 

(b) Leave the connections the same as in (cC), Run the machines 
at full speed, keeping the generator voltage and frequency constant, 
Note voltage across the motor for different values of the motor 
field current. Make the range of variation as great as possible 
without allowing the machine to drop out of step at the lower limit, 
or unduly heating the field coils at the upper limit. Take ten 

N. B. — Keep generator terminal volts and frequency constant. 

Report. — Plot two curves : 

{a) Motor terminal volts as ordinates, and frequency as abscissa- 

(^) Motor terminal volts as ordinates, and motor field amperes 
as abscissae. Draw the curves on separate sheets. 

§ 17. Polyphase Currents and Transformations. 

Experiment (26). Balanced two-phase system. 

With full load on each phase and lagging currents, plot curves 
of current, E.M.F., power in each phase, and total power, assum- 
ing that the currents and E.M.F/s are sine waves. 

A balanced polyphase system is one in which the curve of total 
power is a straight line, for in such a system the total power is 
constant. In an unbalanced system the total power is pulsating 
with twice the frequency of the E.M.F. 

Method, — Run a two-phase generator at full speed and voltage. 
Draw an inductive load from each phase, in each of which a volt- 
meter, ammeter and wattmeter must be connected. Adjust the 
values of inductance and resistance in the two circuits so as to bal- 
ance the system as nearly as possible. If the system is exactly 
balanced, the volts, amperes and power factor of each phase will 
be the same. When a fair adjustment has been attained note the 
instrument readings. 

Report, — ^The voltmeter and ammeter readings give effective 
values. If the volts and amperes are assumed to be sine waves, 



their maximum values will be found by multiplying the effective 
values by 1.41. Their phase relations may be found by means of 
the wattmeter readings, since watts = EI cos 6. 

Plot the volts and amperes of each phase as sine waves. The 
voltage curves of the two phases must be displaced in phase by 90^. 
Plot the watt curve for each phase from the products of the instan- 
taneous values of volts and amperes. Plot the total power curve 
by adding the instantaneous values of the power curves of the two 

Plot all of the curves upon the same sheet of cross-section paper. 

Determine the ratio — ~- • This ratio is called the "bal- 

max. power 

ance factor " of the system. If the system is perfectly balanced* 

the value of the balance factor will be unity. 

Experiment (27). Transformation frotn two-pliase to three-phase by 

means of two transformers. 

It is possible to pass from one balanced polyphase system to 
another, by means of two transformers, no matter what the num- 
ber of phases may be. Both primary and secondary systems 
must be balanced, or both must be unbalanced, for the trans- 
formers cannot store energy. 

Fig. 27. 

Method, — T'j and 7, represent the two transformers. The pri- 
mary of each is connected to one phase of a two-phase generator 
G. The secondary of one of these transformers is connected to 
the middle point of the other secondary. 

The secondary E.M.F.'s may be represented vectorially by E^ 
and £/. The points i , 2 and 3 form the vertices of an equilateral 
triangle, and are therefore a source of three-phase current. The 
triangle will be equilateral only on condition that the terms of the 
secondary winding are so proportioned that, 



EJ = £— ^. 

Connect the transformers as shown and measure the voltages 
E^, £;, £,, El, El and E^, 

Report, — Construct a vector diagram, drawn to scale on cross-sec- 
tion paper, showing the phase relation and amplitude of the vari- 
ous E.M.F.'s as in Fig. 27. 

Experiment (28). Three-phase J with two transformers. 

In three-phase transmission systems it is usual to obtain a lower 

voltage at the distributing end 


of the line by means of three 
transformers, one being con- 
nected in each phase. The 
same thing may be accom- 
plished with two transformers. 
The secondary distributing sys- 
tem will then be J or T, ac- 
cording to whether the secon- 
dary windings are connected so that their E.M.F.'s act in series 
with or in opposition to each other, as shown in Fig. 28. 

Method. — Connect the secondaries of the two transformers so 
that their E.M.F.'s act in series. 


Fig. 28. 

^ z\ 

Fia 39. 

Whether they are acting in series may be determined by first 
connecting the two secondaries in series with each other through 
a voltmeter, the primary line No. 2 being temporarily open. If 




Fig. 3a 

the instrument deflects, the E.M.F.'s are acting in series, and the 
secondaries have been correctly connected. If the voltmeter 
shows no deflection, it is because the voltages are in opposition. 
The connections of one secondary must then be reversed. 



Adjust r,, r, and r, so as to draw equal currents from the trans- 
formers. Note the values of the primary and secondary delta 
voltages and line currents as shown in Fig. 29. 

After this has been done, open line wire No. 2, keeping the 
other connections unchanged. Note the line currents and the 
voltages shown in Fig. 30.* 

Report, — Construct vector diagrams similar to those of Figs. 29 
and 30, using the actual readings obtained. Draw these diagrams 
to scale on cross-section paper. Place the diagrams of this ex- 
periment and those of experiment (29) on the same sheet. 

It will be found that the result of breaking the middle wire has 
been to make the secondary system similar to the Edison three- 
wire system. The middle secondary line wire carries the dif- 
ference of the currents of the two outside line wires. 

Experiment {2g), Three-phase Y tvith two transformers. 

Method.— Conn^cX. the apparatus as shown in Fig. 31, arranging 

the secondaries so that their 

— E.M.F.'s are in opposition, and 
not in series as in experiment 

— (28). Adjust r^ and r, so as to 
make the currents //, // and // 

__ equal. Note the primary and 
secondary line currents, the 
primary delta voltages^and the 
secondary voltages i'-2', 2'- 3' and I'-l'. 




Fig. 32. 
The secondary system is an unsymmetrical balanced three-phase 

1 TT 

Fig. 33. 

K One leg of the Fis missing. It is sometimes called the in- 
verted three-phase system. 

* In Figfl. 39 and 30, the points i, 2, 3 correspond to the points i, 2 and 3 of Fig. 28. 
The same is also true of the points i^ 2' and 3'. 


Break the primary line wire No. 2, so as to make /, equal to zero, 
and again note the ammeter and voltmeter readings. 

Report — Construct diagrams coiresponding to Figs. 32 and 33, 
using the actual ammeter and voltmeter readings. Draw these 
diagrams on the same sheet of cross-section paper as those of 
experiment (28). 

It will be found that breaking the middle primary wire brings 
points i' and 3' to the same potential. The current // becomes 
equal to the sum of // and /,'. 

§ 18. Test of a Three-Phase Induction Motor. 

Experiment (30). 

An induction motor is similar in its action to a transformer in 
which there is magnetic leakage.* 

If the primary impressed E.M.F. E^ is constant and the resist- 
ance of the primary circuit is negligible, the total number of lines 
of force is a constant. At any instant, 

a ' a 

In this expression N is the total flux. A^^ re- 
presents the flux at zero slip and zero secondary 
current. Owing to the small air gap, N^ is nearly 
all mutual to the primary and secondary wind- 
ings. N^ is the stray field which is blown out by 
the secondary current, and linked with the pri- 
mary turns alone. N^ will be in phase with the 
^* primary current, while N^ will be in phase with the 
resultant of both primary and secondary currents. 
The diagram of the induction motor on constant potential cir- 
cuit is shown in Fig. 34. 

OD = A/; 

OE, = impressed E.M.F. 

Let the primary current in phase with N^ be represented by 01,, 
The secondary current, whfch will be assumed to be in phase with 
the secondary E.M.F., and at right angles to iV , is given by the 

line BI,. The current producing the useful flux N^ being the re- 

♦See Hcyland in EUctrotechniscke Zeitsckri/e, 1S94: also «« Polyphase Electric 
Currents " by S. P. Thompson, Chap. VI 11. 


sultant of 01^ and BI^, is represented by the line OB in phase with 

The energy taken by the motor is given by the length O IV, 

The lines 01^ and l)l\\ represent the current and energy taken at 
no load. If there were no losses in the machine, the energy at no 
load would be zero. The secondary current and the slip would be 

zero. The primary current would be represented by Om, produc- 
ing the flux N^, which would then be equal to the total flux N. 
Fig. 35 shows the following important properties : 
{a) The loci of the points I^ and B are circles. This is be- 
cause the angles Dl^nt and mBO must always be right angles. 
(b) The primary and secondary currents, represented by the lines 

(?/j and BI^, increase with the load. 

{c) The useful flux N^ decreases with the load. 

{d) The stray field N^ increases with the load. 

The torque of the motor depends on the product I^N^ , /, being 
the secondary current. As the load comes on it is necessary that 
this product shall increase. This is accomplished by an increase 
in the slip, which increases /, because the secondary E.M.F. and 
therefore /, are proportional to the slip. 

At first, as shown in the diagram, /, increases more rapidly than 
N^ diminishes, so that upon the whole, I^N^ becomes greater, and 
the torque increases with the slip. 

As the slip goes on increasing, however, a point is reached at 
which the product I^N^ ceases to increase. If a still heavier load 
is applied, the motor will stop. This is due to the fact that /, has 
become very great, and has blown out the useful flux to such an 
extent that there is insufficient torque. It is on this account that 
at starting, when the slip is lOO per cent., it is usually necessary to 
insert a starting resistance in the secondary circuit to cut down /,. 

Both /, and N^ are proportional to E^. It follows that the 
torque of the motor varies, within limits, as the square of the im- 
pressed volts. 

From the above discussion of the action of the motor, it appears 

{a) Unless specially designed, the motor will lose its torque and 
stop, if the load causes the slip to increase beyond a certain max- 
imum value. 

{b) The motor is very sensitive to a variation in the voltage, and 



it is liable to lie down when heavily loaded, if the voltage is al- 
lowed to drop below a certain point which will depend on the de- 
sign of the machine. 

The curves to be determined in this experiment are : 
{a) Commercial efficiency. 
{b) Apparent efficiency. 
{c) Current per phase. 
{d) Speed in per cent, of synchronism. 
{e) Torque. 
(/) Power factor. 

Method. — If the three-phase system is perfectly symmetrical it 
will only be necessary to take measurements on one phase. If 
this is not the case, however, the connections for the test may be 

arranged as in Fig. 35, so as to 
make measurements on all three 

Load the motor by coupling it 
to a generator whose efficiency 
curve is known. The output of 
the induction motor will be equal 
to the output of the generator 
°* ^^' plus the losses of the generator. 

Note the ammeter, wattmeter, voltmeter and tachometer read- 
ings at ten diflferent outputs, ranging from zero to 50 per cent, 
overload. In addition to the three line voltages, note the voltages 
between the neutral point and the terminals of the motor. 

In order to read the watts in each phase, and thus obtain the 
total watts, insert the terminals i and 2 of the current coil of the 
wattmeter successively into the correspondingly marked binding 
posts of the switches (a), {p) and (^).* The pressure coil of the watt- 
meter is connected between one of the current coil terminals i and 
2, and the artificial neutral point N. 

/j, /j and /j are 32 c. p. lamps. The energy taken by them should 
be subtracted from the motor input. 

In order to obtain a reading, open the switch to which the watt- 
meter terminals have been connected ; the current will then flow 
through the instrument instead of through the switch. 

Care must be taken to insert the wattmeter in each phase in 
the same sense. This is due to the possibility that with an un- 
symmetrical system of voltages one of the phases may feed back 

* See D. C. Jackson, •* Altemating Current Machinery," pp. 556-570. 


into the line at light loads. A negative wattmeter reading must 
be subtracted from the sum of the other two in calculating the 
total watts input. 

In general, when measuring the watts by this method, the total 
watts are given by the algebraic sum of the three readings. The 
artificial neutral point is provided because the real neutral point 
being within the motor frame, is inaccessible. 

N. B. — Keep the impressed volts and frequency constant. 

Report, — {a) Commercial efficiency™ — -^~— . 

The input is obtained by adding the three wattmeter readings. 
The output is equal to the output of the generator coupled to the 
motor plus the losses of the generator. These may be derived 
from an efficiency curve of the machine in question. 

{b\ Apparent efficiency « < ~ r— . 

^ "^ amperes x volts 

{c) Plot three curves giving the amperes in each phase. 

{d) Plot a curve of speed in per cent, of synchronism. 

The synchronous speed is equal to the frequency multiplied by 
60 and divided by the number of pairs of poles of the induction 

{e) Plot a curve of the values of the torque in ft. lbs. — 

. ,, ^ H.P. output 

ft lbs. torque = . x 33,000. 

^ revs, per mm. x ?r 

(/) Plot the values of the power factor in one phase in the form 
of a curve. If the impressed E.M.F/s are not symmetrical, plot as 

ordinates the values of the fraction 

apparent watts input* 

i^g) Plot the C* R loss in the field windings of the motor in the 
form of a curve. These losses may be calculated from the values 
of resistance and current. 

Plot all of the above curves upon the same sheet, taking horse- 
power output as the abscissa throughout. 



§ 19. Converters. 
I. Running from the A. C. End. 

Experiment (31). Various methods of symhronizing a polyphase con- 

verter. Determine the efficiency and external characteristic 

cufves at the D, C. end. {a) Using a two-phase 

converter ; {b) the same machine 

running single phase. 

In a direct-current motor dynamo the armature reaction is almost 
entirely absent. Except for that portion of the current which over- 
comes the losses of the motor the effect of the amperes input is 
completely neutralized by the amperes output. In a converter 
this is theoretically not the case unless the number of phases is in- 
finitely great. In any particular inductor the current entering the 
A. C. end is a sine wave. That delivered at the D. C. must also 
pass through this same conductor. It will be of opposite sign to 
the current from the A. C. end ; it will also be alternating. Instead 
of being a sine function, however, its value is constant throughout 
each half period. It is for this reason that the alternating current 
supplied is not by any means completely neutralized by the direct 
current output. Both the f^R losses and the armature reaction 
of a converter at a given output vary in an inverse proportion to 
the number of phases. 

A polyphase synchronous motor has the advantage of being 
more easy to synchronize than a single-phase machine, owing 

Fjg. 36. 

to the fact that it acts more or less strongly as an induction motor 
owing to the eddy currents developed in the pole pieces by the 
rotary field. This is a quality which is useful in running the ma- 
chines in parallel. 

Method. — Before determining the curves of efficiency and ter- 


zninal volts, synchronize the converter by the following methods 
and note the action of the machine. 

1. Start the converter as a direct-current motor from a constant 
potential circuit. When the machine is in step, as shown by the 
synchronizing lamp /,, switches I and 2 should be closed. Switch 
i' must have been closed in the first place in order to allow the 
synchronizing lamp to light up. 

2. Open the field circuit of the converter, and start as an induc- 
tion motor. When the machine is up to speed, close the field 
circuit. The converter should then run in synchronism with the 
two-phase generator. Care must be taken before closing the field 
switch to adjust the value of the field resistance correctly. 

3. Keeping switches I and 2 closed, bring the speed of the 
generator slowly from standstill up to its rated value. The con- 
verter should speed up along with the generator. 

{a) Determination of the efficiency and external characteristic 
curves, running as a two-phase converter. 

The field current must be kept constant at the value which makes 
the current input a minimum at no-load. Run the generator so 
that it gives the rated frequency of the converter. 

The machines may be loaded by means of a resistance con- 
nected across the terminals of the D. C. end. It is more economical, 
when possible, to feed into the D. C. power circuit if the potential 
of the circuit is lower than that of the converter. The method of 
doing this is shown in Fig. 36. Vary the load by adjusting r^. 
Take eight readings of input and output, carrying the test from 
no-load to 50 per cent, overload. 

{b) Determination of the efficiency and external characteristic 
curves running as a single-phase converter. 

Open one of the two phases, and the rotary will continue to run 
single-phase. Note the readings of input and output at the same 
loads as in (a). 

N. B. — Keep converter field current, frequency and impressed 
volts constant. 

Report. — Plot the efficiency curves of {a) and {p) on the same 
sheet, taking the values of efficiency in per cent, as ordinates, and 
of horse-power output as abscissae. 

Plot the readings of the voltmeter at the D.C. end in [a) and {d) 
as ordinates, with respect to amperes output as abscissae. The ex- 
ternal characteristics of both curves are to be plotted on the same 


Experiment (32). {a) At Full Load Determine the Relation Between 

the D.C. Volts and t/ie Field Current. 

(Jj) Insert an Equal Inductance in Each Phase ^ and Determine the 

Same Curve, 

In a converter the voltage at the D.C. end is the same as the 
counter E.M.F., except for the IR drop and armature reaction. 
The counter E.M.F., however, is almost equal to the impressed 
volts. It follows therefore that when running from a constant po- 
tential line, the volts at the D.C. end are not appreciably altered 
by varying the field current. 

This is not, however, the case if there is inductance between the 
converter and the A.C. generator (see § 16, exp. 25). The voltage 
at the A.C. and D.C. ends can then be raised by means of the field 
current owing to the fact that the leading current taken by the 
A.C. end causes a resonant rise of potential. This fact is brought 
out in this experiment. 

Metliod {a), — With the same connections as in Fig. 36, at full-load 
direct- current output, vary the field of the converter, and note the 
D.C. voltage. (^) Insert an equal inductance in each phase, and 
note the voltmeter readings at the same field strength. Take four 
readings in each case. 

N. B. — Keep the generator terminal volts, the frequency and the 
direct- current amperes constant. 

Report, — Plot two curves \ci) and {p) on the same sheet with volts 
at the D.C. end as ordinates and field amperes as abscissae. 

§ 20. II. Running from the D.C. End. 

Experiment (33). At Full Load, Non- Inductive, Determine the Curves 
of Speed and A.C Voltage in Relation to the Field Current, 

It will be found that as the field current is increased the speed 
will drop, buc the voltage will remain almost constant. 

Method, — Make the connections as in Fig. 37, running the ma- 
chine single-phase for the sake of convenience. Adjust r^ so as to 
draw full-load current, and leave it unchanged throughout. Vary 
the field current through as wide a range as may be possible with 
safety and note the values of the voltage at the A.C. end and the 
speed. Take four readings. 



N. B. — Keep the impressed volts at the direct-current end con- 
Report, — Plot two curves on the same sheet ; one with volts at 





— I ' ysr^ 


Fig. 37. 

the A.C. end, and one with speed in revs. per. sec. as ordinates. 
Take the field current as abscissa for both curves. 

Experiment (34). With lagging current, vary the output while keep- 
ing the power factor approximately constant. Detennine 
a curve between speed and current output. 

The effect of a lagging current upon the field of an A.C. gener- 
ator was discussed in section 6. Its action is to blow out the 
lines of force far more than if the current were in phase with the 
E.M.F. Increasing the output with a lagging current weakens the 
the field. This, in a converter running from the D.C. end, causes 
the speed to rise. This will still further increase the lag owing to 
the increased reactance, and the converter is liable to run away. 

Method, — Use the connections shown in Fig. 37, except that an 
inductive load must be substituted for r^. Keeping the power 
factor constant by varying the inductance and resistance of the 
circuit, increase the current output, and note the values of the 
speed. Take four readings. 

N. B. — Keep the converter field and impressed volts constant. 

Report, — Plot a curve having speed as ordinate, and current out- 
put at the A.C. end as abscissa. 

Tabular View of Experiments. 

As the experiments in many instances require complicated con- 
nections, the following table is added in order to facilitate the 
work. The experiments are here divided into sets which may be 



performed successively without altering the arrangement of the 
apparatus to any great extent. The approximate time required by 
a student for each set is given in a separate column. 

No. of cxpcrimcnw. 

Hours required. 

No. of experiments. 

Hours required. 

I, 2 

3 : 





21, 22, 23 

5»6, 7 










26, 27 


II "1 


28, 29 


12, 13, 14 




L 15 J 




i6, 17 


33, 34 





will depend on size 

of transformer. 

— . . 


[Contributions from the Havemeyer Laboratories of Columbia University, No. 38.] 


By C. H. JOUET. 

Part I. Lead and Copper Slags. 

This paper is intended as a guide for the analysis of the various 
slags and cinders from blast and reverberatory furnaces, producing 
pig and refined lead, copper and iron, and to condense such infor- 
mation for the assistance of the student and the metallurgical 
chemist. Complete schemes for the analysis of the various types 
of slags are given first and are followed by the rapid methods em- 
ployed at smelting establishments. Slags contain many rare ele- 
ments and from a practical standpoint, it is not necessary to search 
for them but rather to determine the principal constituents for the 
daily use of the furnace manager. 

In the analysis of a lead slag, the constituents usually deter- 
mined for the furnace manager are silver, lead, silica, iron, lime, 
magne.sia and zinc, but alumina, copper, sulphur, arsenic, antimony, 
bismuth, are usually present and are sometimes estimated. 

a- 1 


ud moisten thoroughly with a few drops of water, cover, add lo c.c. of 
I the slag^ and decomposition has begun, shown by the effervescence, this 
I may prevent further decomposition, if the slag has not been chilled and is 
issdved silica with a mixture of Na,CO, 20 parts and NaNO, i part If, 
I I ex. of cone, HNO,, cover and evaporate to dryness on asbestos plate, 
with 5 CO. cone. HQ and dehydrate again, take up finally with lo c.c. cone. 
water to lOO c.c, filter on 7 cm. ashless filter paper with suction pump, 
itinum crucible over burner or in a clay annealing cup in a closed muffle, 
n some cases, as in hard lead slags, the silica may be contaminated with 
tad' 3 drops of cone. H,SO^. Barium sulphate may be occasionally found 

tioQ, in order to estimate these constituents, after filtering the silica, etc., and 
^flien wash once with dilute NH^OH, then thoroughly with hot strong 
oxide and calcium sulphate will be dissolved, if present 

Filtrate contains all the AgCl, PbSO^ and some Sb,0, and CaSO^. Pass 
in H,S gas, acidify with HCl, warm, pass more gas, filter out ppt of PbS + 
Ag^ + Sb,S, on a small paper (Filtrate i), wash on paper with (NHJ,S, 
(Filtrate 2) and dissolve residue in hot dilute HNO, (not over 25 per cent. 
HNOJ, filter and add a pinch of common salt and if any precipitate of 
AgQ, filter on weighed Gooch crucible, wash with water, dry and weigh 
and calculate Ag. Evaporate the solution to fumes of SO, with 10 c.c. 
cone. H,SO^, cool, take up with cold water, settle and filter out PbSO^ on a 
small a^less paper, dry, ignite and weigh.f Wt PbSO^ x .68293 « wt. 
of Pb. 

Filtrate z (from above treatment) contains some calcium sulphate, add 
(NHJ^C,0^ in excess, make ammoniacal and boil, filter any precipitate of 
CaC,0^ and discard filtrate. Reserve ppt. to be added to main ppt. of 
oxalate of lune. 

Filtrate s (from above treatment). Acidify with HCl and pass in some 
H,S. Ppt «■ Sb,S, + S, filter and wash, discard filtrate and dissolve pre- 
cipitate through the paper with strong (NHJ,S, (the reagent should leave 
no residue when a quantity is evaporated to dryness) into a weighed porcelain 
crucible, evaporate to dryness. Cover with fuming HNO,, ignite at first 
gently and after sulphur is oxidized, add 2 c.c. fuming HNO, and ignite at 
high heat Weigh as Sb,0^. Wt. Sb.O^ x .7901 ^ wt. of Sb, or dissolve 
Sb,S, in K,S through the paper into a small Erlenmeyer flask and acidify 
with HCl, add KCIO,, boil out free CI, cool, add j^ gm. KI and titrate the 

free I with- Na.SAsJ^Ps^'"^on- 

I c.c. SB .012685 gm. I. 
I cc. « .006021 gm. Sb. 

tAmni filter, aad.wadi with H,S water. 

PUtrate. Add NH,C1. NH.OH, unHl alkaline and (NHJ,S. digest and 
filter, and wash with diluted ammonium sulphide. 

Filtrate. CaCI,,MgCI,. Acid- 
ify with HCl, boil out H,S, make 
ammoniacal, add (NH^),C,0^ in 
excess, to the hot solution and 
boil }i hour, settle and filter on 
small ashless paper, wash well 
+ hot water, ignite and treat 
+ H,SO,. Weigh as CaSO,. 
Wt. of CaSO, X .41186- wt. 
of CaO or pour on the paper 
hot dilute sulphuric acid (1:5) 
and wash well, titrate the hot 

bS; Precipitate. NiS, CoS. FcS, MnS, ZnS, 
ii Al^OH),. Treat ppt. with cold dilute HQ, 
'ij,'ix)2 sp. gr. filter and wash residue on 
\-^ .paper thoroughly with cold H,S water. 

Filtrate. MnCL. 

Residoe. NiS + 

!.• CoS (black). Ifpres- 
li: ent in small amounts, 
lis dissolve in HNO, + 
hi * HCl, filter, wash well, 
irc addioc.c.conc.H,SO^ 
oe Evaporate to fumes of 
Id SO, in No. 2 beaker. 

FeCI^ZnCI,, A1,CI,. 
oxidize with KCIO, 
after making acid 
with HCl, boil out 
free CI, make alkaline 
with Na,CO„ then 
just acid with acetic 


^f m rt W t t ' ^t ^»lt ^ t»^r#kfca'««I *«»« t M m f^ B^* '«li ■»» J^ft ■ t *« f . • h 

1 • 



Treat one gram of the slag with cone. HCl + HNO, (2 : i)and 
when well decomposed, evaporate to dryness and take up with 5 
c.c. of HCl and 30 c.c. of water, boil, and render alkaline with 
barium hydrate solution, warm and filter out the precipitate and 
wash well with water until the precipitate is free from chlorides. 
Add to the filtrate i c.c. of strong ammonium hydrate, and then 
a saturated solution of ammonium carbonate until all of the barium 
in excess is precipitated. Heat and add in fine powder, o.6cK) 
gram of ammonium oxalate. Filter and wash free from chlorides. 
Evaporate the filtrate to dryness in a weighed platinum dish and 
ignite gently over a free flame, below a red heat, until all volatile 
matter is driven off. Digest the residue with hot water, filter from 
any insoluble matter on a 9 cm. filter paper, and treat the filtrate 
with one or two drops of the ammonium carbonate solution, any 
precipitate is filtered off, and the filtrate evaporated to dryness, and 
again ignited after acidification with hydrochloric acid, cool the 
dish and weigh, deduct the weight of dish. Weight =« KCl +NaCl. 
Dissolve the combined chlorides in a few c.c. of hot water, trans- 
fer to a porcelain casserole and add one drop of cone, hydrochloric 
acid and hydrogen platinic chloride solution in excess. This solu- 
tion should contain 2.1 grams of HjPtCl, in every 10 c.c. Evap- 
orate on a water bath to a thick syrup and take up the mass with 
strong ethyl alcohol (80 %), sp. gr. 0.8645, avoiding the absorp- 
tion of ammonia, filter off precipitate of potassium platinic chlo- 
ride on a weighed Gooch crucible (paper in the base) and wash 
free from the precipitating agent with 80 % alcohol, both by de- 
cantation and after collecting on the Gooch or other form of filter. 
Dry for thirty minutes at 100° C. and weigh the K^PtCl,. Instead 
of filtering on a Gooch crucible, an ordinary filter paper can be 
used which should be balanced against another paper and the one 
placed inside of the other for filtration, or the precipitate can be 
dissolved through the paper with hot water into a weighed porce- 
lain dish and evaporated to dryness at 100° C. and the K,PtCl 
weighed. In any case, calculate the K,PtCI, to KCl and K,0. 
Deduct KCl from combined chlorides. 

Wt. of NaCl X .5307 = Na,0, 
Wt. of K,PtCI, X .30695 « KCl, 
Wt. of K,PtCl, X .19394 = K,0. 

TOi« xxn.— 5. 


Lead is usually determined by fire assay, and if present in less 
quantity than i %, the following modification should be followed: 

Take S-io grams of slag, and weigh out about 0.3CX) gram of 
pure silver and add to the charge of slag and lead flux ; 

Charge. Slag, .... 5-10 grams 
Pure silver, . . 0.300 ** 
Lead flux, . . 20 " 
Borax glass, . cover. 

Melt in clay crucible in muffle furnace. Time of fusion about 
twenty minutes. Pour into scorifier mould, separate button from 
the adhering slag. Increase in weight of silver button gives weight 
of lead in the amount of slag taken. The button should be soft 
and malleable. If the slag has any matte, add to the charge two 
nails hung points down. 

The lead flux is made as follows : 

NaHCO,, . . 16 parts. 
K,CO,. ... 16 " 
Flour, ... 8 «« 
Borax glass, . 4 *' 

Silver and Gold. — ^Take i A.T. of slag, add litharge and suffi- 
cient reducing agent to obtain a button of about 6 grams. Cupel 
button and weigh Ag. 

Charge. Slag, . . . . i A.T. 

I A.T. 
I A.T. 

•5-75 gram, 
points down, 

Sod. bicarb. 
Argol, . 
Two nails, 
Salt, . . 

Fuse in muffle furnace about twenty minutes and pour quiet 
fusion into a scorifier mould. Separate button and cupel. Weigh 
Ag + Au. Part with nitric acid C.P. 1.16 sp. gr. and weigh the 
Au. In case of copper slags, part of the copper will be found with 
the button of lead ; the button must then be scorified with test lead 
until a malleable button of lead is obtained which can then be 

Lead (wet way), Separate Determination. — Treat 2 grams of 
slag with 15 C.C. cone. HNO, in a covered casserole until decompo- 
sition is complete, then add 10 c.c. cone. H,SO^ and evaporate to 


fumes of SO,. Cool and add 50 c.c. water and boil to dissolve ferric 
sulphate, cool, filter and wash the residue on the filter with water 
containing i% sulphuric acid and then with strong alcohol, 80^. 
(Use suction pump if convenient) Dissolve the lead sulphate 
through the filter with a strong and hot solution of ammonium 
acetate, wash out the casserole with the same solution. The lead 
may be now determined, either by acidifying the filtrate with sul- 
phuric acid, filtering ofT the white precipitate on a small paper and 
washing with water containing sulphuric acid i^, and finally with 
alcohol, 80%. Dry the paper and precipitate (see note 2), burn 
off paper in a weighed porcelain crucible. After adding nitric 
acid, any reduced lead should be dissolved in nitric acid and the 
precipitate now freed from carbonaceous matter is treated with 
I c.c. H,SO^, and when nitrate is all decomposed, the heat is raised 
to drive off the excess of H,SO^. Cool and weigh PbSO^. Wt. of 
PbSO^ X .68293 = wt. of Pb, or after the lead sulphate has been 
dissolved in ammonium acetate and the solution acidified, with 
acetic acid, the lead may be titrated hot with a standard solution 
of ammonium molybdate, using tannic acid as an indicator accord- 
ing to the method of H. H. Alexander, as described in Eng, and 
Min. Journal^ 1893, Apr., p. 298, and Ricketts and WC^^Xy Assaying, 
3d edition, p. I $0. 

Sulphur in Lead and Copper Slags. — Fuse i gm. of slag in 
a silver crucible with caustic potash (free from sulphur) until quiet 
fusion. Complete decomposition takes place in about 1 5 minutes. 
Cool and dissolve in hot water, bring solution to a boil and filter 
out precipitated bases, wash until the filtrate comes through free 
from sulphides or sulphates. Add about 2 c.c. strong bromine to 
the filtrate and warm for some time to oxidize sulphides. Acidify 
with hydrochloric acid, remove silica by evaporation to dryness, 
take up with dilute hydrochloric acid and filter, heat to boiling and 
add S-io c.c. of boiling barium chloride (10 % solution). Boil for 
15 minutes and allow to settle, filter and wash well with hot water. 
Ignite and treat precipitate with 3-4 drops cone, sulphuric acid. 
Drive off excess. Cool and weigh BaSO^. Wt. of BaSO^ x . 1 3734 = 
wt. of S. 

This method is modified from Fahlberg-Iles method by Furman* 
See Manual of Assayings p. 88. 




Treat i gram of finely divided slag with 15 c.c. nitric acid 1.42 
sp. gr. until the slag is decomposed, take to complete dryness and 
dehydrate the silica at 1 10° C. for half an hour, if the slag is per- 
fectly decomposed and the silica is white, take up with 20 c.c. HCl 
and 30 c.c. water, boil, filter and wash well ; ignite the silica in a 
weighed platinum crucible or in a clay annealing cup in a muffle 
furnace, brushing out the silica for weighing. Expel SiO, with 
HF + HjSO^. If slag is not decomposed by acid treatment, filter 
out siliceous residue and fuse in platinum crucible with Na,CO, 
and dissolve fusion in dilute HCl, evaporate to dryness and de- 
hydrate, take up with a few c.c. of dilute HCl and evaporate again 
to dryness and dehydrate at 1 10° C. ; take up with dilute HCl, 
filter, wash well, ignite and weigh SiO,. Expel with HF + H,SO^ 
and determine SiO, by loss. Any residue from this treatment 
in either case, fuse with Na,CO, and add to main solution. If 
any BaSO^ in residue, leach fusion with water and determine ba- 
rium in the residue by solution of the residue in dilute hydrochlo- 
ric acid and precipitation with sulphuric acid as barium sulphate. 
Weigh as BaSO^. Any residue not BaSO^ should be added to 
the main solution after solution of the same. 

Pass H,S into the main solution (not over 5 c.c. of free acid in 
300 c.c. of solution) — filter and wash with H,S water. 

Precipitate. CuS 
chiefly, Bi,S,, PbS, 
As,S,, Sb,S, -h S. 
Treat on paper with 
K,S^ solution. 

Residue. CuS, Bi,- 
S,.PbS, -hS. Proceed 
for the separation of 
the above as in 
scheme for lead slag. 

Solution. K,AsS^-h 
KjSbS^. See lead slag 
scheme for separation. 

Filtrate, Fe,Cl,. 
MnCl,. ZnCI,. Al.Cl,. 
NiCl,, CoCl,. CaCl,, 
MgCl,. See lead 
slag scheme for esti- 
mation of above. 


These foregoing schemes will serve for the complete examina- 
tion of a slag, but it is not often that such work is demanded from 
the chemist, and as the daily analysis of the slag is for the guid- 
ance of the furnace managers, some rapid methods must be con- 

The slag is drawn as a chilled sample, rendering it soluble, to a 
great extent, in acids. 

The determinations made daily in lead slag are silver, lead, sil- 
ica, iron, lime, zinc and sometimes magnesia and manganese. 

Sih^er and lead are determined by fire assay. 

Silica and Lime, — Weigh I gm. of slag into a No. 2 beaker, 
moisten with water and add 10 c.c. cone. HCl and a few drops 
HNO,. Mix up well to prevent caking. Evaporate quickly to 
dryness, on asbestos plate, dehydrate at not over 1 10^ C, take up 
with 10 c.c. HCl and 30 c.c. water, boil, dilute to 100 c.c. and filter 
on small ashless paper with aid of suction pump, wash well with 
hot water. (If lead sulphate is present wash with acetate of am- 
monium and discard the washings.) The silica may be dark owing 
to admixed carbon. Burn ofT in mufHe in a clay annealing cup, 
cool and brush out silica on a balanced watch glass, weigh and 
calculate the percentage of silica. The silica is sometimes con- 
taminated with oxide of antimony (especially in the case of slag 
from the hard lead furnace) and sometimes with barium sulphate. 
In this case it should be driven ofT with hydrofluoric acid and sul- 
phuric acid and the silica determined by loss. To the filtrate from 
the silica, add one drop of sulphuric acid to precipitate any barium 
present, then heat to boiling and add drop by drop ammonia with 
constant stirring until the Solution is nearly neutral. The solution 
is dark red in color, but should remain clear, now add 20 c.c. of 
ammonium oxalate solution (i part oxalate in 24 parts water), boil 
the solution for a few minutes, settle and test the supernatant 
liquid with a few drops of ammonium oxalate solution. If all of 
the lime has been precipitated, filter rapidly and wash with hot 
water thoroughly. Throw paper and precipitate into about 200 c.c. 
of a hot solution of water and sulphuric acid containing about 10 ^c 

acid. Titrate hot with — KMnO. solution. i c.c. = .0028 gm. 

10 * ^ 

CaO. Calculate the percentage of lime. Manganese interferes 

with this method. 

Iron, — Take 0.5 gm. of slag dissolve in water and cone. HCl, and 


a few drops of stannous chloride solution (not an excess), and 

evaporate to low bulk to remove most of the free acid, add drop 

by drop SnCl, solution until the solution is decolorized, then add 

lo c.c. HgCl, solution. Dilute with cold water to about 300 c.c, 

the precipitate should remain white, otherwise discard solution 

and proceed with a new portion. Add 10 c.c. ''preventive solution" 

(MnSO^, 160 gms. in 1750 c.c. water; H,SO^, 320 c.c. i.82; 

HjPO^, 330 c.c. 1.7 sp. gr.) and titrate with— KMnO^ solution 

I c.c. « .0056 gm. Fe or titrate with — KjCr^O, solution (leaving 

out use of the preventive solution) i c.c. = .0056 gm. Fe. 

Iron (Alternate Method). — Take 0.500 gm. slag, decompose 
with 10 c.c. hydrochloric acid, add 10 c.c. water and weigh about 
5 gms. pure zinc, sufficient to reduce the ferric chloride and to con- 
sume the free acid, and leave a small amount undissolved, then add 
enough dilute H,SO^ to dissolve the remaining zinc ; pour solution 

into 300 c.c. cold water and titrate with — KMnO. solution, cal- 

' 10 

culate percentage of Fe. 

Run blank test on the same weight of zinc used and determine 

the quantity of permanganate consumed. Deduct c.c. — KMnO^ 

used from total number of c.c. 

Zinc, — Take i gram of slag, dissolve in a casserole in 15 c.c. of 
cone, nitric acid and add slowly two grams of potassium chlorate, 
cover and heat gently to drive off the greenish fumes, avoid cak- 
ing as much as possible and evaporate to dryness, avoid baking, 
take up with 5 gms. of ammonium chloride, 20 c.c. cone, ammonia 
and 30 c.c. water, boil two minutes and break up any lumps, filter 
and wash well with hot water, dissolve precipitate in nitric acid and 
if any manganese binoxide is present, boil down to small bulk, 
adding a few crystals of potassium chlorate as before ; if no man- 
ganese, the evaporation and chlorate treatment may be omitted, 
then add ammonia in excess, filter and wash well with hot water. 
Combine filtrates which should not exceed 200 c.c. in bulk. If 
blue from presence of copper, acidify with hydrochloric acid, add 
30 grams of test lead (free from zinc) and shake well until all the 
copper is thrown down. Bring solution to slightly acid state 
(6 c.c. in excess) with hydrochloric acid. 


Titrate hot with standard potassium ferrocyanide solution 
(K^FeCN^, 3H,0, 43.2 gms. in i liter) with indicator of uranium 
acetate, i c.c. K^FeCN^« .010 gm. zinc. Calculate zinc per- 
centage, Cd, Cu, Fe, Mn, Ni and Co interfere and should be re- 
moved before titrating. 

A blank test should be made with same bulk and acidity of solu- 
tion. These operations for silica, iron, lime and zinc can be per- 
formed in two hours. 

Magnesia. — ^Treat i gm. of slag with hydrochloric acid and 
water in the usual way, dehydrate the silica, take it up with HCl 
and, without filtering it out, add ammonia until quite alkaline and 
filter and wash well, precipitate the lime with an excess of ammo, 
nium oxalate, filter it out and evaporate the filtrate to small bulk, 
cool and add excess of hydro-sodium phosphate. Stir well and when 
the crystalline precipitate of magnesium ammonium phosphate has 
separated completely it may be filtered, washed with ammonia 
(dilute) and weighed as Mg,P,0^ after ignition. 

Manganese. — Decompose i gm. of slag in 2 c.c. of hydrochloric, 
4 c.c. of nitric and 7 c.c. of sulphuric acid in a casserole ; if the slag 
has not been chilled, make a fusion with sodium carbonate and 
dissolve fused mass in diluted hydrochloric acid, evaporate to dry- 
ness and heat. Then take up in dilute HCl and evaporate with 
HjSO^ to fumes of sulphuric anhydride. Transfer the contents of 
the casserole to a 500 c.c. flask, washing it with boiling water. 
Then add emulsion of zinc oxide until the acid is neutralized and 
the iron completely precipitated, shaking the contents violently. 
Dilute to SCO c.c. with distilled water, and mix well. Draw off 
with a pipette 100 c.c. of the clear liquid and transfer to a casserole, 

bring the solution to a boil and titrate with — KMnO. solution 
^ 10 * 

until a faint permanent pink color appears in the solution after 
very vigorous stirring when viewed against the white background. 
The value of the KMnO^ in iron multiplied by 0.2946 gives its 
value in Mn. 

Manganese (Alternate Method). — Dissolve i gm. of slag in 
20 c.c. of nitric acid 1.4 sp. gr. in a casserole. Evaporate to dry- 
ness to dehydrate silica, take up with nitric acid and add potassium 
chlorate in crystals about two to three grams in small amounts at 
a time, boil out the free chlorine, and filter through an asbestos filter 
with aid of suction pump (the asbestos should be treated with 


hydrochloric acid to free it from soluble lime, manganese and 
magnesia and should have been previously washed and ignited), 
wash out the casserole with strong nitric acid on to the filter and 
continue the washing on the filter until the washings are colorless, 
and free from lime and magnesia ; discard the filtrate. 

The binoxide of manganese (hydrated) is dissolved in hydro- 
chloric acid + I c.c. H,SO,, filtered from the asbestos, and washed 
well with hot water on the same filter tube and pump, nearly 
neutralize with ammonia, add a few crystals of acetate of soda, 
and boil, filter, wash with hot water (repeat treatment on basic 
acetate precipitate by solution in hydrochloric acid and neutraliza- 
tion and precipitation). Combine filtrates which should be free 
from iron, if not, evaporate nearly to dryness after nearly neutral- 
izing with ammonia and filter out any precipitate, heat to boiling 
and add excess of sodium ammonium phosphate, render slightly 
ammoniacal, and stir until precipitate becomes crystalline. Settle 
and filter, wash well with warm water, dry, ignite and weigh as 
manganese pyrophosphate. Wt. Mn,P,0^ x .38723= wt. of Mn. 

In the examination of copper slags, the following methods will 
be found useful : 

Estimation of Copper. — ^Take i gram of rich slag and 5 grams 
of poor slag, finely ground, moisten in casserole with 10 c.c. of 
water, add 20 c.c. of cone, hydrochloric acid and i c.c. cone, nitric 
acid and boil for several minutes, evaporate to dryness and heat at 
1 10^ C. until the silica is completely dehydrated, cool and add 
10 c.c. cone, hydrochloric acid. The liquid is boiled, diluted with 
water and filtered, washing the residue several times with water. 
If the slag has not decomposed well, the residue should be treated 
in platinum (preferably) with hydrofluoric and cone, sulphuric 
acids to remove the silica and decompose the residual slag, evap- 
orated to sulphuric anhydride fumes, and the mass treated with 
hot water and added to the main solution. To the filtrate is added 
sufficient sodium thiosulphate solution to reduce all of the iron 
and a piece of sheet zinc is added and the solution warmed until 
all of the zinc has dissolved and the copper sulphide precipitated. 
The copper sulphide is filtered off, washed with warm water and 
dissolved in hot dilute nitric acid (1.2 sp. gr.). The copper can 
be determined by making the solution slightly ammoniacal and 
titrating with a standard solution of potassium cyanide of about 
the strength i c.c. = .0050 gm. of copper or the copper can be 


quickly estimated by first adding 20 c.c. of a saturated solution of 
zinc acetate, cooling the solution, diluting to 50 c.c. and adding 
2~3 grams of potassium iodide and titrating the iodine set free 
with a standard solution of sodium thiosulphate (about 19.6 gms. 
per liter of Na,S,0,, 5H,0) using starch paste as an indicator. 

I c.c. of Hypo. = .0050 gm. Cu approximately, it should be 
standardized against pure copper. 

Silica, — Take 0.500 gram of slag, treat with 10 c.c. cone, hydro- 
chloric acid and i c.c. cone, nitric acid. Evaporate to dryness and 
dehydrate, take up with 10 c.c. dilute hydrochloric acid and filter, 
wash twice with hot water, ignite and fuse residue in platinum 
crucible with sodium carbonate until quiet fusion, dissolve in dilute 
hydrochloric acid and evaporate to dryness and dehydrate silica at 
110° C, take up with 10 c.c. hydrochloric acid and 25 c.c. water 
and filter with aid of suction pump. Wash thoroughly with hot 
water and ignite in clay annealing cup in muffle furnace. Cool and 
weigh SiO,. 

Iron, — ^The combined filtrates from the silica are made alkaline 

with ammonia and the precipitated ferric hydrate filtered ofT, 

washed well with water and the precipitate dissolved in dilute hy- 

drochloric acid, and, after reduction, the iron is titrated with — 


KMnO^ solution, i c.c. — KMnO^ = .0056 gm. Fe ; or pass the 

hydrate of iron dissolved in dilute sulphuric acid through a 10 inch 

column of zinc two or three times to reduce the iron and titrate 

with — KMnO. solution. 
10 * 

Ume. — ^The filtrate from the iron is treated with ammonium ox- 
alate in excess for the lime, which is weighed as CaSO^ or the ox- 

alic acid is titrated with — KMnO^ solution. 

10 * 

Gold and Silver. — ^The sample is fused with the following 

charge : 

Slag, ... I A.T. 

Litharge, . i A.T. 

Soda bicarb., i A.T. 

Borax glass, 10 gms. 

Argol, . . . .5-.8 gm. 

Salt, .... cover. 


The resulting button is scorified with test lead until copper is 
removed and the lead button is malleable and soft and then 
cupelled. Weigh Au and Ag, or the button can be dissolved in 
nitric acid and the gold filtered off and the silver precipitated by 
sodium chloride solution. The gold and chloride of silver can be 
treated by scorification with test lead, cupelled and the Au and Ag 
weighed. Part the gold and silver with nitric acid 1.20 sp. gr.^and 
weigh Au. 




By J. F. KEMP. 

ITie importance of chemical analyses of rocks in general and of 
igneous varieties in particular has only been properly appreciated 
within the last ten years. It is indeed true that in the time of 
Abich and Bunsen in the fifth and sixth decades of the present 
century much attention was given to this branch of investigation, 
and that the work and influence ofthe latter made available many 
results ; but interest languished with the passing away of faith in 
his two fundamental magmas — the normal-trachytic and the nor- 
mal-pyroxenic — in the igneous rocks, and the analyses which were 
subsequently made and recorded were either prompted by their 
practical applications, or were merely intended to give a general 
idea of the composition of the rock in question. They were sel- 
dom employed for close mineralogical computations. Those 
geologists who gave the matter attention believed that analyses 
were so variable and were so largely a function of the sample 
taken, that they might differ greatly if the materials were de- 
rived merely from opposite ends of a hand-specimen. They 
therefore gave them comparatively small attention. Even when 
analyses were to a certain extent recast, as for instance in the Re- 
pofts of the Swvey of the Fortieth Parallel, only the percentages 
of oxygen in the SiO,, A1,0,, Fe,Oj, FeO, etc., were deducted 
from the total percentages of the oxides and were used to cal- 
culate the so-called " oxygen ratio ; " that is, the continued ratio 
of the percentage of oxygen combined with the silicon, to that 
combined with the monad and dyad bases, to that combined with 
the triads. The quotient obtained by dividing the sum of the last 
two by the first was called the oxygen quotient and was esteemed 
to be characteristic of the several groups of igneous rocks. 

The following ranges of oxygen quotients would not be far from 
the truth. It is well to add that the higher the silica the lower the 


quotient. Ultra-basic rocks would run even higher than the values 
here given. 

Rhyolites — Granites . 1 75-.350 

Trachytes — Syenites -SSO-.S/S 

Dacites — Quartz-diorites •27S-.350 

Andesites — Diorites .3SO-.500 

Basalt — Gabbro .540-.675 

To a certain degree these values are characteristic, and being in 
each case a single number which summarizes a whole analysis 
they are more easily employed than are the equally characteristic 
percentages of several oxides, but, after all is said, the contrasts 
are based upon no very fundamental or at least no very definite 
principle, and they give no clew to the mineralogy of the rock. 
The same quotients may be obtained with widely differing aggre- 
gates of minerals and from very dissimilar rocks. 

From the introduction of microscopic methods of investigation 
up to a date about eight of ten years ago, the energies of practi- 
cally all students of the subject were devoted to observing and re- 
cording mineralogical and textural differences and the subject of 
chemical composition received but slight attention. It was re. 
vived, however, toward the close of the eighties by W. C. Brog- 
ger, then in Stockholm, and in the course of time has received 
wide recognition and employment from many others. 

Petrographers are now accustomed to recast an ordinary chem- 
ical analysis by dividing the several percentages by the molecular 
weights of the corresponding molecules, so as to obtain a series 
of numbers, which are called the " molecular proportions " or 
" molecular ratios." These quantities indicate the relative numbers 
of the several molecules in the rock magma, and in that respect 
are more significant than are the percentages. Using the molecular 
proportions as fundamentals, curves or diagrams of various sorts 
can be plotted, which will indicate in a graphic way the variations 
in composition of a series of igneous rocks in a single district, or the 
variations in a single family, the specimens coming from various dis- 
tricts. Many interesting conclusions may be drawn and many char- 
acteristics shown, a review of which will be given in a later paper. 
The molecular compositions of the common rock-making minerals 
are now quite accurately determined and understood, and using 
them it is often possible to calculate from the molecular proportions 


furnished by a rock analysis, the percentages of the several miner- 
als in the rock. The calculations are usually checked in a general 
way by a study of thin sections. 

The commoner rock-making minerals and their molecular com- 
positions are given below in the tables, p. 80-81 , to which reference 
may be made in following the accompanying illustrations, but it 
may be remarked that petrographers are accustomed to regard 
minerals of complex compositions as made up of combinations in 
varying proportions of simple molecules. Thus labradorite is a 
lime-soda feldspar, but it is conceived to be formed by a combi- 
nation in the proper proportions of the albite molecule, Na,0, 
Al,0,,6SiO,, with the anorthite molecule CaO,Al,0,,2SiO,. 
Hypersthene is a silicate of magnesia and ferrous oxide, but we 
think of it as a combination of MgO,SiO,, with FeO,SiO,. 
Olivine is also a silicate of magnesia and ferrous oxide, and is 
regarded as a combination of (MgO),SiO, with (FeO),SiO,. 

In a recent joint report by W. H. Weed and L. V. Pirsson,* the 
latter presents recalculated analyses of a large number of igneous 
rocks. The following example is selected from pp. 466-467. A 
syenite was gathered at the Wright and Edwards mine. Barker, 
Mont, and was analyzed by W. F. Hillebrand of the U. S. Geo- 
logical Survey. A number of minor and relatively unimportant 
determinations were made, in addition to those here quoted, as for 
instance TiO,, H,0, CO,, BaO and SrO, all amounting to 1.50. In 
the citation below, the molecular proportions are given under the 
respective percentage values. 

SiO, A1,0, rc,0, FeO MgO CaO Na,0 K,0 Vfi^ Q Total. 
64.64 16.27 242 1.58 1.27 2.65 4.39 4.98 .37 .05 98.62 
1.077 .158 .015 .022 .031 .047 .070 .053 .002 .0014 

From an examination of thin sections with the microscope it 
was observed that the minerals in the rock were the following. 
The quartz, it may be remarked, was inconspicuous, so that the rock 
is called a syenite, the total silica being at the same time below 
the percentages of a possible feldspar, albite. 

* Geology of the Little Belt Mountains, Montana, by W. H. Weed, with a report 
on the Petrography by L. V. Pirsson. XX. Ann, Rip. Dir, U. S. Gi9l, Surviy, 
III.,.257. The anal3rsii is taken from p. 466. 




Orthoclase, K,0, Al^O,, 6SiOj 

MgO, SiO, 

CaO. SiO, 

FeO, SiO, 
Magnetite, FejOj, FeO 
Quartz, SiO,. 

From an inspection of these formulas it is evident that all the 
K,0 is in the orthoclase ; all the Na^O is in the albite ; all the re- 
maining AI3O, is in the anorthite and requires an equivalent num- 
ber of molecules of CaO. The remaining CaO is in the horn- 
blende and apatite. The apatite can be calculated on the basis of 
the PjOj. All the MgO is in the hornblende. All the Fe,0, is in 
the magnetite and an equivalent number of molecules of FeO are 
required by it. The remainder of the FeO is the hornblende. 
The excess of SiO, then remains for the quartz. The molecular 
proportions are hereafter employed as whole numbers. 























The total CaO is 47; CaO in anorthite 35 ; therefore of the CaO 12 
remain for the apatite and hornblende. The expanded formula 
for apatite is gCaO, CaCl,, 3P,0^, but from this expression we are 
never to infer that the CaCI, exists as such in the mineral. The 
Ca in the CaCI, has been weighed as CaO. Having therefore 
abstracted the necessary CaO for the apatite the residue will go to 
the hornblende as shown in the next tabulation, which also em- 
braces all the remaining minerals. 





















In order to turn these results into percentages of the minerals 
in the rock, we multiply the several molecular proportions by the 
respective molecular weights. 


Thns^lftgnetite, .015 Fe,0, X 160 = 2.42 

.015 FeO X 72 = "oS Total, 3.50 





40 _ 1.24 




60— 1.86 




56= .28 




60— .30 




72= .50 




60 — .42 Total, 4.60 











Grand Total, 


In order to raise these individual percentages so that they will 
make an even hundred, they should each be increased about 1.5 
per cent 
















The above values differ slightly from those obtained by Profes- 
sor Pirsson, because apatite was not reckoned by him, the lime be- 
ing attributed to the hornblende and anorthite. 

In one respect the numerical labor may be shortened. Thus 
the percentage of orthoclase is .0S3K,0x94 + .053A1,0, x 102 + 
6 (.o53SiO,) X 60, an expression which may be factored into 
•053(94+ 102+ 6 X 60). This latter is merely the molecular weight 
of orthoclase multiplied by the molecular proportion of the K,0, 
the oxide which gave us the clue to the original calculation of the 
orthoclase. For this purpose the molecular weights of the sev- 
eral rock-making minerals are later given. 

If the albite molecules were all combined with the anorthite 
ones in order to yield a plagioclase — the relative amounts of each 
in the plagioclase would be proportional to the sums of the mo- 
lecular proportions of the component oxides, as given in the 
tabulation, p. 78 ; 1. ^., albite, 70 + 70 -j- 420 = 560 and anor- 
thite 35 + 35 + 70—1 140. This would be Ab^An.* But some 
of the albite is in the orthoclase. Pirsson found by determination 

* In tbe cnstomarj abbreriations Ab*means albite. An, anorthite, and Or, ortboclase. 


of the optical properties of the plagioclase that it was approxi- 
mately AbjAn. Half the albite molecules were therefore in the 
orthoclase or present in microperthite. From these deductions 
we can calculate the ratio of alkali-feldspar to soda-lime feldspar, 

424 Or + 280 Ab = 704 Alkali feldspar. 

280 Ab + 140 An = 420 Soda-lime feldspar. 

This ratio 704 : 420 is almost exactly 5 : 3. One can readily ap- 
preciate the accuracy with which a result of this character will en- 
able us to classify rocks as orthoclase or alkali feldspar rocks and 
as plagioclase rocks. 

In the actual performance of these recalculations, the minera- 
logical composition of the rock is not always so simple as in the 
case cited. For instance, when biotite is present with orthoclase, 
one cannot say how much potash and alumina belong with each ; 
and if hornblende is also present, the distribution of the magnesia 
and iron oxide presents difficulties. In such cases it may be nec- 
essary to separate and analyze one of the minerals in order to 
furnish the clue, by which the analysis may be unraveled. When, 
at some future day the necessary data shall have been accumu- 
lated, there is little question that rocks of similar textures will be 
classified and defined on the basis of their percentages of the sev- 
eral component minerals. 

The molecular compositions of the more important rock-making 
minerals are here given together with their molecular weights. 
Then a series of tables similar to tables of logarithms is appended 
by means of which molecular proportions can at once be looked 
up and set down for all percentages of the more abundant oxides 
which are likely to occur in the usual run of analyses. It is hoped 
that by their use the recalculation of analyses may be facilitated 
and more often performed. 



Molecular Weight 






K3O, AljO,. esiOj 



Na^O, AljOj, 6SiO, 



CaO, AljO,, 2SiOj 



Neph elite. 

Na^O, AiPs. 2SiO, 



K,0, AljOj, 4SiO, 



i2CaO, aAljO,, 9Si02 



Analdte, Na,0, A1,0„ 4SiO, + <H,0 440 

Sodalite, 3[Na,0, A],0„ aSiO,] + ^NaQ 969 

Haaynite, 3[Na,0, AljO,, aSiO, | + aCaSO^ 1 124 

NoseUte, 3[Na,0, A1,0„ aSiO, | + aNajSO^ 1 136 

MnscoYitc, ( K, H ),0, Al,0,, aSiO, 

Biotite, 2[(H,K),0, (Al, Fe),0, .aSiO,] 

2 (Mg. Fc)0, SiO,. 

Amphibolet and Pj- MgO, SiO, 100 

roxenes contain in FeO, SiO, 132 

the different vari- CaO, SiO, 116 

eties different pro- (M^Fe)0,(Al, Fe)jO,, SIO, 

portions of the Na,0, Fe^O,, 4SiO, 461 

Inoleculet here 

OUrinc \ (MgO),.SiO, 140 

^"^°*^' \ (FeO)j.SiO, 204 

Magnetite, FeO, FcjOj 232 

Anatite \ ^CaO, 3P,0,. CaCI, 1041 

^P*^^' 1 9CaO. 3P,05. CaF, 1008 

In the group consisting of haiiynite and noselite (often called re- 
spectively haiiyne and nosean) neither mineral occurs pure, of the 
formula given, because the two molecules always replace each 
other — the combination rich in lime being called haiiynite, that 
rich in soda, noselite. In cases where as in muscovite, biotite and 
one of the pyroxenes, two elements, such as K and H, Al and Fe, 
or Mg and Fe replace each other in indefinite amounts, no molec- 
ular weight can be calculated. We must then assume separate 
and relatively simple molecules; for instance in muscovite K,0, 
A1,0,, 2SiO, and Hp, A1,0,, 2SiO,. It does not follow however 
that these are known in nature. 

In the following tables the oxides are arranged in the order sug- 
gested by H. S. Washington in the American Joutnal of Science^ 
July, 1900, p. 59. It is the most convenient and significant one 
for petrographers and it emphasizes the most important features, 
even if it separates oxides of like chemical properties, such as 
SiO, and TiO,, Al,Oj, Fe,0, and Cr,Oj, etc. In using the table the 
units of percentage are in the left line, the decimals then follow 
horizontally to the right as in logarithms, but by a proper use of 
the decimal point the samp values will answer for tenths or multi- 
ples by ten, of these percentages. 

TOL, xxn.— ^ 



Silica, SiO,. Molec. Weight 60. Log. 1.718151. 






1 4 



































! .543 






1 .551 
















: .576 






, .585 





; .593 











1 .610 

















i -635 
























































































































































































































































I. on 







1. 021 


I 025 











1. 041 







1.05 1 






1. 061 







1. 071 













1. 09 1 






1. 100 


1. 103 

1. 105 

I. 106 

1. 108 

1. 1 10 

I. Ill 




1. 116 

1. 118 

1. 120 

1. 121 


1. 125 

1. 126 

1. 128 

1. 130 

1. 131 


1. 133 


1. 136 

1. 138 

I. 140 

1. 141 

1. 143 

1. 145 

1. 146 

1. 148 


1. 150 

1. 151 


1. 155 

I. 156 

1. 158 

1. 160 

1. 161 

1. 163 

1. 165 


1. 166 

1. 168 

1. 170 

1. 171 

I. 173 


1. 176 

1. 178 

1. 180 

1. 181 


1. 183 

1. 185 

1. 186 

1. 188 

I. 190 

1. 191 

1. 193 

1. 195 

1. 196 

1. 198 



1. 201 





1. 210 

1. 211 


1. 215 


1. 216 



1. 221 













1. 241 







1. 251 




1.258 ■ 


1. 261 







1. 271 













1. 291 







1. 301 





1. 310 





1. 316 



1. 321 









iliBUia, AI,0» Molec. Welgkt Its. Log. «.«waM. 





5 ' 


7 i 






•003 ; 




" 1 







.012 ' 




.015 , 






.021 ' 






.026 1 

















.041 i 



.044 , 


.046 : 


















.062 , 








.069 I 






■074 ! 








.081 , 



.084 1 




.088 1 







•094 1 









.lOI , 












.III 1 


.112 1 






.117 1 






.123 1 








.130 1 



•»33 1 





•137 1 












.148 1 
















.163 1 









































.199 ' 











.209 1 



.212 ' 




















































































1 .289 






1c Oxk 

le, Jtti 


W« WC 

\%\L% 10 

0. Los, 

























































. .025 




! .028 





























i .044 











1 .050 











1 .057 











• .063 











; .069 











; .075 


1 .077 


































1 .097 

' -097 






Ferrous Oxide, FeO, Molec. Weight 79. log. 1.857S3S.. 



















































.051 . 

053 . 










.065 1 











•079 ; 
























































.149 . 

150 ' 











,164 i 










.176 1 . 











.190 . 























.214 ! .215 

. Weight 40. 

4 1 5 

.010 .0x2 

.217 1 .218 
Log. 1.602060. 

.015 1 .017 

.220 ! 


.020 j 


















.042 1 











.067 , 

.070 . 










.092 1 























.145 ; 






• 157 




























































.292 ' 






















.342 1 

345 ! 










•367 . 

370 1 










•392 . 

395 1 









, .415 

.417 1 ■ 











.442 . 











.467 ; . 






















•517 ; ■ 

520 . 










.542 . 

545 . 











570 . 










.592 . 

595 ■ 


























Use, CaO, Helec Weiskt fit. Utg. 1.748188. 









6 < 













' .025 













.064 1 . 















, .071 





i .080 





















1 ""^ 







' .125 




' .132 









' .146 













, .168 











1 .185 

' .187 






, .196 


' .200 


, .203 


.207 ; 









\ .221 

1 .223 







' .234 




' .241 

•243 . 









' .257 


: .259 





Soda, ira,0» Molec. Welghl fS. Log. 1.7MHKL 



























































































































































































PolMh, KtOy Molec. Weight 94. log. 1.97Sld8. 







.OOO y 
.OIO j 
.02I ! 

' .064 

\ .074 

1 .085 

I .096 












, .159 

001 .002 
012 , .013 
022 I .023 

033 I .034 













.003 I .004 









' .024 



































































































1 .167 



Water, H^Oy Molec. Weight 18. log. 1.255278. 




























.011 { 
.066 ' 

.122 ' 



•5" I 







.300 I 

.355 ' 














478 .483 
533 .539 

i 8 


; .044 

' .155 

! .211 

I .267 


; .378 


Carbonic Acid, C0„ Molec. Weight 44. log. 1.54S453. 








1 8 



























1 .052 











1 .075 





I .086 






1 .098 






















1 .143 


.148 ! 




















1 .189 




















TtUalc ield, T10„ H*Iee. Weight 8). Log. 1.91B814. 




' 3 







' .000 



•003 , 

































.040 , 











.052 1 











' .064 





















.089 , 





























ZIrcontt, ZrO„ Moler . Weight 1^. Log. 9.080800. 


■ .OOO .OOI .OOI .002 .003 .004 .005 .006 , ^006 ' .007 

PMosphorlc pentoxlde, PA* Molec. Weight 142. Log. 2.152288. 




.000 .001 
.008 I .008 
.015 .015 

.002 ' .003 
.009 , .010 
.016 I .017 












Snlphnrlc anhydride, SO,, Molee. Weight 80. Log. 1.908000. 

01 a 3 4 .s6 78 

o i .000 I .001 .002 .004 .005 .C06 I .007 .009 ' .010 I .011 

I I .012 I .014 .015 .016 .017 I .019 .020 I .021 ' .022 ; .024 
Chlorine, CI, Atomic Weight 8S.5. Log. 1.5S0228. 

oil a 3 4 5 6 7I8 9 

' I 

.000 1 .002 .006 .008 .011 .014 .017 .020 .022 I .025 

Fluorine, F, Atomic Weight 19. Log. 1.278754. 

oi|a|34 5 6 7|8|9 

.000 ' .005 .010 I .016 , .021 .026 .031 I .037 I .042 I .047 

Snlphor, 8, Atomic Weight 32. Log. 1.505150. 




i 031 

! .062 

















Gtaromie oxide, Gr,Ow Molec. Welgbt 1S9.8. Log. 3.184198. 


2 1 



5 . 

o ija 3-4 56 7 » 

.000 .000 .001 .002 .002 .003 , .004 .004 .005 
.006 .007 .008 .008 .009 .010 .010 .011 .012 
.013 .014 .014 \ .015 .016 .016 .017 .018 .018 
.020 .020 j .021 .022 .022 .023 .024 i .024 .025 
.026 1 .027 \ .028 .028 .029 .030 .030 .031 .032 
•033 034 .034 .035 .036 .036 .037 .038 .038 

Nickel oxide, XiO, Molec. Weight 75. log. 1.875061. 
Cobalt oxide, CoO, Molec Weight 75. Log. 1.875061. 




















6 7 

.008 1 .010 
.021 .023 
.035 .036 







Cnpric oxid 

le, CqO, Molec. Weight 79.1. Log. 1.898176. 


I, a 345617 

1 1 

.001 ! .002 .004 . .005 .006 .007 1 .009 





Manganons oxide, MnO, Molec. Weight 71. Log. 1.851258. 


I 2 3 4 

5 6 














.003 .004 

.017 .018 

.031 .03a 

.045 046 
.059 .060 

•073 074 



.007 .008 
.021 1 .022 

•035 036 
.049 ' .050 

.063 1 .065 

.077 j .079 






BaO, Molec. Weight 








3 ; 4 

.002 .002 
.008 .009 











.003 .004 
.010 ' .010 





, 8rO, Molec. Weight 108.5. Log. 2 










5 1 6 1 7 









.005 .006 .007 
.014 < .015 .016 



. Ll,0, ] 

Weight 80. U 

»g. 1.41 



X a 

.003 .006 
.036 .040 


4 5 









.013 .016 .020 
.046 .050 ' .053 





Through the untimely death of Dr. Herman Andreas Loos, who 
died on July 17, 1900, at the age of twenty-four years, chemistry 
has lost an able student and a promising investigator. He was 
the son of Mr. August Loos, of New York, and received his earlier 
education at the College of the City of New York, where he was 
graduated with the degree of Bachelor of Sciences in 1895. Dr. 
Loos then taught for several years in the day and evening classes 
of the New York Public School system, but finally, in 1897, re- 
signed his day-school position to become a graduate student in 
chemistry at Columbia University. While thus a student under 
the Faculty of Pure Sciences he at the same time ably filled the 
Instructorship in Chemistry in the East Side Evening High School* 
He was a pioneer in this branch of our public instruction, and at 
once by his success in organizing his classes and ability to interest 
his pupils gained the respect and commendation of the Board of 
Superintendents of Public Instruction. His name is well known 
in the Board of Education circles, and those in authority know 
that they have lost a worker who made a splendid record for him- 
self as a teacher. His scholarly industry and ability for research 
were soon recognized at Columbia, where he was honored with the 
University Fellowship in Chemistry for 1899- 1900. His principal 
contributions to the literature of chemistry are : " The Electro- 
lytic Determination of Zinc in Amalgam" (thesis for M.A.); "A 
Study on the Metallic Carbonyls and their Decomposition " 
(School of Mines Quarterly, 21, 182); " The Decomposition of 
Nickel Carbonyl in Solution " {Journal American Chemical Society, 
22, 144) ; and " A Study on Colophony Resin " (thesis for Ph.D.). 
His doctorate thesis was his last and is perhaps his most important 
work. An officer of the Chemical Department at Columbia speaks 
of it in the following words : " In the study on Colophony Resin 
he has decided two controverted points, viz: that abietic acid will 
form an anhydride on heating, and that it is not an oxidation prod- 
uct of turpentine. He has also developed a new method for the 
preparation of pure abietic acid, and established its formula by a 
number of analyses. Many salts were prepared and their decom- 


position both by water and sunlight noted. The whole work is of 
great theoretical and practical interest." 

Dr. Loos was a strong student of theoretical science but at the 
same time showed those sound business ilit es and that respect 
for practica things which mark the highest and most successful 
type of the professional man. His work in applied and industrial 
chemistry was very considerable and he had a surprising knowl- 
edge of the chemistry and methods of manufacture of resin, oils, 
varnishes and allied products, partly no doubt because his father 
and other members of his family were interested in this class of 

Immediately after receiving his doctorate degree, Dr. Loos, in 
June, was appointed Assistant in Analytical Chemistry at Colum- 
bia University. He resigned this position, unfortunately however 
as it now appears, to accept a flattering offer from the Copper 
Corporation of Chili, Limited. To act as assayer and chemist to 
this corporation he left New York in June last and it was while on 
his way to Chanaral, Chili, that he was stricken with yellow fever. 
There is now no doubt that he contracted the disease in Panama 
early in July when the fever was raging there and it was shortly 
after leaving this port that he died and was buried at sea. 

He was a Fellow of the Society of Chemical Industry of Lon- 
don, a Member of the American Chemical Society, and Secretary 
and one o he . ounders of the present Chemical Society of Co- 
lumbia University. Dr. Loos was also highly interested in the 
study of entomology and made during his li e some excellent and 
valuable collections of insects. He was a member of the New 
York Entomological Society. 

The death of so young and promising a student is indeed a sad 
one. The loss of so useful and intelligent a man brings always a 
shock to his circle of friends but that shock becomes doubly se- 
vere when we reflect that a young man has been stricken at the 
very beginning of his ife of usefulness, at the very moment he 
became a value to the State. Mr. Loos was a man of singularly 
lovable character. It is no exaggeration to say that by his agree- 
able personality he made friends wherever he went and that he 
had no enemies. In his death Columbia has lost an honorable 
Alumnus and the community at large a chemist of rare ability, a 
student of sound learning and a man of unusual energy, perse- 
verance and practical judgment. 

Charles Derleth, Jr. 

New York, Sept. 15, 1900. 



By ELWYN waller. 

Potassium Xanthaie. Campbell {J. Am, Chem. Soc.^ XXII. , 307). The 
reagent as provided by the dealers is usually unsatisfactory in quality, and 
at Ann Arbor the students are required to prepare their own salt for use 
thus: A weighed amount of fused KOH is digested cold in a tightly 
stoppered flask with absolute alcohol (2.5 cc. per gm. of KOH). When 
solution is complete (except a small amount of KjCO,) the clear solu- 
tion s decanted into a beaker which is stood in ice water. Pure CS, is 
then gradually added with a stirring to the extent of i cc. per gm. of KOH. 
The stirring is maintained after all the CS, has been added, the tempera- 
ture being kept at 10** C. or below. The salt when fully formed is fil- 
teted by use of the pump, then pressed down firmly, and washed once 
with absolute alcohol, then twice with ether. The Mlt is then dried at 
100** C. and pulverized for use. 

Cobalticyanides, Miller and Matthews (y. Am, Chem, Soc, XXII., 62). 
The potassium salt is possibly o^ use as a reagent in quantitative analysis 
The Pb salt is very soluble, while the combinations with Ag, Cu and Bi 
arc insoluble in waier or HNO3. Of these three, the Bi compound 
only is dissolved out by HCl. 

In presence of sufficient amounts of (NH 4)3804, ferric iron is not 
precipitated, whereas Zn, Mn, Ni and Co are. From a mixture of these 
the Zn may be separated by digestion with KOH solution. 

Standardizing Acidimetric Solution, Threte and Richter (Z/j. Angew, 
Chan,,, 1^00, 486) note that on standardizing with Iceland spar the 
results are a little higher than when NajCO, is used. De la Source 
{Ann, Chim, Anal. AppL^ V., 121) observes that phenol phthalein is more 
sensitive in the cold, than when the solution is hot. Moreover, hot so- 
lutions dissolve alkali from the surface of some kinds of glass. 

New Indicator — Luteol. Gloess {Morit Sci. , XIV. , Mar. , 1 900) . Luteol 
is oxy-chlor-diphenyl-chin-oxaline, a derivative of phenacetine. It is in- 
soluble in water and difficultly so in cold alcohol and ether, though soluble 
in the two last when hot. In alkalies it dissolves with strong yellow 
color. With the slightest excess of acid the solution becomes colorless, 
and cloudy from separation of the luteol (white flocks). It is very sharp, 
more sensitive than phenolphthalein or litmus, can be used with ammo- 
nia, in hot solutions, or in titrations of carbonates. 

JVirw Indicator. Formanek {Zls. Anal. Chem., XXXIX., 99). The 
substance is Alizarine green B from Dahl & Co. It is soluble in water 
with a dirty green color — less soluble in alcohol. It gives green with 
alkalies, carmine red with acids. The aqueous solutions when much 
diluted take a flesh red color, apparently resulting from hydrolysis.. It is 
extremely delicate. Sensitive to COj. Uncertain in presence of NH4 
or Al salts. Paper saturated with solution of dyestuff" is excellent for 
** spot" testing. 


Alkaline Carbonates, Cameron {Am, Chem,/,, XXIIL, 471). Dilu- 
tion of alkaline carbonates causes hydrolyzing, e, g., NajCOg 4- H^O 
rsNaHCOg + NaOH. To estimate NajCOg in mixture with bicar- 
bonate, the use of standard KHSO4 was found to give satisfactory 
results, using phenolphthalein as indicator. 

Potassium Estimation. Adie and Wood {J", Lond, Chem. Soc, 
LXXVIL, 1076). A new method depending on the insolubility of the 
cobalti-nitrite K2NaCo(N02)8, HjO — in acetic solution. The 
reagent used is the sodium cobalti nitrite made by mixing a solution of 
113 gms. Co(C2H303)2 in 300 cc. water and 100 cc. HCgHgOj with 
320 gms. NaN02 in 400 cc. water. After settling and filtering the solu- 
tion is made up to one liter. The solution to be tested should contain 
0.5 to I per cent. K2O, a smaller proportion gives difficulty in collecting 
the precipitate. Add 10 cc. of the reagent in i cc. strong HCjHjOj. 
Allow to stand over night. Filter and wash with 10 per cent. HCjHjOj 
and finally once with water. Dry at 125° and weigh. The precipitate 
is soluble in the prescribed solution to the extent of 1:20000. It con- 
tains 17.2 per cent. K. The determination may be made volumetric by 
filtering through asbestos, boiling up with NaOH, and, afier filtering out 
sbestos and C02O3, etc., making up to 100 cc. and titrating aliquot 
pornons with K2Mn20g after acidifying. 

Qualitative Tests for Alkaline Earths, E. Dumestril {Ann, Chim. 
Phys,f XX., 125). The use of the chromates is proposed instead of the 
sulphates as ordinarily laid down in text-books. The solution of the 
earths is obtained as neutral chlorides by precipitation with carbonate 
dissolving in HCl and evaporating to dryness. Addition of saturated 
solution of SrCrO^ solution will show a precipitate if Ba is present ; in 
another portion a saturated solution of CaCrO^ will give a precipitate if 
Sr (or Ba) is present. If Ba is present, it, together with possible SrCr04, 
is precipitated out completely by CaCrO^ and the precipitate boiled with 
NH4CI, which extracts the SrCrO^. From this the Sr is precipitated as 
carbonate, dissolved as before and detected in neutral solution by CaCr04. 

Calcium oxalate — Conversion to Sulphate. Hess {J, Am. Chem. Soc, 
XXII., 477). After obtaining CaC204 in the usual manner, the precipi- 
tate is ignited directly to destroy the filter paper. An amount of pure 
dry NH4NO3 equal in bulk to the mass of lime is then added together 
with about twice as much (NH4)2S04, and the crucible provided with 
a close fitting cover is heated cautiously up to complete ignition. The 
heat should first be applied so far as possible at the top of the crucible, 
by tilting, etc., or the violence of the action may cause losses. 

Free Lime in Portland Cement. Michaelis (Than, indust. Ztg,^ 
XXIV., 860). The use of the reagents proposea for this purpose is 
criticised adversely. Such are AljCl^, alcoholic I solution, and H^S. 
The proposers of these reagents do not appear to have studied their effect 
on the constituents of a Portland cement. 

Lime in Presence of Iron and Alumina. Blum {Zts, Anal. Chem,^ 
XXXIX., 152). To determine CaO in slags and iron ores in an iron 
master's laboratory when results within 0.2 to 0.5 of the truth are suffi- 
cient for the guidance of the work, a rapid method consists in adding 
tartaric acid to the filtrate from' the silica in amount sufficient to hold up 
FejOj and AI2O3, rendering ammoniacal and precipitating with(NH4)2- 


C2O4. Investigation showed that the error due to solubility of the 
CaC,04 was very nearly counterbalanced by the small amounts of Fe^O,, 
etc. , carried down with the precipitate. When Mn is present the results 
are somewhat high; some MnC204 accompanying the CaCjO^. In one 
case, in a slag containing 1.91 MnO, the result was 0.9a per cent. high. 

Determincuion of Alumina. Allen and Gottschalk {^Am, Chem, Jour,^ 
XXIV., 292). Precipitation by COj from the solution in KOH, was 
found to be the most expeditious and accurate method. The method 
prescribed consists in nearly neutralizing the acid solution with ammonia, 
and then running into the solution from a burette, a solution of KOH 
(in which the amount of impurity is known) in quantity just sufficient to 
redissolvc the precipitate first formed. Then pass in a current of CO^, 
twenty minutes will suffice for 0.2 jjm. AljO,. Filter and wash several 
times. Then rinse back into the beaker and boil two or three minutes 
with 150 to 200 cc. of water containing a little NH^Cl or NHjNOj. 
Let settle and decant onto a fresh filter. Repeat this once or twice, 
finally ignite, cool in a desiccator and weigh. Deduct correction for the 
impurities in the KOH. 

Alumina and Ferric Oxide in Natural Phosphates. Veitch (y. Am, 
Chem. Soc , XXII., 246). The result of a series of experiments indi- 
cates that by the NajSjO, method AIFO4 can be accurately separated 
from an HCl solution containing besides Al, some Fe, Mn, Ca. Mg, Na 
and K, if sulphates are not present in large quantity and fluorides are 
absent. Silica also should be absent. 

Method. Treat i gm. of the phosphate in Pt dish with 5 to 10 cc. 
HF, let stand two to three hours, heat to dryness, then add 2 cc. 
H2SO4 and heat for some time, cool, add 10 to 20 cc. cone. HCl and 
warm for a few minutes. Boil in a beaker for fifteen to twenty minutes, 
filter and wash, add 50 cc. of a 25 per cent, solutirn of NH4CI and 
NH4OH until just alkaline, then HCl until the precipitate just redis- 
solves, cool, dilute to about 250 cc. and add 50 per cent. NajSjOg so- 
lution drop by drop until the solution is colorless (about 20 cc.) cover, 
boil about thirty minutes, 61ter, wash back into the same beaker, dissolve 
in HCl and reprecipitate as before adding 2 cc. of a 10 per cent, solu- 
tion of (NH4)2HP04, wash twenty times with 5 per cent, solution of 
NH^NOg. ignite, etc , >nd weigh. The use of NajSjOj in the second 
precipitation is not absolutely necessary. 

Fe may be determined in another portion by solution, reduction and 
titration with K^MujOg. 

Iron in PuddU Slag. Blum {Zts. Anal. Chem., XXXIX., 156). Vd 
is often present in iron ores, and is concentrated in the slags, with the 
result that the estimation of Fe in those slags by reduction with Zn or 
SnClj and titration with KjMujOg or KgCrgO^ gives too high results. 
The method recommended is to precipitate twice as basic acetate, dissolve 
the second precipitate in HCl, add tartaric then ammonia and (NH4)3S. 
The Vd then remains in solution. The FeS precipitated (after thorough 
washing with dilute (NH4)2S) may be dissolved, oxidized and reprecipi- 
tated and weighed as FcjOg. 

Chromium in Steel. Mahon (y. Am. Chem. Soc.y XXL, 1057). 3 
gms. of the sample are dissolved in 50 cc. cone. HCl and evaporated to 
a moist cake. Add 50 cc. cone. HNO3 and boil out nitrous fumes. 


Then add 4 gms. KCIO3 and boil down to 30 cc. Dilute to 300 cc. 
add 15 cc. ammonia mix, and filter when cold. Dilute filtrate and wash- 
ings to 450 cc. and titrate with standard SQlutions of Fe2(NH4)2(S04)j 
and KjMnjOg, CrOg will be reduced by the filter paper unless the solu- 
tions are dilute and cold. Too large an excess of HNO3 interferes with 
the titration. 

Manganese in Manufactured Irons, Mignot (Ann. Chim» Anal, App.y 
v., 172). In HNOg solution Mn reacts with ^\%0^ to form H^Mn^Og 
which may be titrated with standard HjO^. To prepare the ^\^0^ fuse 
together in an iron crucible 2 parts NaOH, and one part each of BiONO, 
and KClOg, stirring well, cool and wash with water until no longer al- 
kaline, and pulverize. Dissolve the material to be analyzed in 25 parts 
HNO3 (gr. 1.2) heat to boiling, dilute and add 3 parts of the Bi204. 
Filter the rose-colored solution through asbestos, and titrate with standard 
HjOj. The method is inapplicable if over i per cent. Cr is present. 

Estimation of Manganese in Steels, Jervis (^Chem, News^ LXXXL, 
171). The process in which the Mn is oxidized to permanganate by 
Pb304 or PbOj is not in all cases accurate. Ordinarily the presence of 
the iron seems to favor the production of permanganate, but when less 
than o.oi gm. Mn is present the lesult is low. On increasing the propor- 
tion of Mn, a point is reached where the more Mn there is added the 
lower the result obtained. A case is cited where a 13 per cent. Spiegel 
was reported as containing 0.4 Mn and a 20 per cent. Spiegel as well as 
an 80 per cent, ferro manganese gave no indication at all by this process. 

Electrolytic Separation of Manganese, Hiorns ( Chem, News, LXXXL , 
15). Separation from HNO3 solution which is kept at 60^ C. is the best. 
Not over 0.03 gm. Mn should be present. For an ore containing about 
55 per cent. Mn 0.05 gm. of the sample is all that should be used. The 
current should act for 1 2 hours. 

Determination of Zinc. Meade {/, Am, Chem, Soc,, XXII., 353). 
The method was tried of precipitating Zn as ammonia arsenate reducing 
the AsjOr of the precipitate by means of HCl and KI, and then titrating 
the I set free by standard NajSjOj. The results were very satisfactory. 
In ores the Mn was separated by precipitating with KCIO3 in HNO3 so- 
lution separating Fe^O. and AI2O3 by double ammonia precipitation, 
when an excess of Na2HAs04 was added to precipitate Ca and Mg, the 
ZnNH^AsO^ being dissolved in the ammonia, but precipitated after fil- 
tering by acidulation with HC2H3O2. By heating the precipitate at first 
flocculent was brought to a dense granular condition. After filtering and 
washing the precipitate was dissolved back into the beaker in which the 
precipitation was made by use of 50 to 60 cc. dilute HCl, and 2 to 3 
gms. KI added. After standing a few minutes the I set free was titrated 
with standard Na2S203. 

Estimation of Zinc, Low (J, Am, Chem, Soc, XXII., ig8). A re- 
statement of the method originally published in 1892 (Vid. Quarterly, 
XIV., 40), with such modifications as have be^n found advantageous. 

Standard K^FeCy^ solution 22 gms. of the crystallized salt in one 
liter. Standardize by dissolving o.ioo gm. pure Zn in 6 cc. HCl, then 
adding 10 gms. NH4CI in 200 cc. of boiling water. Titrate hot, using 
UO(N03)2 as indicator (spot test). The value of i cc. should be about 
0.005 ^^' When near the end reaction, a little time must be allowed 


for the complete reaction between the salts or the titration will be over- 
run. It is best therefore to reserve a portion of the solution to be added 
to the main solution after the first titration has been carried to an appar- 
ent immediate termination which after a short time will be found to be a 
little overrun. A correction should also be applied for the amount of 
standard solution required to give a positive brown to the indicator. 

Anafysis, — ^To 0.5 gm. ore m a 250 cc. flask, add about 2 gms. KNO, 
and 5 cc. cone. HNO,. Heat imtil acid is about half gone and then add 
10 cc. of cold saturated solution of KCIO3 in cone. HNO,. Boil down 
to complete dryness. This is most readily efifected by using a holder and 
manipulating the flask over a naked flame. Cool and add 30 cc. of a 
solution made by dissolving 200 gms. NH^Cl in 500 cc. strong ammonia 
and 350 cc. water. Boil gently for 2 minutes ; then filter. Wash with 
solution containing about 100 gms. NH^Cl and 50 cc. NH4OH solution 
per liter. Neutralize closely with HCl, and then add 6 cc. HCl in ex- 
cess. Dilute to 150 cc. and add 50 cc. saturated H,S water. Heat to 
boiling, and without filtering, proceed to titrate as already described for 
standardizing. Cu and Cd are the chief interfering metals, but their 
sulphides or the surplus H^S cause no interference. 

ZinC'EUctrofytic. Paweck(Z/x./. ^/f^/rrA^w., V., 221). The cathode 
is made of one (or two) discs of copper about 6 cm. in diameter, the 
wire attached to the center. They should be cleansed by rubbing with 
wet chalk, and then after immersing for a moment in H2SO4 amalga- 
mated by electrolytic deposition of Hg from dilute HgCL solution 
acidified with HNO,, using a weak current, 0.2 to 0.3 amp. Wash with 
dilute HCl ; following that up successively with water, alcohol and ether. 
An alkaline solution may be used if it contains 0.16 to 0.52 gm. Zn as 
sulphate; 7 to 12 gms. Rochelle salt and 5 to 8 gms. KOH or NaOH. 

An acid electrolyte should contain 0.23 to 0.31 gm. Zn, as ZnS04 14 
gms. K3SO4 or Na2S04, and 3 to 5 drops cone. H2SO4. A current of 
3.6 volts should be used. When finished the cathode is washed succes- 
sively with water, alcohol and ether — dried and weighed. 

Nickel in Nickel Steel, Sargent (J, Am, Chem, Soc), Dissolve 2 
gms. in HCl, adding 1 cc. cone. HNO2 to oxidize the Fe, and evapo- 
rate to hard dryness. Take up with 20 cc. HCl, evaporate to 10 cc. and 
then transfer to a 250 cc. separating funnel. Use as little as possible of 
warm HCl (gr. i.i) in rinsing the solution into the funnel. Cool, and 
then add 40 cc. of ether which has been previously agitated with 5 cc. 
cone. HCl. Shake up vigorously, cooling under the tap from time to 
time, then run off the aqueous solution. Wash the ether by shaking up 
with two successive lots of 5 to 10 cc HCl each. Boil out the ether 
from the aqueous solution and precipitate out the small remaining 
amounts of Fe by ammonia, adding a little Br water to bring out Mn as 
well. Resolution and reprecipitation will afford a complete separation of 
Ni from Fe and Mn. If the original metal contained Cu, that element 
must be separated at this point by H^S. If (as is usual) Cu is absent, 
evaporate off the excess of ammonia, finally obtaining a solution of about 
100 cc. which contains but slight excess of NH4OH (i cc.). Then add 
5 cc. of AgNOj solution (0.5 gm. per liter) and 5 cc. of a 2 per cent, 
solution of KI. Titrate with standard KCy (strength should be such that 
I cc. = about 0.00 1 gm. Ni) until the opalescence is entirely removed. 


(Best work with a dark background.) Make a blank test on the AgNOg 
and KI solutions used, deducting the amount from that used to titrate 
the Ni. The method is rapid and accurate. The distinctive feature 
consists in shaking but once with ether, and using the ammonia precipita- 
tion to complete the separation. 

Separation of Nickel and Iron, Ibbotson and Brearley {Chem. 
News, LXXXL, 193). The adherence of Ni to the precipitateid Fe,- 
(OH)j is not due to the formation of an FeNi compound^ but to the 
absorption of the NiNH^ compound just as sand, filter paper, etc., will 
absorb certain salts from their solutions. The presence of considerable 
amounts of NH^ salts (chloride, sulphate, etc.), as well as of NH^OH 
gives FegCOH)^ free from Ni. The addition of KCN in amount but 
slightly in excess of that necessary to form the double cyanide, 2KCN, Ni- 
(CN)^ readily effects a clean separation. In steel works where the 
amount of Ni present is known approximately a very rapid separation 
and determination can be effected, thus — add NH4OH to the acid solu- 
tion of the sample until at least all free acid is neutralized, add excess of 
standard KCN, pour into an excess of NH4OH, and titrate a portion of 
the filtrate with AgNOj and KI. The results will be untrustworthy if 
too large an excess of KCN is used. 

The separation by (NH4)2C08 described by Fresenius — (vide Cairns, 
Quant. Anal., 3d ed., p. 195) can be made uniformly successful if the 
solution is kept slightly acid throughout. 

Smaii quantities of Cobalt in presence of Nickel. Moore ( Ch, News^ 
LXXXIL, 73). The precipitate obtained by NaClO has been variously 
stated as CogOg, Coj^Oig and CojOg. Adding Br or CI and alkali 
to a mixture of the two metals in solution precipitates both as the higher 
oxides, if, however an emulsion of ZnO is substituted for the alkali, Co 
is precipitated immediately, Ni after some lime in the cold or sooner by 
boiling. The more Zn in the solution the slower the action. With a 
mixture of higher oxides of Ni and Co HjOj reduces the Ni to Ni (OH)^ 
leaving Co as CojOg. 

The solution of the two metals should not contain much free acid. If 
Fe is present it should be separated by adding a little ZnO. Then dilute 
300 or 400 cc, add one drop HCl, bring to a boil, add Br water and 
(not too much) ZnO. Boil about five minutes, filter and wash with boil- 
ing water. Rinse the precipitate into a small beaker, dissolving off the 
last traces from the filter paper by HCl with a very little SO 3 into a cap- 
sule, when the solution is evaporated to dryness, then dissolved in water 
and added to the main precipitate. After boiling up to disintegrate the 
precipitate, and then cooling, add 5 to 10 cc. of HjOj rendered strongly 
alkaline by NaOH, stir, boil to remove HjOj, cool and digest a few 
minutes in a stoppered flask containing KI and HCl. Then titrate the I 
set free by standard NajS^Og. I x .046511 = Co. 

Cobalt Estimation, Moore (CA. News^ LXXXIL, 65). 2.5 gms. of 
the finely pulverized ore was treated with cone. HCl, evaporated on it 
until it became syrupy. This was diluted and SiO, and Chromite fil- 
tered off. 100 cc. of saturated solution of NH4CI, was added and the 
volume brought up to 400 cc. Weak NH4OH was then added with stir- 
ring until the color of the solution was a deep red from presence of basic 
ferric salts. The neutralization was completed by cautious addition of 5 


per cent, solution of Na^CO,. When the precipitation is just complete, 
the solution being still acid, the cloudy liquid is diluted to exactly 500 
cc.y and, after thorough shaking, filtered through a ribbed filter. 400 
cc. of the clear filtrate were taken, 20 cc. of a saturated solution of 
NaCjHjOj and 10 cc. HCljHjOj added, the temperature raised to boil- 
ing, and HjS passed in until it was nearly cold. The mixture of CoS, 
NiS and ZnS thus obtained was filtered off, ignited, and then dissolved 
in HCl with addition of a few drops of HNO3. After evaporating twice 
with HCl to remove HNO,, and removing any ¥t^{Oli)^ which re- 
mained by use of emulsion of ZnO, HjO, and KOH w^'e added, which 
separated CojOj, (as well as Ni(OH)2, etc.). After boiling out excess 
of HjOj the solution was cooled, placed in a stoppered fiask, and KI 
with HCl added. The free I was then titrated by NajSijOj. IX 
0.4654 = Co. 

Molybdenum in Iron and Steel, etc, Ibbotson and Brearley {Ch 
News, LXXXL, 269). NaOH is the best reagent for perfect separation 
of Fe (in ferric form) from Mo. At a certain stage of the precipitation 
an insoluble ferric molybdate forms, which requires more NaOH to de- 
compose it. A psitive excess of the reagent is therefore necessary. 
The method consists in dissolving a sufficient amount of the sample (2 
gms. rr so) in HCl, oxidizing with HNO3, neutralizing free acid with 
NajCO, just short of the formation of a precipitate, and then adding 
NaOH solution to the extent of 30 or 40 cc. of double normal in ex- 
cess. (54 cc. normal NaOH will precipitate i gm Fe in the ferric con- 
dition.) Heat to boiling, agitate violently and filter. The filtrate is 
acidified with HCl, and Mo precipitated as PbMoO^. 

The effect of Mo in steel has been studied with reference to its influ- 
ence on other determinations. No effect was observed on Si or S deter- 
minations. Mn separated by Br and acetate was not affected. In the 
Ford Williams method it causes the results to run a little low. 

Fe if reduced by Zn or by SnCJj is made to appear high ; if reduced 
by SO 2 the results are correct. 

P by precipitation is unaffected. By the rapid method the precipitate 
appears to be contaminated by MoO, and high results are afforded. 

Molybdenum powders for adding to furnace charges. Almost all of 
these are claimed to contain 95 per cent. Mo, but none have been en- 
countered containing over 90 per cent. Tungsten is a frequent con- 

In these C is determined by a direct combustion in a current of O. 
To determine Mo, mix the powder with Na^CO- and a little KNO, in a 
platinum crucible, cover with a layer of NajCOg. Fuse, keeping only 
the bottom of the crucible red hot, extract with water, filter off insoluble 
FcjO,, etc., precipitate Mo and W with Pb (^2^8^2)2- 

Molybdenum in Alloys, Borntrager (Z/x. AnaL Chem,, XXXIX., 91). 
Dissolve I gm. of the alloy in 50 cc. aqua regia evaporate with HCl to 
expel HNOg and treat the dry residue with 50 per cent, alcohol. Most 
of the MoOg is left undissolved. A second evaporation of the filtrate 
and treatment of the residue with alcohol as before, will separate the re- 
mainder of the MoOg. 

Separation of Tungsten from Molybdenum, Ibbotson and Brearley 
{Chem, News, LXXXI., 13). Separation by volatilizing off MoOg in a 

VOL. XXII.— 7. 


muffle is unsatisfactory. About one- tenth of the M0O3 present is re- 
tained by the WO3. Separation depending on insolubility of WO, in 
HCl cannot be made perfect, since under conditions where M0O3 is dis- 
solved some WOj is taken up by HCl. A fairly good separation can be 
effected by precipitating both together as Pb salts, dissolving in hot cone. 
HCl, precipitating WO 3 by dilution with boiling water, settling out, etc., 
redissolving in hot cone. HCl again, and reprecipitating again in the 
same manner. If Mo largely predominates better results can be obtained 
by evaporating the HCl solution of the Pb salts (with a few drops of 
HNO3) until nearly pastry, when much of the WO3 will have separated, 
then diluting with 100 to 200 cc. of diluted HCl (3 water ; i acid) 
boiling and filtering off the WO 3. 

Tungsten in Ores, etc, Bomtrager (^Zts. Anal, Chem,, XXXIX., 361). 
Fuse I gm. for an hour with 10 gms. dry NagCOj, leach out with water, 
filter off the residue, ignite and weigh. This weight deducted from i 
gives within 5 per cent, of the proportion of WO3. The solution is di- 
luted to 250 cc, of which 100 cc. is mixed with 15 cc. of cone. HNO3 
and 45 cc. HCl and evaporated to dryness. Take up with a mixture of 
100 gms. NH4CI and 100 gms. HCl in i liter of water and filter. Treat 
the residue with NH4OH, running the solution into a mixture of 15 cc. 
HNO3 and 45 cc. HCl as before. On evaporation to dryness again pure 
WO 3 remains free from SiOj and SnO,. 

Precipitation with Hg(N03)2 is untrustworthy. 

Arsenic in Alloys, Hollard and Bertiaux {Bull. Soc, Chim,, XXIII., 
No. 8). Place 5 gms. of the alloy in a flask with 50 gms. Fe2(S04), 
(absolutely free from nitrous compounds). Then run in 150 cc. pure 
HCl, and heat gently. Pass the vapors through a U tube filled with glass 
beads, immersed in an oil bath kept at 150° to 175° C. The Sb is held 
back in this U tube while AsCL goes on to be caught in a flask beyond 
which should be well cooled and contain 50 cc. of water at the start. 

Arsenic in Insecticides, Haywood {Jour, Chan, Soc,^ XXII., 568). 
Paris green, London purple and other insecticides of the kind should not 
contain free arsenic as that substance <' scorches" the foliage of the 
plants on which it is used. The best method for determining free AsjO, 
in Paris green, etc., was found to be digestion of i gm. of the prepara- 
tion at room temperature in 500 cc. of water with occasional shaking for 
about 10 days. Heating causes decomposition of the Paris green. To 
determine total AS2O3 and Cu, the material was boiled up with NaOH, 
which gives CuOin the precipitate, and AS3O3 in the filtrate, both deter- 
mined by iodometrical methods. 

Qualitative for Gold in Ores, Doring {Berg, u. Hutm, Ztg,, LIX., 
49, 73, 97). Extract with Br and ether, impregnate a piece of filter 
paper with the solution, and incinerate. If Au is present the ash is purple. 
The paper should previously have been impregnated with Mg(N03) and 
exposed to fumes of (NH4)2C03. With ores containing pyrites roast 
100 gms. of the ore, agitate with a few cc. of the Br ether mixture, dilute 
with 50 cc. water after 2 hours standing, evaporate the filtered solution 
down to about 10 cc, and test with a little Br water followed by SnCl,. 
The purple of Cassius reaction will show even if 0.5 gm. Au per ton is 
present The presence of Te may give a coloration which will mask the 


Volumetric for Gold and Platinum, Petersen (Z/x. Anorg, Chem,^ 
XIX. y 59). KI solution when sufficiently concentrated and cold reacts 
with RCI4 affording Ptl, and Ij. The I set free may be titrated with 
standard Na^S^Oj. With AuCl, the reaction is similar, Au I and I, being 
produced. In titrating this solution, however, with NajSjO,, it is only 
after NaAuS^O, has formed that the reagent acts upon the free I. Every 
molecule of Au requires therefore, 3 mol. of NajSjO,. 

Commercial SeparcUion of Platinum and Gold, Priwoznik ( (Est, Zts, 
Berg, u. Ifuttenwesen, 1899, 356). Digest the filings with HNO, (of 

sp. gr. 1-199) ^°^^^ ^^ more Ag is dissolved A small amount of Pt 

also goes into solution. After washing the residue, it is treated with the 
following mixture : 

Cone, HCl 100 vols. 

Cone. HNO, 43 vols. 

Water 143 vols. 

When the AgCl formed on the surface of the metal stops the action — 
decant off the solution, dissolve off the AgCl by NH^OH, and treat again 
with the acid. After six successive treatments the metallic residue con- 
sists of pure Pt. The acid solution is evaporated down with excess of 
HCl, until AuCl, crystallizes out. It is then diluted a little, and a small 
amount of Pt separated by adding NH^Cl. The Au is finally separated 
out by FeSO^. 

It may be necessary to fuse up the metals with 3 parts of Zn prelimi- 
nary to the operations here described. On dissolving out the Zn with 
H3SO4 the metals are left in a spongy mass suitable for the above treat- 

Mercury Cyanide, Vincent {four, Pharm, et de Chim.y X., No. 12). 
The method recommended is combustion with soda lime, which affords 
the Hg in metallic form, and the N of the CN as ammonia. The test 
analyses show very good results. 

Estimating Lead by Electrolysis. Marie {Bull. Soc, Chim,, XXIII., 
No. 12). To obtain from PbSO^ a solution which can be used for 
electrolysis it has been found possible to dissolve by heating on a water 
bath with HNO,, adding crystals of NH^NO, from time to time. After 
dissolving, dilute with warm water, and conduct the electrolysis at 60 to 
70°. 0.3 gm. PbSO^ requires 5 gms. NH^NO,. HNO^ must be used in 
such proportion that after dilution 10 per cent, of the acid should be 
present. PbCrO^ dissolves still more readily in these reagents. The 
process is especially convenient for determining lead in flint glass, etc. 
The pulverized glass is treated with HF and H^SO^ to remove SiOj, 
and the remaining sulphates dissolved as described. The presence of 
free H^SO^ prevents the solution by this means. 

Red Lead, Tocher {Pharm, f,y LXIV., 310). In the cold normal 
HNO3 (sp. gr. 1.05) resolves the oxide into PbOj and Pb(N0g)2 
2.064 gms. of the red lead are treated cold with 50 cc. HNO, of the 
above strength. When the change is completed, the mixture is brought 
nearly to boiling, and 50 cc. of fifth normal HgCjO^ added, it is then 
boiled, and after adding dilute H3SO4 titrated back by standard 

Separation of Bismuth from Lead, Clark {f, S. C. /., XIX., 26). 

y boiling the chloride solution with steel turnings Bi alone is precipi- 


tated. All the Fe must not be allowed to dissolve or some Bi will be 
brought back into solution. Filter off the mixed Fe and Bi, dissolve in 
HCl with KCIO,, and precipitate out Bi by HjS. 

Bismuth CobalHcyatiide. Mathews {J, Am, Chem, Sac, XXII., 214). 
Precipitate with potassium cobalticyanide (vid. supra) for separation of Bi 
in pig leads. Unless i cc. of the solution contains at least 0.0079 §^* 
Bi the precipitation is incomplete. Best when boiling. Large excess of 
the reagent also free acetic increase the insolubility of BiCo(CN) ^ . When 
it is once formed it does not readily redissolve. 

Bismuth Titration, Reichard {Zts, Anal, Chem,y XXXVIII., 100 . 
The Bi is converted to bismuthic acid by oxidizing with CI, in a solution 
containing excess of alkali, which is meantime boiled until the color of 
the precipitate changes to the deep red of HBiO, (pentad Bi). This is 
filtered off, well washed and then boiled with known amounts of standard 
NaOH solution of AsgOg (0.01 gm. As^O, per cc.) until the color has 
been reconverted to the white of Bi(OH),. Then acidulate strongly 
with H^SO^ and Alter hot. Estimate the ASjO, remaining unoxidized 
by the reaction with standard K^Mn^Og. 

Copper in Cyanide solutions. Clennell (/. S. C, /., XIX , 14). The 
presence of Cu in an ore makes a great difference in the feasibility of ap- 
plying the cyanir'e extraction. The amount of Cu in the ore is not so 
important as the question of its being in a condition to use up KCy. To 
determine readily the amount of Cu taken up by the cyanide under 
given conditions of treatment is therefore desirable. The method pro- 
posed depends on i. Precipitation of CuCyj by addition of dilute min- 
eral acids. 2. Non-sensitiveness of methyl orange to HCy or COg. 
3. That the addition of mineral acid to a KCy solution containing 
double cyanides of Cu, gives no permanent precipitate of CuCy, until 
all alkali and free KCy have been neutralized. At this point the solu- 
tion is neutral to methyl orange. 

Place a mt^asured amount (10 to 50 cc.) of the clear solution to be 
tested in a 100 cc. flask, and add standard H^SO^ from a burette with 
constant shaking until there is just no permanent precipitate (a slight 
milkiness may be perceptible but no granules of precipitate). Note the 
reading of the burette, and then run in enough H^SO^ to completely 
precipitate all the CuCyj : until the precipitate ceases to increase, as 
near as may be judged. Filter off a little of the solution, and test it : 
by methyl orange to which it should show acid ; and by dilute HjSO^, 
whi'^h should give no more precipitation. Read the burette again. Fill 
the flask to the mark, filter off 50 cc. and titrate the clear solution back 
with NajCOg solution standardized against the HjSO^ used (methyl 
orange indicator), so as to determine the proportion of HjSO^ nec- 
essary to precipitate the CuCyj. 

Commercial Analysis of Copper, Holland {^Bull. Soc, Chem,, XXIII., 
No. 8). 10 gms. are dissolved in HjSO^ and HNOg in a tall beaker, 
in which the first electrolysis is effected. Crude coppers usually afford an 
insoluble residue which is not filtered out at this stage. 

Cu and Ag. Electrolyze. Volume 200 cc; current 1 amp.; time 24 
hours. In any case weigh the deposited metal with the usual precau- 
tions. If As and Sb are present, the action of the battery is stopped 
when the blue color disappears (leaving 0.05 gm. Cu or less in solution). 


Sb, By evaporating with HCl, passing in HCl gas and finally adding 
HCl solution saturated with H^S, conditions are obtained under which 
the major part of the As only is precipitated. After filtering through 
glass wool, the Cu and Sb are precipitated by diluting and adding H^S 
water, leaving a solution to be treated for, Ni Co and Fe. The precipitated 
sulphides (Cu and Sb with a trace of As) are redissolved in KCy and Na^S 
solution, and Sb electrolyzed out. Volume, 220 cc; current, 0.05 
amp.; time, over night. 

Residual Cu, D^ompose sulphide and cyanide by H,S04 and evap- 
orating, remove the As remaining by heating with bromized HCl, and 
electrolyze. Volume, 100 cc; current, i amp. 

Ni and Co, The bases in the solution referred to above are converted 
into sulphates by evaporation with H3SO4 (adding a little HNO, to 
peroxidize Fe,) diluted, over-neutraliz«l with NH4OH, and without fil- 
tering off Fe2(0H)j electrolyzed. Volume, 250 cc; current, 0.5 amp.; 
time, over night. 

Fe, The precipitate of Fe,(OH)j if filtered off, and Fe determined 
by titration. 

Pb, Separate portion of 10 gms. dissolve in HNO,. Electrolyze. 
Volume, 350 cc; current, 0.5 amp; time, 18 hours: weigh anode 
with adhering PbO,. 

Ag, Dissolve the Cu (obtained by electrolyzing as in first paragraph 
or if Ag is very small, 20 to 50 gms. of the original sample) in HNO3, 
precipitate by a little HCl, keeping the solution at 70® C. Dissolve the 
AgCl in KCy and electrolyze. Volume, 250 cc; current, 0.05 amp.; 
time, over night. 

Instead of weighing the deposit, dissolve off with HNO,, and use 
Volhard titration. This titration is rendered more exact by just over- 
running the addition of NH^CNS, (Fe,(S04), indicator) and then 
cautiously titrating back with standard AgNO, (2 gms. Ag per liter) 
until there is a sharp and complete disappearance of the red color. 

As, Put in a flask 5 gms. of the sample with 50 gms. pure Fe2(S04)g. 
Connect up for distillation and run in through a stopcock funnel 150 cc 
of HCl. Between the flask and condenser is interposed a U tube con- 
taining glass beads, which is immersed in an oil bath kept at 150^ to 
175° which holds back Sb. Determine As in the distillate; best by 
neutralizing and titrating with standard I solution. 

Au, Dissolve 100 gms. of the sample in 750 cc HNO,. Filter, 
burn, add Pb, cupel and part the insoluble portion. 

S. Dissolve 5 to 20 gms. of the sample in aqua regia. Precipitate by 
BaCl,, etc. 

Commercial Coppers, Clark (^J. S, C, /., XIX., 27). The Cu may 
be separated free from all Sb, As or Sn, by precipitating in dilute HCl 
solution with KI, adding afterward Nag SO, until all free I has been re- 
moved. Unless free HCl is present some Sb may remain with the pre- 
cipitate. This should also be washed with dilute HCl. To the filtrate 
add H2C4H4O3, render alkaline and separate traces of Cu, etc., by 
Na^S, filter and acidify to obtain As, Sb and Sn as sulphides. Distilling 
the mixed sulphides with strong HCl into water, drives over the As which 
reforms AsjS, in the distillate. Boiling the contents of the flask with 
steel turnings precipitates metallic Sb, and in the solution the Sn may be 
separated by H2S and determined as SnO,. 


Copper and Chilled Slag. Shelby {^Eng. and Min. Journ., LXIX., 
70^)- 5 S^^* o^ ^^% ^^ moistened with 2 or 3 cc. of water, and then 
are added 10 cc. of cone. HCl and i cc. HNO3, The mixture is evapo- 
rated and heated until the Si02 is dehydrated, when it is cooled, 10-15 
cc HCl added, and after boiling and dilution, the solution is filtered. To 
the filtrate is added satm^cted solotton of NajS^O, sufficient to reduce 
Fe, and then a strip of Zn 1x3 inches. This causes formation of enough 
HjS to completely precipitate the Cu as CuS. When the Zn is all dis- 
solved, filter off the CuS, dissolve in HNO, and determine Cu by cyanide 
or by iodide method. 

Determining Thallium as Sulphate. Browning (Am, J, Set , Feb., 
1900). Heating precipitated TlCl with excess of H2SO4 in a crucible at 
first exposed to a temperature of 220° to 240° and then to the tempera- 
ture produced by placing the crucible 5 cm. above an iron plate heated 
to dull redness, until the weight was constant, afforded TIHSO4. Heat- 
ing to constant weight at dull redness afforded TI2SO4. 

Colorimetric for Vanadium, Maillard (Bull, Soc, C^/m. , XXIII. , No. 
12). For estimating small amounts the red coloration produced by 
H3O2 or ozonized ether was found to be proportionate to the amount of 
V2O5 present. A standard solution of sodium metavanadate was pre- 
pared for comparison. 

To the comparison solution as well as to the solution under examina- 
tion were added 2 to 5 cc. of pure HCl, and 3 to 10 cc. of ether satu- 
rated with H202» and shaking. After bringing the aqueous solutions up 
to the same volume (15 cc.) the comparison was made in a Duboscq 
colorimeter. The ether is prepared by shaking together equal volumes of 
ether and a lo- volume solution of H2O2. With solutions containing 
i-iooth of V in 10 cc, it was found necessary to add at least 4 cc. HCl 
and 10 cc. ether; with more dilute solutions the use of a sufficient excess 
of the reagents did not need special consideration. The coloration de- 
velops immediately, and lasts for hours, even days. Experiments showed 
that the reaction was unaffected by presence of KCl, NH4CI, K2SO4, 
KNO„ Na2HP04 or CaClj. 

Volumetric for Cerium, Browning (^Am, Jour, Sci., VIII. , No. 48). 

The CeO, obtained by ignition of the oxalate in the air dissolved by 
heating with HCl in presence of KI sets free I, CeClj being formed. 
After cooling the free I can be determined by standard NajS^Og. 

In Bunsen's method the decomposition was effected in a sealed fiask. 
The author finds it can be accomplished in a stoppered fiask, the air in 
the flask being dispossessed by passing in CO 2 before applying the heat, 
which need only be such as is obtained by standing the fiask on a steam 
radiator for an hour. 

Stolba's assertion that CeC204 may be titrated with standard KjMnjOg 
in the same manner as calcium (warm H2SO4 solution) without any oxi- 
dizing power of the K2Mn20g being destroyed by the presence of the 
cerous compounds was found to be correct. 

Volumetric Determination of Hydrogen. Colson (C. Eend,^ CXXX., 
330). AgOH, prepared by precipitating the nitrate by KOH, and dry- 
ing in vacuo at 1 10° C. was found to take up H, affording Ag-f HjO. The 
reaction is slow in the cold but rapid with the moist gas at 100^ C. A 
separation from CH4, C2H4 orO may thus be affected. 


Chlorine in Bleaching Powder. Wolowski (Z/x. Anal, Chem,^ 
XXXVni., 711). The solution of the bleaching powder is run into a 
standard solution of KI acidified with H^SO^ until a few drops of chloro- 
form placed in the containing flask at the start, shows only a faint pink 
color on vigorous shaking, the I having been converted to ICI3 thus 
HI+HCl=HCl+IClg. If the titration is overrun, a fresh test rou t be 

Bromide in Presence of Chloride and Iodide, Weszeliky (Z/x. 
Ancu, Chem., XXXIX., 81). If no I is present add about 1 gm. KjCOg 
and CI water sufficient to oxidize all Br to KBrOg. Evaporate to dry- 
ness, dissolve in 100 to 150 cc. water, acidify, add KI and titrate with 
Na^SjOg. I cc. N/ 1^ NajSjOg =■ 0.001333 Br. 

If I is present, put the solution in a distilling flask with ground glass 
connections, acidify — add enough CI water to set free Br, and convert I 
to HIO3. Distil into a solution of 0.5 to i gm. KOH. Before the dis- 
tillation is quite complete, pass a current of CO^ through the apparatus. 
Rinse out the delivery tube by allowing the flask to cool slightly until it 
has drawn up some clean distilled water from a beaker substituted for the 
receiver, and forcing it out again by the CO 3. Treat the distillate as de- 
scribed above for Br (evaporation, etc.). 

The contents of the flask now containing HIO, may receive the addi- 
tion of KI and (in case of the absence of Fe, As or Sb) be titrated for 
I. One-sixth of the I indicated by Na^SjOj titration represents the I 
originally present. 

Pettenkofer' s test for Carbon Dioxide, Letts and Blake {Sci, Proc, 
Roy, Dubl, Soc, N.S. II., 107). Ba(OH)j when kept in glass alters in 
strength in consequence of action on the glass. It is recommended that 
it be kept in bottles coated inside with paraffin, and that the test be con- 
ducted in a jar lined in the same way. 

Carbon in Iron and Steel Re-use of Copper solution. Sargent (J, Am, 
Chem, Soc, XXII., 210). The solution of 2KCI, CuClj for dissolving 
off" the Fe to leave " Total Carbon ** in a condition to be estimated, may 
be re-used several times until the Fe salt accumulates too far. Re-oxida- 
tion of the Cu salt is, however, necessary. The author finds that the 
simplest and easiest method consists in passing CI through the solution for 
some time. In laboratories where many determinations of the kind are 
made every day the saving in the reagent is very considerable. 

Ferrocyanide Determination. Donath and Margoscher {Zts, Aongen- 
Chem,f 1899, 345)* I^ ^^^ case of old purifying materials, grind finely, 
and take 50 gms. for the test. Treat in a liter flask with 100 to 150 cc. 
of 15 per cent, solution of KOH. Keep in a warm place with frequent 
shaking for some time. Then fill up to 1030 cc. the volume of the in- 
soluble material having been found to be 30 cc. Mix well, and filter 
through a dry filter. Take an aliquot portion of the filtrate for the esti- 
mation. Add solution of bromized soda (proportions : 80 gras. NaOH 
made up to i liter and 20 cc. Br added). By prolonged heating the 
cyanogen compounds are completely destroyed, and Fe2(0H)g separates 
in which the Fe may be filtered off" and determined by well-known 
methods. The amount of Fe thus found multiplied by 7.5476 gives 
amount of KjFeCy^, 3H3O. Using the factor 6.5833 gives the anhy- 
drous salt. 


Methods for Estimating Fluorine. Harker (CA^/«. News^ LXXXII.,56 
and 64). In examining topaz and other minerals comparative trials were 
made with different processes for estimating HF. Berzelius method — 
fusion with alkaline carbonates — gave low and in some cases irregular 
results. Liversidge*s method (Crookes Select Methods ^ edition of 1884, 
p. 580) decomposition with H^SO^ and SiOj passing the gas into 
NH^OH and obtaining from this KjSiFg for weighing. Affords uniform 
results but rather low. Berzelius* method as modified by Rose, fusion 
with alkaline carbonate and SiOg afforded very good results. 

Qualitative for Boric Acid. Borntrager (Zts. Anal. Chem.^ XXXIX., 
92). BO 8 heated on platinum foil gives a green flame, borates do not. 
Heating borates with HF only or with NH4NO3 and NH^Cl, or HjSO^ 
and HCl, or H2SO4 and HNOg, or HCl and HNO,. The coloration 
is more immediate and intense than when alcohol is used with one acid 

Boric Acid in Boracite, Schwartz {^Chem. Ztg.y 1899, 497). Decom- 
pose c to 2 gms. of the finely powdered mineral in a flask fitted with re- 
versed condenser by use of 5 to 10 cc. HCl in 100 cc. of water. Let 
stand several hours, filter, wash thoroughly, neutralize exactly by use of 
fifth normal soda and methyl orange, make up to exactly 200 cc. or some 
convenient bulk, take out 50 cc. or some other measured portion, add 50 
or 60 per cent, of glycerin, then a little phenol-phthalein and titrate 
B3O2 with fifth normal soda. 

Errors in Volumetric for Boric Acid, Stock (C Rend.^ CXXX., 
516). The presence of CO^ in the reagents for titration of B^Oj is a 
source of serious error. It can only be avoided by boiling out the CO, 
from all the reagents used, and titrating with standard NaOH from which 
CO2 has been separated by addition of BaCl2. 

Phosphoric Acid. Sherman and Hyde (J, Am. Chem. Soc, XXL, 
652). Ihe method is a modification of that of Woy, i. e., heating the 
** yellow precipitate" at a temperature just below red heat (the crucible 
being supported inside of another, and the outer one heated to dull red), 
and weighing the residue — P205,24MoOg. For precipitation the 
HNO3 solution 150 cc. was kept at 50° C. and a 3 per cent. (NH4)2Mo04 
solution added gradually with stirring. The success of the operation 
appears to depend largely upon the care exercised at this stage. The 
principal variation from Woy's method consists in omitting re-solution 
and reprecipitation of the *' yellow precipitate" which he seemed to re- 
gard as essential. 

Phosphorus in Steel. Ibbotson and Brearley (Ch. News, LXXXIL, 
55). After obtaining the yellow precipitate the Mo therein may be 
separated as PbMo04 and weighed as such. Dissolve 2 gms. of the 
sample in 45 cc. HNOgCgr. 1.2) add permanganate until a permanent 
pink color or MnO, precipitate persists after 2 minutes' boiling. Clear 
with FeS04, add about 4 cc. NH4OH, then 30 cc. molybdate solution. 
Shake, let stand a few minutes at 70 to 80° C. Filter, wash, dissolve 
in a little NFI4OH, passing the solution a second time through the filter, 
add to the solution 10 to 12 cc. HCl and 10 cc. Pb(C2H302)2 (40 gn^s. 
per liter). Heat in the flask in which precipitation was made 10 to 12 
gms. NH4CI with 50 cc. strong NH4C2H3O2 solution. Mix with the 
other solution, shake, filter, ignite and weigh PbMo04. This weight 
multiplied by 0.007 gives that of the P. 


Basic Siag. Dafert and ReitmaT (Z/x. landw, Ver* station 6st,^ 1899, 
75). It is proposed that the slag should be valued by the per cent, of 
total P2O5, of which 90 per cent, should be soluble in formic acid. 

Total PjOj is determined in HNOj extract of the slag. 

Of another portion 5 gms. is agitated with a 5 per cent, solut'on of 
HCHO2 for \o minutes, which is then filtered off and the dissolved 
PjO^ determined in the filtrate. 

Titrating Per sulphates. Le Blanc and Eckhardt (^Zts,f, Ele, Ktro, 
Chem,, 1899, 355)' Simple heating of the salt effects its decomposition, 
^. g.y pure KjSjOg affords K^SO^ + SOj + O^. The oxidizing power 
is the important feature, and if any of the NH^ compound is present, re- 
sults by the method would be fallacious. Addition of a known quantity 
of ferrous salt to be partially oxidized by the persulphate, and then titrat- 
ing the excess of ferrous iron present, when performed at ordinary tem- 
peratures, gives low results unless the ferrous salt is very largely in excess. 
If, however, the titration is performed at a temperature of 70 to 80® C, a 
moderate excess of ferrous salt may be used, and correct results obtained. 

Detection oj Nitrogen. Jacobsen^s reaction. Tauber {Ber.^ XXXI., 
3150). The Jacobsen reaction for nitrogen in substances containing S 
consisted in heating with about five volumes of iron filings, and sub^- 
quently with metallic potassium, thus eventually producing ferrocyanide 
which in aqueous solutions was recognized by the ordinary test. 

The author finds that under the conditions mentioned the metallic iron 
fixes the nitrogen of the atmosphere, and if the action is continued long 
enough considerable quantities of cyanide are formed. )t appears that 
some years ago, patents for this process of manufacturing cyanides were 
granted in Germany but have been abandoned. 

Kjeldahl Process. Procter and Turnbull (y. S. C. /., XIX., 130). 
Dry CUSO4 was found to replace satisfactorily the use of Hg when the 
operation was conducted thus : In a 500 cc. fiask of Jena glass were 
placed 0.5 gm. of the material to be analyzed, 2 gms. dry CUSO4 and 
20 cc. cone. H2SO4. Heat was then applied. After the first vigorous 
action was over (about fifteen minutes) 10 gms. dry powdered K2SO4 
was added and the heating continued until the solution was clear (requir- 
ing about thirty minutes) After cooling about 70 cc. of water was added 
with some pure zinc, and the distillation, etc., conducted essentially in 
the ordinary manner after adding slowly 100 cc. of 50 per cent. NaOH 
solution. The distillation requires about thirty minutes. 

Nitrite Determination. Gailhart (y. Pharm. Chim. [6], ,XII. No. i). 
If a neutral solution of a metallic nitrite is added to an excess of a neu- 
tral, boiling concentrated solution of NH4CI the reaction occurring is : 

NH4CI + MNO2 = N2 + 2H2O + MCI. 

The reaction can be conducted in a Schlossing apparatus and the evolved 
N measured. Presence of nitrites does- not interfere. 

Nitric Acid in potable waters, fertilizers, etc. Pool (through Chem. 
NewSf LXXXI., 47). Evaporate in an Erlenmeyer to dryness with ad- 
dition of excess NaCl. Treat the dry mass with H2SO4 in an atmosphere 
of CO2 and boil off the CI into a solution of KI in which the I is after- 
ward titrated. Operating in presence of CO2 the NO is not oxidized into 
products capable of liberating I and only the CI shown in the equation 


does that work. 6HC1 -f 2HNO, == 2NO + 4HjO + 3CI2. Hence 
every \^{=^ 162) corresponds to 2HNO =126. 

Kjeldahl method Applied to Nitrates. Field's modication. Veitch (/. 
Am, Chem. Soc, XXL, 1094). H2SO4 containing 34 gms. salicylic 
acid per liter is used, 35 to 40 cc. for each determination. After leaving 
it in contact with the nitrate in the cold until the latter is dissolved, 607 
gms. pulverized itj^ ^^ added, and the mixture heated over a low flame 
for 15 minutes, then at full beat until clear, the rest of the operations 
being as usual. 

Nitrates, Commercial, v. Wissel (^Chem, Centr.^ II., 144). Com- 
parative tests were made with: i (Mockern's Method). Reduction in 
alkaline solution with Znand Fe. 2 (Ulsch's). HjSO^ and Fe filings ; 
3 (Devarda's). In alkaline alcoholic solution with alloy of Al, Cu and 
Zn ; 4 (Forster's). Heating with salicyl sulphuric acid. 

Devarda's method proved the best, next came Ulsch's and then For- 
ster's. Results by Mockern's method were always low. 

Nitric Acid and Mixed Acid. Van Gelder (/. S. C. /., XIX., 508). 
The method of testing acids, in connection with the manufacture of nitro- 
glycerine and other nitro-compounds is given. Nitric acid. About 10 
gms. are weighed in a bottle, the bottle opened under water, and the so- 
lution diluted to i liter. Lots of 100 cc. each arc then taken out for the 
tests. One lot is titrated with standard NaOH for total acidity. For 
the " low oxides " (lower than N2O5) 10 cc. are diluted to 100 cc. and 
titrated with tenth normal K^MnjOg. (In the discussion it was stated 
that the usual method was to run the acid from a buiette into the 
KjMnjOg). The results of the titration with NaOH calculated to HNO^ 
gives total acidity. 

The results by K2M2nOg titration are calculated to NOj. 

To obtain '* origmal " HNO, deduct from the HNO3 by alkali test the 
HNO3 equivalent to NOg determined by KjMuoOg. 

To obtain "available " HNO3 deduct from HNO3 by alkali test half 
of the HNO3 equivalent to NOj by K^Mn^Og. 

For mixtures of H 2 SO4 with HNO3, the dilution ^^ managed in the 
same way as above. On one 100 cc. lot the total acidity is determined 
by alkali test. Another lot of 100 cc. is evaporated on the water bath. 
When HNO3 has been driven out, a few drops of water are added and 
the evaporation continued. After ten minutes this is repeated and again 
after the same interval. After three repetitions the acid is diluted, and ti- 
trated with alkali, which gives HjSO^. The low oxides are determined 
as given above. The presence of As in the HjSO^ causes some error, 
but its influence is generally disregarded. 

Calcium Carbide. Erdmann and Unruh (y. Pr. Chem., LXI., 233). 
The quality of this material is usually determined by the volume of gas 
yielded by a known weight of materials. This method is usually erro- 
neous as the gas almost invariably contains CH^ and Hj. The error is 
much reduced if the valuation is effected by determining the loss of 
weight in an apparatus constructed similarly to those for determination of 
CO 2 by loss. The residue from the reaction can be titrated by standard 
HCl aod phenol -phthalein and the insoluble residue filtered off and 

Commercial Calcium Carbide. Moissau {Bull. Soc. Chim.y XXI.). 


Theoretically i gm. CaC should give 349 liters of C^H, gas on decom- 
posing various saniples with milk of lime saturated with €3^2 ^^^ ^^' 
umes obtained were (at o^ and 760 mm.) 392.8 up to 318.77. If the 
carbide lacks a crystalline appearance the figure may be much less (228.6 
to 260.3). To examine the residue left by the reaction the carbide was 
decomposed by use of sugar solution in order to keep Ca(0H)2 in solu- 
tion. The residue was washed successively with water, alcohol and ether 
and then dried in vacuo. It was found to consist of silicides of carbon, 
calcium and iron together with a little CaS and graphite. Treatment 
with dilute HCl removed Ca, Fe, Al and P. CSi and graphite remained 

Naphthalene in Coal Gas. Colman and Smith (y. S. C /., XIX., 
128). It has been found that naphthalene forms a picrate C^qH,, 
CgH,N,07, which is insoluble in aqueous solution of picric acid. How- 
ever passing gas containing naphthalene through a picric acid solution 
gives a precipitate containing some free naphthalene, the complete com- 
bination of which with the picric acid must be effected by heating. The 
method consists in passing the gas through a known amount of solution of 
picric acid in a series of three or four bottles, the strength of the picric 
acid solution having been accurately determined by titration with standard 
alkali, using lacmoid as indicator. The rate is about i cubic foot per 
hour. 10 to 15 feet maybe necessary. The contents of the bottles were 
then united and a stopper inserted cariying aglass tube closed at the 
lower end but having a small opening at the side below the stopper. The 
glass tube is then connected with an exhaust pump, and after the air has 
been exhausted, the tube is drawn up until the side opening is closed by 
the stopper. The bottle is then heated to complete the combination, after 
which it is cooled, the contents filtered and the picric acid remaining 
titrated back by standard alkali. 

Examination of Coals for Gas Manufacture. Rhodin (y. S. C /., 
XIX., 12). The author uses charges of about 100 gms. of the coal care- 
fully sampled, in an unglazed porcelain retort set in a Fletcher injector 
furnace. For hydraulic main and condenser he uses Woulfe bottles con- 
nected by means of long U tubes (inverted). After these comes a puri- 
fier made of a Fresenius tower, and filled with soda lime. The gas is 
then conducted into a bell gas bolder having a capacity of about 30 
liters, provided with graduations, from which the gas made can be drawn 
for tests on illuminating power. It is well to run a preliminary charge 
of 50 gms. in the apparatus to drive out the air or other gases in the 
system, as also to gain an approximate idea of the size of the charge best 
adapted for the actual test. 

New Coal Calorimeter. Parr (y. Am. Chem. Soc.^ XXII., 646). The 
combustion chamber consists of a short copper cylinder, with caps which 
screw on. One cap is provided with a valve through which a red-hot cop- 
per wire may be dropped to start the combustion. The charge is made 
by mixing i gm. of the pulverized coal with 16 to 18 gms. of Na^Oj, or 
in that proportion. Half those quantities were used in many of the de- 
terminations of the author. The correction for the hot wire was found 
to be per half inch of No. la copper wire 0.012° C. or o.oai° F. The 
correction for the combination of the products of combustion with the 
NajO was found (by experiment with sugar carbon) to be 27 per cent, 
of the total indicated heat. For deta\ls of the apparatus the original 
paper must be comsulted. 




Professor Chandler returned from Europe at the opening of the term 
after an extremely interesting and successful summer. 

On July 7th he was honored by receiving the Degree of Doctor of 
Science, honoris causa, from the University of Oxford, this being the 
second occasion upon which the degree has been awarded to an American. 

About ten days later, as the first foreign president of the Society of 
Chemical Industry, he presided at the general meeting of the Society in 
London, and on July i8th, in the historic theater of the Royal Institution, 
he delivered his annual address before a large audience. The address, 
which is published in full in the Journal of the Society of Chemical In- 
dustry iox July, 1900, was entitled "Chemistry in America," and dis- 
cussed chemical and technical education, chemical research and chem- 
ical industries in the United States. 

The next night he presided at the annual dinner of the Society and 
proposed the health of the Queen, and also, in the absence of Mr. 
Choate, responded to the toast of "The President." During the follow- 
ing week, the society made numerous excursions to places of interest in 
and near London. On July 20th they were handsomely entertained at 
Oxford by the authorities of both the city and the university ; and on the 
2 1 St the Society adjourned, to meet at Paris the next day, for another 
week's session. 

After leaving Paris, Professor Chandler travelled in the south of France 
and Italy, and has brought back to the school, for the Museum of Chem- 
istry, some rare chemicals from the Paris Exhibition, a series of beautiful 
incrustations from the calcareous springs at Clermond Ferrand, and sam- 
ples of ancient pottery and building material from Pompeii. 

At the end of last June, the American Chemical Society, in conjunc- 
tion with the Section of the American Association for the Advancement 
of Science, held their general meeting at Havemeyer Hall and at the 
Chemists' Club. Of the forty- eight chemical papers upon the program, 
nearly one-fourth were by officers of Columbia University, thereby illus- 
trating the amount of original work which is being done in the Chemis- 
try Department. 

The papers were as follows : 

1. "The Composition of Abietic Acid." (In abstract.) Hermann A. 

2. "The Direct Synthesis of Ketodihydroquinazolines from Ortho- 
amido Acids." Marston Taylor Bogert and August Henry Gotthelf. 

3. "The Direct Synthesis of Ketodihydroquinazolines from Ortho- 
amido Nitriles." Marston Taylor Bogert and August H. Gotthelf. 

4. "Experiments with Some Substituted Benzoic Acids and their 
Nitriles." Marston Taylor Bogert and August H. Gotthelf. 

5. " Some Analyses of Milk and Cream, with reference to the Condition 
of Fat Globules." H. C. Sherman. 


6. '<Note on the Ferrocyanides of Lead and Cadmium/' Edmund 
H. Miller and Henry Fisher. 

7. " The Direct Preparation of Imides of the Bibasic Acids from the 
Corresponding Nitriles." Marston Taylor Bogert. 

8. " Investigation as to the Nature of Corn Oils." Second paper : 
'< Determination of the Constitution." Herman T. Vulte and Harriet 
Winfield Gibson. 

9. ''Note on the Determination of Phosphorus as Phospho-molybdic 
Anhydride." H. C. Sherman and H. S. J. Hyde. 

10. <*New Methods for the Separation of Some Constituents of 
Ossein." William J. Gies. 

11. '' Notes on the Constituents of Ligament and Tendon." William 
J. Gies. 

The Chemical Library. 

The library, which has for the last three years been situated on the 
ground floor in Havemeyer Hall, is to be converted, in the immediate 
future, into a " Science Seminar Room " under the immediate supervision 
of an officer of the Library Department, and will be open to all visitors 
between the hours of 8:30 a. m. and 11:00 p. m., or, in other words, 
during the same hours as the main library. It is proposed not to limit 
this room to the literature of chemistry alone, but to concentrate in it the 
books and journals relating to other branches of science. 

General and Industrial Chemistry. 

At Barnard College, Miss Margaret £. Maltby, Ph.D. (Goettingen), 
formerly professor at Wellesley College, has been appointed instructor in 
the place of Dr. H. T. Vulte, who resigned his position there last year 
so as devote himself more exclusively to his work at the College of 
Physicians and Surgeons. Dr. Maltby is assisted by Miss Eleanor 
Keller, A.B., a graduate of Barnard College. 

At the College of Physicians and Surgeons Dr. Vulle, who is in charge 
of the laboratory of General Chemistry, is assisted by Mr. G. A. 
Goodell, A.M. (Columbia), who was connected with him for some years 
at Barnard. Mr. Goodell takes the place, as tutor, of Dr. J. A. 
Mathews, who last year was appointed to the Barnard Fellowship, and is at 
present studying abroad. Dr. H. Chambliss, Ph.D. (Johns Hopkins), takes 
the place as assistant, of Mr. L. L. Watters, who is now studying at 
Columbia for a Ph.D. in analytical chemistry. Professor Pellew has been 
rewriting his laboratory book on general chemistry, which is being 
issued to the students in the form of advanced sheets. It will appear in 
book form next spring. 

In the laboratory of industrial chemistry, quite a number of special 
and post-graduate students are working this term. On Friday, October 
19th, Mr. S. A. Tucker, instructor in industrial chemistry, and Mr. 
H. R. Moody, who is pursuing post-graduate studies in the depart- 
ment, read a paper before the Society of Chemical Industry, embodying 
the results of their elaborate investigation, last year, upon the electrolysis 
of calcium chloride solutions, for the formation of chlorate. This year 
they are preparing to investigate the influence of gases at the temperature 
of the electric arc, in the production of new compounds. 


Analytical Chemistry and Assaying. 

Since the resignation of Professor Ricketts last May, the temporary 
arrangement which existed during his absence on his sabbatical year,« has 
been continued. Dr. Wells is in charge of the qualitative laboratory and 
Dr. Miller in charge of the quantitative laboratory, the assay laboratory, 
the graduate work in analytical chemistry, and the apparatus room. 

Qualitative Analysis, — Dr. Lenher, for the past two yeats assistant in 
qualitative, has resigned to accept the position of assistant professor at the 
University of Wisconsin. Dr. W. A. Dreyfus, a graduate of the Univer- 
sity of Munich, takes his place. On account of the large increase in the 
number of students, an additional assistant has been provided, Mr. A. C. 
Neish, A.M., a graduate of Queen's University, Toronto. The capacity 
of the laboratory has been increased during the summer by the addition 
of thirty-two new desks so that there are now places for two hundred and 
twenty-eight students, all of which are now occupied. 

Quantitative Analysis, — Last spring, Dr. Hermann A. Loos, formerly 
University Fellow in chemistry, was appointed assistant in place of Henry 
Fisher, who is now in the employ of Ricketts and Banks. Soon after re- 
ceiving the appointment. Dr. Loos resigned to accept a position in Chili, 
but to the deep sorrow and regret of all who knew him, died of yellow 
fever on the voyage from Panama. The place was filled by the appoint- 
ment of Dr. Alfred Tingle, a graduate of the universities of Aberdeen, 
London, and Pennsylvania. 

There is a notable increase of students in quantitative work. The 
laboratory is full, every one of the seventy-two desks being occupied, and 
it has only been possible to accommodate all the students by doing away 
with the alternate week attendance of the mining engineers and by taking 
students into the instructors' laboratories. 

Assaying. — The increase in students holds good for the assay labora- 
tory ; the course is given twice a year ; the first section contains thirty- 
two men, and there will probably be twenty more in the second section. 

Some changes have been made in the course. The lectures have been 
cut down from two a week to one, and the practical work has been 
altered, to accord more nearly with the present western practice, by the 
introduction of more work in the mufHe furnaces. 

Graduate Work, — ^There has been a remarkable increase in graduate 
students in analytical chemistry. Besides nine students taking their 
major for M.A., or minor for Ph.D. in analytical chemistry, there are 
seven new men all of whom are doing quantitative work for their theses 
for the doctor's degree. Dr Sherman gives this year, for the first time, 
a new course (Chemistry 15) which is an advanced course in proximate 
organic anal)rsis, including tanning materials, wines, liquors, carbohy- 
drates, proteids, foods, asphalts, explosives and plants. 

The plan of filling the students' desks during the summer with the 
apparatus for the different courses, as practiced in the German labora- 
tories, has now been in operation for two years. This enables the stu- 
dent to begin his work as soon as he receives the key to his desk. 

The greatly increased work of administration and instruction leaves 
scarcely any time at the disposal of the officers for original research, how- 
ever, the following work is planned for the year : 

Dr. Miller : The completion of the analysis of the ferrocyanides of cad- 


mium to determine the variation in the composition of the precipitate 
under different conditioDs. 

Dr. Sherman : The completion of the work on the structure of the fat 
globules in milk ; and the heat of combustion as a factor in the examina- 
tion of lubricating and drying oils. 

Dr. Joilet is preparing an extensive article on the analysb of slags and 
cinders, the first part of which appears in this number. 

Mr. Hogarty will continue his work on the assay of zinc blendes. 

Dr. Tingle will finish an investigation on unsaturated organic acids, 
already well under way. 

The thesis work of the graduate students, and fourth- year analytical 
chemists, has not yet been assigned except in two cases : 

Mr. Watters has started an analytical investigation of the sage plant. 

Mr. Kern will work on the analytical behavior of uranium with 
special reference to its separation from iron. 

Organic Chsmistrv. 

During the past year several changes have occurred in the laboratory of 
Organic Chemistry. At the close of last year's work Mr. E. W. Scherr, 
Jr., the assbtant in this laboratory, withdrew, to take up the study of 
patent law. His place has since been filled by the appointment of Dr. 
Chas. £. Caspari, a graduate of Johns Hopkins University. Dr. Cas- 
par! has had a thorough training in the subject of Organic Chemistry un- 
der the able guidance of Professor Remsen, whose assistant he has been 
for the past year, and will have special charge of the undergraduate work 
in the laboratory. 

Another change of importance is that of transferring the course in ele- 
mentary organic (Chem. 20) from the second to the third year. Under 
the old arrangement the studies of the third year made a serious break 
between the organic chemistry of the second and fourth years. Here- 
after the student in chemistry will begin the study of this branch in his 
third year and continue it without interruption throughout his entire 
fourth year. Another advantage of this transfer is the equalizing of the 
work of the second and third years, the second year having been much 
the hardest, and the third year the easiest, of the four under the old 

Mr. A. H. Gotthelf, who last year received his Ph.D. from Columbia, 
has returned to continue his researches with Mr. Bogert in the organic 
laboratory, several papers upon these investigations having already 

Four researches in organic chemistry, papers concerning which were 
read at the June meeting of the American Chemical Society, are being 
continued in the organic laboratory this year in addition to the following : 

Some experiments with anthranilicnitrile. By Mr. Bogert. 

The synthesis of new quinazolins. By Dr. Gotthelf. 

Researches on Lindera benzoin. By Dr. Caspari. 

A critical study of lauric acid. By Dr. Caspari. 

Experiments with 5-aminosalicylic acid. By Dr. Dreyfus. 

A new synthesis of secondary amines. By Dr. Tingle. 

The reactions of organic bases upon hydroxy- and unsaturated com- 
pounds. By Dr. Tingle. 


An investigation of the berry of Melia azedarach^ Linn6. By Dr. 

A study of the derivatives of roetanitro paratoluidin. By H. St. J. 
Hyde, A.M. 

The nitrophthalic acids and their derivatives. By L. Boroschek, A.M. 

The direct synthesis of cyclopentadiazenons. By D. C. Eccles, B.S. 

With the beginning of the second half year Messrs. Bernheim, Falk, 
Moflfatt and Pickhardt will start original work, so that during the second 
half year there will be a total of at least 19 original researches being con- 
ducted in the organic laboratory, a total far in excess of that of any pre- 
vious year in the history of the school. 

The establishment of several new courses is at present under consider- 
ation. One of these probably will be an elementary course designed es- 
pecially for students not intending to pursue chemistry as a profession. 
It will be somewhat similar in nature to the existing elementary course 
(Chem. 20) for the chemists, but will be more popular and less scientific 
in character, the aim being to make the course both interesting and in- 
structive, touching only upon the most important classes of organic com- 
pounds, and illustrating the lectures by numerous experiments. It is 
believed that such a course will attract many students from the College, 
College of Physicians and Surgeons, etc., who cannot afford the time 
required to master Chemistry 20, but who yet would be very glad of the 
opportunity to take a less detailed course. It will give to the student a 
sufficient knowledge of the subject to take with advantage the elementary 
laboratory course (Chem. 30), or, with outside study, to take up the more 
advanced course (Chem. 21). In the past two years numerous requests 
have come from college students for the establishment of such a course. 

The following advanced courses, embodying special lectures and lab- 
oratory work, are also in contemplation : 

Color chemistry. 

Essential oils, terpenes and perfumery. 

Synthetic drugs. 


With the constantly increasing number of postgraduate students in the 
School of Chemistry, it is believed that before long the demand will be 
sufficient to warrant the establishment of one or more of these advanced 

Physical Chemistry. 

This laboratory is also overcrowded with students, having nearly 
twice as many, both regular and post-graduate, as in any previous year. 
Part of the increase is due to the change in the course by which the regu- 
lar second-year chemists, seventeen in number, take their laboratory 
physics under Dr. Morgan. But the number of post-graduate students, 
working for Ph.D. degrees, is also much larger than in any former year. 
In order to give greater facilities, an adjoining room, hitherto occupied 
by one of the professors, has been vacated, and is being fitted, as rapidly 
as possible, as an annex to the laboratory of ph3rsical chemistry. But, 
in spite of this, the department is seriously cramped in its work by lack 
of space. 

Four investigations have already been started, and others will be com- 
menced as soon as there is room enough. During the last three years. 


since work in this laboratory was commenced, one book and fifteen 
articles have been published, and it is hoped that this number will be 
largely added to, this year. 


We regret to record the severe and dangerous illness of Frank C, 
Hooper, who was recently promoted to be Instructor in Mining, and 
who expected to begin his new duties in October. After his return from 
the Summer School held in Colorado in July, Mr. Hooper was taken 
ill with a low fever, apparently resulting from overwork and a run-down 
condition of his system. While slowly recovering from this, a severe 
attack of pneumonia set in about the middle of August, and for several 
weeks his life was in danger. Although he is now improving satisfac- 
torily, his complete recovery will necessarily be slow, and some time must 
elapse before he can undertake any woik. His place in the department 
is being filled temporarily by Edward B. Durham, E.M., who graduated 
from the School of Mines in 1892. 

The Summer School of Mining was held this year in the Cripple 
Creek district of Colorado. On May 29th Adjunct Professor Peele and 
Mr. Hooper lefl New York, and on June 2d reached Victor, Colorado, 
about five miles from the town of Cripple Creek. Victor was selected as 
headquarters for the session, on account of its convenience to the mines 
visited. The total number of students was 26. As four members of the 
class had already visited the mining regions, either for study or practical 
work, they were permitted, by special arrangement, to make independent 
trips. Messrs. Lunt, F. H. Morley and Lidgerwood spent several weeks 
in Aspen and Leadviile, and W. R. Morley did his work at the Mary 
Murphy Mine. The regular session at Victor was attended by Messrs. 
Adams, Armstrong, Carrington, Cary, Colman, Cornell, Emerson, Eyer, 
Flint, Goode, Goodman, Mills, Roberts, Rubidge, Schimper, Stanley, 
Stewart, Thayer and Yung, of the fourth year, Bateson, of the third 
year, and Smith and Stroffregen, entering the course from Cornell and 
Princeton respectively. 

Three weeks and two days were devoted to surface and underground 
detail work at the Vindicator and Lillie Mines, on Bull Hill. In caring 
for the class Messrs. Peele and Hooper were assisted by Mr. Chas. H. 
Fulton, who had special charge of the underground surveying. Mr, 
Fulton, a graduate of the School of Mines, has been during the past year 
Instructor in Mining and Metallurgy at the University of Wyoming. In 
the Vindicator mine, which is one of the large and successful properties 
in the district, the officers allowed us the utmost freedom in carying on 
our work. The mine is being operated through half a dozen shafts, 
ranging from 300 to 900 feet in depth. We were also given valuable 
privileges in the adjoining Lillie mine, whose 1,135-foot shaft in one of 
the deepest in the district. During the last week of the session short 
visits were made to the Victor, Wild Horse, Independence, Portland and 
Granite Mines. We were unable to go underground either at the Inde- 
pendence or in the main workings of the Portland, but excellent ex- 
amples of wide stopes and square-set timbering were found in the Granite 
mine. On the Fourth of July a number of members of the class visited 

VOL. ZZXI.— 8. 


hiformally the interesting Doctor mine, which latterly has been produc- 
ing extremely rich ore. The surface plants of one or two other mines 
were also seen. 

As a large part of the class was due at Litchfield on July 21st, for sur- 
veying, the time available this year for mining was shortened by several 
days, to allow all to take field geology and a portion of the metallurg- 
ical work. Three days were therefore devoted under Dr. Bentley, of the 
Department of Metallurgy, to studying the Gillett and Economic Gold 
Extraction Company's chlorination mills. The latter has been recently 
built and is a good example of modern plant. Dr. Hollick, of the De- 
partment of Geology, took charge of the class on July nth, for one 
week's field geology. Afterward a part of the class, comprising those 
who had completed their Litchfield surveying, continued the metallurgical 
work at Florence, Pueblo and Colorado City. The required work of the 
Summer School being finished several members of the class made ex- 
tended trips to other mining regions. 

Our sincere thanks are due to Messrs. John S. Gary and J. Wilson 
Gary, of Denver, for their kind assistance and influence in arranging for 
the summer work; to Messrs. F. J. Campbell, Secretary, A. F. Holman, 
Superintendent, and the foremen of the Vindicator mine, for their cordi- 
ality and liberality during our long stay at their property; and to Messrs. 
W. F. De Camp, Superintendent of the Lillie mine, F. M. Woods and 
other officers of the Wild Horse mine, and Lee Wood, Manager, 
and the superintendents, of the Victor and Granite mines. 

The Cripple Creek district is an interesting one, both geologically and 
in the methods of mining employed, and the class was given abundant 
opportunity for efficient study. 

RoBiRT Peelb. 


The staff of the Department of Civil Engineering remains unchanged in 
its personnel with the exception that Mr. L. Le Count, C.E., has taken 
the position of assistant in consequence of the resignation last June of Mr. 
L. McHarg, C.E. The laboratory equipment, however, has been consid- 
erably increased and actively employed even during a considerable por- 
tion of the summer vacatiori. Through the generosity of the Fairbanks 
Scales Company the cement-testing laboratory has had its plant materi- 
ally increased by the gift of the latest pattern of the Fairbanks cement - 
testing machine, and the complete remodeling of the older machine made 
by the same company. The laboratory is now fitted to make the widest 
range of examinations in cement, mortar and concrete ; and work of that 
kind is now in progress the greater part of the time. 

The same general observations apply to the road materials testing lab- 
ratory. A very large amount of this class of work has been performed 
during the last six or eight months for Dr. F. J. H. Merrill, State Geol- 
ogist, and for the Honorable E. A. Bond, State Engineer and Surveyor, 
as well as for a number of private parties. It is the desire of the manage 
ment of the Department to extend the use of all its laboratory equipments 
so as. to be in close touch with the practice of the civil engineer's profes- 
sion throughout the Country at as many points as possible. 


The special eflforts which have been made by the department to develop 
in an efficient and broad manner its work in the railroad field are sus- 
tained in every possible direction. Its railroad subjects are being con- 
stantly advanced so as to include the broader fields of railroad adminis- 
tration, as well as construction and maintenance. These considerations 
have had a very effective influence upon the shaping of the railroad field 
work of the summer school, held near Litchfield, Connecticut. The last 
session of the Summer School of Surveying closed in September and has 
been most successful in every respect, not only in the instruction work but 
in numbers, which have been greater than ever before. 

Owing to causes entirely beyond the control of either the Department 
or the University the hydraulic laboratory of ihis Department is not yet 
completed, although very considerable progiess toward its completion has 
been made during the past two months. It is. hoped that it may be in 
condition to be leed by the present fourth-year class. 

W. H. B. 


Dr. Joseph Struthers resigned the tutorship in metallurgy on January 
I, 1900, and was succeeded by Mr. W. A. Bcntley, S.B. (Columbia), at 
first as lecturer and later as instructor in metallurgy. 

The Summer School was held in two sections, the first of which, for 
students in metallurgy and mining engineering, studied ths metallurgy of 
lead and silver in Colorado ; the second, for students in metallurgy and 
in mechanical engineermg, studied the metallurgy of iron and steel in 
Pittsburgh and Johnstown, Pa. 

The metallurgical laboratory received from Messrs. Fraser & Chalmers 
a cupola furnace for smelting copper ores, and a large English cupel fur- 
nace, and from the Wilbraham Baker Blower Company a Green blower. 
These important gifts will enable the department to widen the labora- 
tory instruction very materially. 

The Department has purchased many new and important pieces of 
apparatus, including many special furnaces, a Roberts- Austen autographic 
Le Chatelier pyrometer, a portable Le Chatelier pyrometer, and Le 
Chatelier's new microscope for opaque objects, with monochromatic 
illumination by mercury arcs. 

The metallurgical laboratory, as in previous years, has been used 
chiefly by advanced students ; but a new course, metallurgy 6A, of 
laboratory instruction for students in mining engineering has been organ- 
ized and will be given during the coming winter. Students in metallurgy 
continue to take the longer laboratory Course 6. 


In the School of Architecture, the year has opened with an entering 
class more than half as large again as last year's. It is apparently of ex- 
cellent quality, containing, as it does, a number of graduates of col- 
leges and technical schools. Several of last year's special students have 
passed their examinations and joined the regular classes. The work has 


started with promptness and enthusiasm, and is profiting by the changes 
lately made in the curriculum, especially by the concentration in the 
third year of all the work in speci^cations and building materials, and 
the assignment of the whole of Thursday, every week, to alternate ex- 
ercises in office practice and in engineering design. 

Each of the four years of the course has now come to have a char- 
acter of its own. The first year is devoted to the more elementary 
branches of design, drawing and mathematics. In the second year the 
drawing and design take on a more practical and professional character, 
and the mathematics gives place to theoretical mechanics. The third year 
is distinctively technical, the practical work just mentioned being accom- 
panied by the study of architectural engineering. Historical studies, 
with stated work in historical research and historical design, run through 
these years, with weekly practice in writing upon architectural subjects. 
These culminate in the fourth year in a series of papers of considerable 
importance which are read to the class and which afford its members a 
valuable experience both in investigations and in putting into shape the 
results of their investigations. This is the principal class work of the 
fourth year, mainly occupying the evenings, so that the days are left free 
for the study of problems in architectural design of a somewhat advanced 
character. Those who elect to do so substitute for this work advanced 
work in architecture engineering. 

Fifteen students graduated in June, as against twelve in 1899. 

The Columbia Fellowship in Architecture was awarded to Theodore 
Blondel, Jr., of the class of 1899, by a jury nominated by the Municipal 
Art Society. The subject of the competition was **A Monumental 
Fountain in Central Park.'' Honorable mention was awarded to Bayard 
S. Cairns, of the class of 1899, and William E. Parsons, of the class of 

Messrs. Balb, Cook and Willard have presented to the school a model 
of Mr. Andrew Carnegie's house, and Messrs. Bowrig and Tilton a large 
drawing of the United States Hospital at Ellis Island. 

A paper on **The Artistic Designing of Tall Buildings," written in 
French by Professor Hamlin, was read before the International Congress 
of Architects at Paris in August, by G. O. Totten, Jr., a graduate of this 
school in 1 891, 'delegate to the Congress. It will be published in the 
Compte Rendu of the proceedings. 


The Summer School in practical geodesy carried on its work at Oster- 
ville, Massachusetts. The class was in charge of Professor Jacoby, as- 
sisted by Dr. S. A. Mitchell, J. R. Wemlinger and L. Le Count. The 
work done this summer covered the usual ground, but improvements were 
made in the part of the course relating to base measurements. In the 
first place, a standard 100 base meter was laid down and monumented 
permanently in the spring by Professor Jacoby and Charles Derleth. C.E., 
who visited Osterville for that purpose. The base proper was then meas- 
ured by the students with a steel tape, standardized upon the above- 
mentioned loo-meter standard base. The new arrangement proved very 


Daring the smniiier Pr o fe aMr Recs was at the Paris ExpositioD, from 
June 7th to August 2d, as United States juror in the department having 
charge of instiuments of precisioo He had a most excellent opportunity 
for examining the latest types of astronomical and phy-sical apparatus. 
Professor Rees is very enthusiastic in his praises of the German exhibits. 
He states that the German instruments were so much superior to those 
exhibited by any other nation that the juror for Great Britain, Profes- 
sor Boys, of London, proposed to the jury that Germany should receive 
a prize which should recognize this fact. Professor Boys's motion was 
seconded by the United States juror, but the rules relating to awards did 
not allow of any such prize. The German Government published a vol- 
ume of 250 pages, illustrating and describing all the instruments of pre- 
cision exlubited by the German makers. Catalogues were printed in the 
French, English and German languages, so that the jurors could read the 
descriptions in the language which b^t suited them. 

Professor Rees also was United States delegate to the Congress on 
Chronometry, which met at the Observatory at Paris, and he attended by 
special invitation the Astronomical Conference which met to discuss the 
work now going on with reference to cataloguing and charting the stars 
by photography. This Conference also discussed the way-s and means for 
making observations during the coming winter of the planet Eros, for the 
purpose of obtaining with the greatest accuracy the value of Solar Parallax. 

Unfortunately, as Columbia has not provided the astronomical depart- 
ment with the means for mounring the Rutherfurd photographic telescope, 
it was impossible for the Columbia Observatory to offer to assist in this 
scheme so far as taking photographs was concerned, but the Director 
offered to measure and reduce photographs that might be taken by other 
observatories, especially in the United States. In this connection, it may 
be properly stated that the Lick Observatory has requested the astronom- 
ical department to examine certain photographs taken with the Crossley 
reflector (mirror three feet in diameter), for the purpose of determining 
whether that ins: rument can be used with accuracy in the above-men- 
tioned work. 

Professor Rees spent a month in England,' and visited the observa- 
tories at Greenwich, Cambridge and Oxford. 


As soon as examinations closed in May, Professor Kemp conducted a 
Summer School of Geology for several days in the vicinity of Peekskill. 
The class studied and mapped a large part of the eruptive area called 
the Cortland series among geologists. After this work, Mr. Van Ingen 
took charge of the students for a study of the strata neir Rondout. An 
excellent opportunity was afforded to observe folds, faults, unconform- 
ities and other structural phenomena. The Department has already re- 
ceived two reports based upon this work which are very satisfactory. 
The officers are growing to realize more and more the importance of field 
experience to the mining engineer, and are, therefore, endeavoring to 
provide opportunities for observation of structural phenomena both on 
the Saturday excursions and during the Summer School. 


Iq the ea ly part of July, Dr. Hollick conducted an additional Sum- 
mer School for those students of Mining Engineering who had partici- 
pated in the Summer School of Mining at Cripple Creek, Colorado. Dr. 
Hollick made his headquarters at Canyon City and the work was per- 
formed in the foothills of the Rocky Mountains near that place. The 
field proved to be a most instructive one and excellent results were at- 
tained. During the last week in June, The American Association for 
the Advancement of Science was in session at Columbia aod Professor 
Kemp delivered the Vice-presidential address before the Section of Geol- 
ogy. His subject was *• Pre- Cambrian Sediments in the Adirondacks." 
In the address he gave a description of the most recent results which 
have been reached in the study of the metamorphic rocks in these moun- 

Professor Kemp passed the summer in British Columbia, and expects 
to prepare later on a scientific paper upon the basis of his observation 
there Dr. Julien was busied with the geology of Cape Cod during the 
vacation, and has already presented a short account of his observations 
before the New York Academy of Sciences. Mr. Van Ingen performed 
some field work in northern New Jersey and in the vicinity of Rochester. 
During the summer the Geological Survey of Louisiana issued Special Re- 
port No. 5, on ** A Collection of Fossil Plants from Northwestern Louis- 
iana," by Dr. Hollick, illustrated with sixteen plates. In the forthcom- 
ing annual report of the Geological Survey of New Jersey, now in press, 
is another report from Dr. Hollick on ** The Relation between Forestry 
and Geology in New Jersey,'* illustrated by a map, showing forestral dis- 
tribution in the State. 

Dr. Hollick was traveling in the West during the vacation, visidug the 
Grand Cafton of the Colorado, the Yosemitc Valley and other points of 

\ Jena 

Normal #^ 

^j^g^w/ Glass. \^ 

C. A. F. Kahlbaum's Chas- Schleicher & 

ChBHilcally Pure Oroanlc j j SchUGll*S 

iBPrflanlc Chamloali. | C. P. FILTERPAPER 


ISth St and 3rd Avs.. New York 

Chemicals S Chemical Apparatus 


SCHOTT & GENOSSEN, Manufacturcis of the famous Jena Glass 

C. A. F. KAHLBAUM, Cherokals. 


CARL ZEISS, Microscopes. 

FRANZ SCHMIDT & HAENSCH, Polariscopcs and Spcctro3COi>es. 

DESMOUTIS & CO., Purest Haramereri Platinum. 


GREINER & FRIEDRtCH'S, Fine Stopcock Ware. 

MAX KOHL, Physical Apparatus. 

TENDER, Reagent Bottles with Black Indelible Letters. 

KAVALIER'S Famous Bohemian Glassware. 

Tlie ibove Grms tie known to furnish the best goods in their lespeclive lines 
We are their representatives for the United StXel and Canada and carry a consider 
able slock of these goods on hand. Kohl'* Apparatus and Stender's Botilei are 
nporled to order onlj. 

Jena Gla»B tha Glass of the Fntnre 

Neumann Bros. 

Book Binders 

Sand7EA8TieTH ST., 



Books bound In any style. 
Single volumes or in quantities. 

Woodbridge School, 


School of Mines Preparatory School 

417 Madison Avenue, 

Between 48th and 49th Streets, NEW YORK CITY 

Twentieth Year Begins October ist, igoi. 

THE school is well equipped with physical and chemical laboratories, in which the 
students are required to perform a complete set of experiments illustrative of their 

recitations in physics and general chemistry. A special laboratory is devoted to 
qualitative chemical analysis for advanced students. 

Five hundred Students of Columbia School of Mines have been instructed in the 
Woodbridge School. Also a large number have been prepared for Massachusetts In- 
stitute of Technology, Stevens Institute, Sheffield Scientific School, Lawrence Scien- 
ific School, Troy Polytechnic Institute, Cornell University, and the Classical, Medical, 
and Law Departments of Harvard, Yale, Columbia and Princeton. 

A summer sch«x)l for students who have failed in June begins August 13th and 
coaches men for the Fall examinations. All classes are limited to five. College men 
are coached in Freshman and Sophomore Mathematics and Quantitative Analysis. 

An advanced course for older students prepares them to enter the second year class 
Last June one of our students received his degree in Electrical Engineering in 
three years. 

d A^ / 5" ^0' 1 1 

Vol. XXII. No. 2. JANUARY. 1901. 






A. J. MOSB8, Prof, of Mineralogy. JOB. 8TRUTHBRB. Ph.D., 

J. P. KEMP, Prof, of Geology. Lecturer in MeUUurgy. 

R. PEELB, Adj. Prof. Mining. I. H. W00L80N. Instructor In 

A. D. P. HAMLIN, Mechenical Bnglncering. 

Adj. Prof, of Architecture. W. H. PREEDMAN. E.E., 

R. B. MAYER, Inetn^or in Drawing. B. O. MILLER, AssitUnt in 

B. AVALLER, Annlsrticel Chemiet. Mechanical Engineering. 

ManMTlnff Editor, R. B. MAYER, . 
BuBln«BB ManMr^r, 8. O. MILLER. 


Blectrochemistry and Blectrometallurgy. By Francis B.Crocker 119 

The Analysis of Slags and Cinders. By Cavalier H. Joiiet 140 

Notes on the Assay of the Zincy Precipitates obtained in the Cyanide 

Process. By Chas. H. Fulton and Chas. H. Crawford 153 

The Powering of Ships. By William Ledyard Cathcart 163 

A Method of Cychc Analysis of Heat Engines. By Charles E. Lucke, 

M.S 223 

Book Reviews 253 



Registered at the New York Post Office as Second Class Matter. 

AU Remittances should be made payable to Order of "The School of Mines Quarterly " 



Ill* well-knawi 

Tool and Di« Steel of the Best QuaUty, Sheet Steel, 
Spring and Macbinery Steel. Steel for Bdge-Toole and 
Hardware of Every Description. Poliahed Drill Rods 
and Fine Wire. Steel Forgings, Called Sptinga, etc. 

WntErn OBcc and Watetaoa*e, SasUro OOea mnd Warehou 

M and 66 Bo. OlinMn Btreet, ISO Faarl Bti«^ 


16,000,000 BARRELS 






E. R aCKERMAN, PMia . Aeaoe AM Sac. C. E. 

Compare Weights. 


Anrigt W<<gM 1-B Inch "lanklnt >9B," II Ifet. et 
I ha iqum ytnt. 

Anriga Walghl 1-B Inch Rid Packing. 14 Iba. I* 
Itia tquarc yard. 

Al BOe. par poind. "1ENKINS"SB" la n»t onlj 
nrymnsk ohaipar, bat tka batt Jelnt pack- 
ing mniilacturad. 

JENKINS BROS., NewYork.Phlladelphia.ChfCBgo, Boston 



Vol. XXII. JANUARY, 190 1. No. 2. 



Part I. Theory. 

Introductory. — Electrochemi.^^try may be defined as that branch 
of science relating to the electrical production of chemical sub- 
stances and chemical action or vice versa, to the generation of 
electrical energy by chemical action. 

Electrometallurgy is the branch of science that relates to the 
electrical production and treatment of metals. 

The two subjects are based upon the same principles, the theory, 
laws and data of one being equally applicable to the other. Hence, 
these principles as given in Part I. are common to both branches, 
no distinction being necessary unless specially noted, in fact it is 
not possible to draw any sharp line of separation between them. 

Electrochemistry may be subdivided as follows : 

I. Electrolytic Chemistry , which consists in separating or 
producing other action upon chemical substances by the decom- 
posing effect of an electric current. Since the electrolyte is usually 
in the liquid state, we have : 

*• Wet methods " with solutions, 

" Dry methods " "wxyHa fused materials. 

Electrolytic chemistry is applied to the following purposes : 

{d) Electrolytic production of chemicals, as, for example, caustic 
soda and chlorine from a solution of common salt. 

vouxxii— 9. 119 


(p) Electrolytic refining of chemicals by the elimination of im- 

[c) Electrolytic cliemical effects, such as bleaching, tanning, etc. 

{d) Electrolytic chemical analysis, as in copper determination. 

(/) Primary batteries including various forms of voltaic cell in 
which electrical energy is generated by chemical action. 

(/) Secondary batteries are quite similar to the foregoing, but in 
this case the chemical action must be reversible so that after pe- 
riods of working the cell may be charged or brought back to an 
active condition by sending through it a current opposite in direc- 
tion to that which it generates. 

2. Electrothermal Chemistry includes those methods in which 
electric current raises the temperature of materials, usually to a 
high degree, in order to produce chemical action, but the effects 
are not due to electrolysis, alternating currents being generally 

Cliemical action with electrical heating, as for example, the produc- 
tion of calcium carbide from lime and carbon in an electric furnace. 

Electrical Fusion of Chemicals is similar to the preceding 
method, being accomplished by heating in an electric furnace, but 
is merely a physical process since it involves no chemical action, 
as, for example, the fusion of some refractory substance such as 
silica or alumina. It has been proposed to make brick by actu- 
ally melting instead of baking the clay. This would naturally be 
included under the head of applied chemistry. 

Electrometallurgy may be subdivided in substantially the same 
manner as electrochemistry ; in fact it could be included directly 
with the latter. The distinction is mainly nominal, those methods 
in which metals play the prominent part being called metallurgical. 
On the other hand primary and secondary batteries are put under 
the head of electrochemistry, although their active plates usually 
consist of metal. Since it is customary, however, to consider the 
two sciences separately, the following classification of electrometal- 
lurgy is given. 

I. Electroljrtic Metallurgy, which consists in separating or 
producing other actions upon metals by the decomposing effect of 
an electric current. Since the electrolyte is usually in the liquid 
state, we have : 


'• Wet methods " with solutions, 

** Dry methods " with /used materials. 

Electrolytic metallurgy is applied to the following purposes : 

{a) Electrolytic production of metals^ being the art of obtaining 
them from their compounds by electrically decomposing the latter 
in a state of solution or fusion, as, for example, the Hall process 
of extracting aluminum from alumina dissolved in fused cryolite. 

(^) Electrolytic refining of metals which consists in eliminating 
impurities by electrodepositing the metal itself, the foreign sub- 
stances being left behind in the anode or bath, or vice versa ; as 
for example, the well known electrolytic refining of copper from 
the crude metal. 

(r) Electroplatings which is the art of coating articles with an 
adherent layer of metal by electrodeposition, as in nickel plating. 

(rf) Electrotyping, which is the art of reproducing the form of 
type, medals and other objects by electrodepositing metal on the 
object itself or on a mould obtained from it. 

2. Electrical Smelting, which consists in reducing metallic com- 
pounds by chemical action at a high temperature produced by an 
electric current, as, for example, the Cowles process for the produc- 
tion in an electric furnace of aluminum bronze from a mixture of 
alumina, carbon and metallic copper. 

3. Electrical Heating and Working of Metals, which consists 

in treating metals mechanically with the aid of heat generated by 
electric currents, as in electrical welding, forging, rolling, casting, 
tempering, etc. 

Historical Notes. — Immediately following the invention of the 
primary battery by Volta in 1800, Nicholson and Carlisle dis- 
covered the chemical action of the electric current in decomposing 

In 1807 Sir Humphrey Davy gave his famous lecture " On Some 
Chemical Agencies of Electricity," he having, the same year, dis- 
covered the metals sodium and potassium by reducing their com- 
pounds electrolytically. 

In 1834 Faraday* established definite laws and nomenclature 
for electrochemistry. From 1836 to 1839 Jacobi, Spencer, Jordan 
and Elkington applied these principles to practical use in the mak- 

* « Experimental Retearchss," Vol. I., pp. 195-35S. 


ing of electrotypes. Since that time the theory as well as appli- 
cations of electrochemistry and electrometallurgy have made steady 
progress in electrotyping, electroplating, in the production, refining 
and working of chemicals and metals and in the development of pri- 
mary and secondary batteries. The advance, however, has not been 
as great or as rapid as it promised to be, and several other branches 
of electrical engineering, such as telegraphy, telephony, electric 
lighting and electric power, have all far outstripped electrochemis- 
try and electrometallurgy in the extent and value of their results, 
in spite of the fact that the two latter were the first practical appli- 
cations of electricity, and three out of the four other applications 
have been developed during the last twenty years. But, in the 
opinion of those most competent to judge, electrochemistry and 
electrometallurgy are certainly destined to become of enormous im- 
portance in the near future. 

The Physical and Chemical Principles, as well as the quanti- 
tative relations of electrochemistry and electrometallurgy, are more 
exact than those of almost any other branch of science, since the 
two most important electrical units — volt and ampere — are both 
directly based upon electrochemistry, the former being defined in 
terms of the E.M.F. of a Clark cell, and the latter being defined as 
the current which deposits .cx)iii8 gram of silver per second 
These definitions were selected by the International Electrical Con- 
gress, Chicago. 1893, as being the most reliable, and have since 
been legalized by nearly all civilized countries, which goes to show 
the accuracy and certainty of electro* hemical science. 

Nevertheless, knowledge concerning these sciences is not general 
and most electrical engineers are not very familiar with their theory 
or even practice, while they may be well informed concerning the 
mechanical side of their profession. A similar statement is ap- 
plicable to most chemists, familiarity with chemistry is far 
out of proportion to their knowledge of electrochemistry. 

The laws and data of electric heating, upon which the other 
branches of these sciences are based, are also very definite, and 
these will be taken up and discu^sed after those of electrolysis. 

It is assumed that the reader has a general acquaintance with 
electrical terms, units and principles, and, if not, he is referred to 
some elementary work from which he should obtain this knowledge 
before attempting to study any special branch of electrical science. 


Otherwise it would be necessary to follow the too common practice 
of devoting the first half of each article or book to these same 
general principles. 

Tlie principles of electrolysis are based upon Faraday's laws which 
may be combined and stated as follows : 

The amount cf chemical action produced {for example^ the weight 
of metal deposited) by an electric current^ is directly proportional to 
the strength of the current^ to the chemical equwalent of the material 
acted upon and to the time during which the current flows. To 
explain this law physically, modern theory assumes that an 
electrolyte is composed of two sets of particles some of which are 
tree and are called ions^ one set carries positive and the other nega- 
tive electric charges. The positive ions, which include all of the 
metals as well as hydrogen and a few other elements, move or 
** migrate " toward the negative electrode and there neutralize or 
deposit their charges. The negative ions which include oxygen, 
chlorine, bromine, iodine and the acid radicals migrate toward the 
positive electrode and there neutralize or deposit their charges. 

These ions on reaching the electrodes are liberated as solids 
liquids or gases or produce secondary reactions according to their 
ordinary physical and chemical properties. It is further assumed 
that the electric charge is equal for all monad ions, half as much 
for dyads, and so on, hence the same quantity of electricity trans- 
ferred, that is, the same current for the same time will always re- 
quire the deposition of the same total valency of ions on each 
electrode. It follows that the weight deposited is proportional to 
the weight divided by the valency ions which is the same as their 
chemical equivalent, thus agreeing with Faraday's law. 

This law may be put in the form of a simple expression. Call. 

ing C the current in amperes, / the time in seconds, w the weight 

in grams of the given material and Z its chemical equivalent, we 


w:CZt, (i) 

This establishes a fundamental and definite relation between 
electrical and chemical science. The relation is made still more 
direct by calling the weight in grams of each element liberated by 
one ampere in one second (/. e. one coulomb) its electrochemical 
equivalent. Hence designating this quantity as z^ we have 

w 2= Czt, (2) 


With this expression, calculations of the weight of metal de- 
posited by a given current and other similar problems, may be 
easily solved. 

One ampere deposits .001 118 gram of silver per second, in fact 
it is thus that the ampere is legally defined, as already started. 
The chemical equivalent of silver is 107.11, the same as its atomic 
weight since it is a univalent element. By Faraday's law the 
weight of any other material liberated will be proportional to its 
chemical equivalent, to the current and to the time, hence 

.001 1 18 za 

107. 1 1 ^^ ^^' 

The chemical equivalent of hydrogen being unity, the weight 
of hydrogen set free by one ampere in one second would be 
.00001044 gram; therefore by definition this constant is the elec- 
trochemical equivalent of hydrogen. Its product with the chemical 
equivalent of any element is obviously the electrochemical equiva- 
lent of that element. 

The foregoing expressions involve awkward decimal fractions 
and the second as a unit of time. It is more convenient in most 
cases to consider the weight liberated in one hour, so multiplying 
the constant in (3) by 3600 we have 

w = .0376 ZCT, (4) 

in which 7 is the time in hours, C7 being therefore ampere-hours. 
From (4) by transposition the current required to liberate a given 
weight in a given time is found to be : 

26.6 w 
^^- ZT~' (5) 

\{ E is the E.M.F. employed the power consumed in watts is : 

26.6 wE 
EC^ —2t~' (^^ 

If the weight is given in pounds W, we have, since there are 453.6 
grams in one pound : 

W^ .0000829 zcz (7) 

C. if^, (8) 


12065. WE 
EC^ jf-' (9) 

In these expressions, all of the quantities are easily and defi- 
nitely determined with the exception of the equivalent Z, To the 
electrician this is one of the most puzzling points in electrochem- 
istry. The chemical equivalent Z of any element is the weight 
which is equivalent to one gram of hydrogen in chemical combi- 
nations, hence : 

... . , ^ atomic weight , . 

chemical equivalent = _ ^^ J^^ — . (lo) 

effective valency ^ ' 

For example, 35.18 grams of chlorine combine with one gram of 
hydrogen to form hydrochloric acid (HCI), and since one atom of 
chlorine combines with one of hydrogen, the valency or ratio of 
the number of atoms is one. Hence both the atomic weight and 
the chemical equivalent of chlorine are 35.18. But it requires two 
atoms of chlorine to satisfy one of zinc and form zinc chloride 
(ZnCI,), that is 2 x 35.18 grams of chlorine combine with 64.9 
grams of zinc; therefore the atomic weight of zinc is 64.9, its 
chemical equivalent is 64.9 -*- 2 = 32.45 because its valency or com- 
bining value of each atom is 2. In the case of iron there are two 
compounds with chlorine : FeCI, and FeCl,. The former, or fer- 
rous chloride, is analogous to zinc chloride and the chemical equiv- 
alent is 27.8, being one-half of 55.6, the atomic weight. But in the 
latter, or ferric chloride, the number of chlorine atoms are three 
times those of iron, hence the effective valency is 3 and the chem- 
ical equivalent is 

55.6 -.3= 18.53. 

Strictly speaking, the latter compound is Fe,CI, since there are 
two atoms of iron in each molecule or group of atoms, and in this 
case the total valency of iron is 4, one •* bond " or unit of the val- 
ency being required to hold the two atoms together. The ratio of ' 
the number of atoms, however, is 3, which is the effective valency. 

Thus certain elements are monads, having valency = i and 

chem. equiv. = atomic weight, such as sodium, potassium, silver, 

chlorine, iodine and bromine; others, such as zinc and oxygen, 

J , , . , , , atomic weight, 
are dyads, having valency =» 2 and chem. equiv. = — 

while many others are triads, tetrads or have two or more different 
valencies, with corresponding chemical equivalents. 


This matter is of great importancejboth theoretically and practi- 
cally. P'or example, copper forms two chlorides, Cu,Cl, and CuCI, 
In the former the chemical equivalent = atomic weight = 63.1. 

In the latter the chemical equivalent =« — =31.55. 

Hence the actual weight of copper deposited would be twice as 
great when the former is electrolyzed, the current and time being 
the same. Similar facts apply to the case of other elements which 
have two or more different valencies. 

Ihe equivalents of compound substances are similar to those of ele- 
ments, being the proportions in which they combine. For ex- 
ample, the amount of the radical SO^ that tends to be set free when 
a solution of copper sulphate (CuSOJ is electrolyzed, would be 

Z= 31:^ +J4 X '5-88) ^ 47 68, the atomic weight of sulphur 

being 31.83, of oxygen 15.88, and the valency of SO^ is 2, since it 
would require 2 atoms of hydrogen to combine with it and form 
sulphuric acid (H,SOJ. In other words, there is one equivalent 
= 47.68 grams of SO^ for each equivalent = 31.55 grams of cop- 
per. As a matter of fact SO^ is not a staple compound and it 
would either combine with the anode to form a sulphate, or if 
that were carbon, and, therefore, not acted upon, the SO^ would 
break up into SO, + O, the former combining with the water of 
the solution to form H,SO^ and the latter being liberated as a gas. 

The chemical equivalent of caustic soda (NaOH) being 22.88 
+ 15.88 + I = 39.76, there would be 39.76 grams of that com- 
pound produced for 35.18 grams of chlorine and i gram of hydro- 
gen set free, when a solution of salt (NaCl) is subjected to electrol- 
ysis. The reaction is NaCl + H,0 = NaOH + CI + H. 

To find the weight of caustic soda produced by a certain current 
in a certain time, substitute in (3) the number of amperes and sec- 
onds for Cand / and the chemical equivalent 3976 for Z* 

The actual weight of deposit or amount of chemical action is some- 
times less than the theoretical value, /'. e.y that calculated by equa- 
tions (2) or (4). Indeed, some authorities even go so far as to deny 
the truth of Faraday's law. In the opinion of the writer the dis- 
crepancies can usually be explained, and are no greater and often 
less than the modifications which have to be made in all laws on 
account of the conditions in each practical case. 

The most serious cause which reduces the weight of metal de- 


posited is the setting free of hydrogen in place of the metal. This, 
however, is perfectly covered by Faraday's law, since the apparent 
loss of metal is accounted for by the amount of hydrogen. The 
liberation of hydrogen is likely to occur : First, if the current den- 
sity (amperes per square centimeter) is too great : second, if the 
solution is weak, and third, if the metal has a great affinity for 
oxygen and decomposes the water, thereby setting free the hydro- 
gen. In regard to current density it is found that a certain value 
gives the best results as to the character and weight of deposit, 
and if this rate is exceeded there is a tendency to evolve gas and 
produce a pulverulent deposit. If the solution is too weak a simi- 
lar eflfect is produced, since the compound becomes used up near 
the surface of the cathode and the current is forced to liberate 
hydrogen instead of the metal. In other words the velocity of 
migration of the ions is not sufficient to supply them rapidly 
enough at the electrode when the current density is excessive and 
the number of ions are inadequate when the solution is too dilute. 

Another reason for loss of metal in electrodeposition is local 
action^ by which it is dissolved after being deposited. If, for ex- 
ample, free acid is put in the bath to increase its conductivity, as 
is often done, this tends to dissolve some of the metal deposited, 
and the weight finally obtained maybe considerably less than that 
which the current actually produces. The tendency to local action 
is increased by lack of homogeneity in the deposit or bath, as, for 
example, dissimilar chemical composition in different parts of the 
deposit, whereby voltaic action is set up between different portions 
of the plate. 

The actual weight of caustic soda obtained by the electrolysis of 
salt solution may be less than that calculated in the manner de- 
scribed on the preceding page. The discrepancy, however, is 
usually due to the conversion of part of the caustic soda into sodium 
hypochlorite (NaClO) by the action of the chlorine which is aiso 
produced. This secondary reaction can be avoided by interposing 
a porous diaphragm to prevent the chlorine and caustic soda from 
mixing. It is evident that none of these cases shows any real viola- 
tion of Faraday's law ; they are merely due to Improper conditions 
which can usually be corrected. 

Hastening chemical action by electric currents sometimes produces 
peculiar results. If one cell of a battery is generating current in an 
electric curcuit, and another c£ll is added in series, then the current 


in amperes is doubled as well as the E.M.F., provided the internal 
resistance is insignificant; hence two cells produce four times the 
power of one. The explanation is that the addition of a second 
cell causes the first to work and consume zinc at twice the rate of 
a single cell and also adds its own power. 

This principle may be utilized to hasten the solution of a metal 
or other electrochemical action. Copper is not quickly dissolved 
by sulphuric acid, but, if it be used as the anode of an electrolytic 
cell containing the dilute acid, the solution can be effected with 
great rapidity by means of a large current, the weight dissolved 
being given by equation (4). The E.M.F. may be small, being 
only that required to overcome the resistance, since the combina- 
tion of the copper evolves more energy than the liberation of the 
hydrogen consumes, so that the E.M.F. of the cell would aid instead 
of oppose the current. 

General Relation Between the £lectrical Quantities. — The 

electrical quantities in a circuit in which electrolytic work is being 
done are related in a manner which may be expressed by the fol- 
lowing form of Ohm*s law, 

E^ e 

'^'-r-+r^+r: (") 

in which C is the current in amperes. E is the direct E.M.F. in 
volts of the dynamo or other source of current, e is the counter 
E.M.F. set up by the cell. R^ is the internal resistance in ohms ot 
the generator, R^ is the resistance of the conductors connecting 
the generator with the cell, and R^ is the resistance of the elec- 
trolytic cell itself. These quantities can be pre-determined or 
calculated even before the apparatus is made, or they may be 
actually measured by electrical instruments in the case of an exist- 
ing plant. 

The direct EM.R^ which must be produced by the source of cur- 
rent in any given case, is found by solving equation (ii) with re- 
spect to E, 

Er^e+ C{R^ + R, + R;). (12) 

If E be raised or lowered, the current will be increased or de- 
creased accordingly. In this way the strength of the current may 
be regulated, usually by varying the field strength of the dynamo 
by means of a rheostat. A variable resistance inserted in the 


main circuit would have a similar effect, but is wasteful of energy. 
In most cases it is simpler to consider the voltage F measured be- 
tween the terminals of the cell, in place of £ the total E.M.F. of 
the generator. This voltage is less than E by the amount of the 
drop in the generator and in the conductors; that is 

F= £•- C{R^ + R^ from which £= r+ C(R^ + /?,). (13) 

Substituting this value in (i i), we have 

By regarding the voltage Fas the important quantity, a com- 
mon cause of confusion is avoided. For example, an electrolytic 
cell is often supplied with current from a lighting circuit operating 
at no volts. Incandescent lamps or other forms of resistance are 
Usually put in series with the cell to prevent an excessive flow of 
current. The drop in this resistance may use up 105 volts so that 
the pressure Fat the cell is only 5 volts. The result is exactly 
the same as if a generator of 5 volts were "connected directly to the 
cell, the drop of 105 volts having no effect upon the latter. It is 
a common error, however, to suppose that the total 1 10 volts acts 
upon the cell or at least produces conditions different from those 
due to the direct application of low voltage. 

The resistance of the generator {R^") is the resistance in ohms of 
the armature of a shunt-wound dynamo and the sum of the re- 
sistances of the armature and series field magnet coils in the case 
of a series or compound wound machine. In a primary, second- 
ary or thermoelectric battery it is its internal resistance. The 
value of R^ is usually between one and five per cent, of the total 
resistance of the circuit, which latter is R^ + ^,+ R , 

It may be measured by some of the well-known methods, but 
ordinarily it is so low that the fall of potential or *' drop " method 
is most applicable.* The resistance of the armature may also be 
calculated if the size, length and arrangement of the armature 
winding be known or can be ascertained by examination. 

The resistance of brush contacts is often a large factor in elec- 
trochemical and metallurgical dynamos, since the currents are usu- 
ally very large. Assuming the brush contact to be .001 ohm 

**< Practical Managemeit of Dynamos and Motors," by Crocker and Wheeler, 
page 90. 


and the current amperes, the loss of potential is one volt, 
and of energy i,ocx) watts, or one and one-third horse-power, 
which losses are excessive unless the machine generates at least 
I oo volts, that is lOO kw. which is equal to 1 34 h.p. Large brushes 
made of copper wire gauze may be used to make this resistance 
as low as possible. 

Tlu resistance of the line or conductors connecting the cell with 
the generator may also be measured by the " drop " method, or it 
may be calculated by the formula 

^ Lr , . 

in which L is the length of the wire in feet, d its diameter in 
mils (thousandths of an inch), and r is the resistance of one mil- 
foot (a wire one foot long and one mil in diameter). The value of 
r for copper, according to Matthiessen's standard adopted by the 
American Institute of Electrical Engineers {^Transactions, Vol. X,), 
is 10.35 international ohms at 20^ C, 11.57 ohms at 50° C, and 
1 2.82 ohms at 80° C. 

The measurements are expressed in feet and inches, since these 
are almost universally employed in England and America, the 
metric system being rarely used for electric wires or conductors of 
any kind. 

To determine the resistance at other temperatures add to or sub- 
tract from the resistance at 20^ which is 10.35 ohms per mil-foot, 
four tenths of one per cent, for each degree above or below 20^ » 
for example, add two per cent, for 25^ and subtract two per cent, 
for 15°. 

This result agrees with Matthiessen*s formula within a small 
fraction of one per cent, for temperatures between o^ and 100^ 
C, and is far simpler; it is also probably of equal if not greater 
accuracy, since the latest results show a linear relationship between 
the resistance and temperature of pure copper. 

T/ie cmrent capacity or maximum current density that it is allow- 
able for a conductor to carry, varies from about one ampere per 
500 circular mils for copper wires one tenth inch in diameter of 
thereabouts, to 1 ,000 amperes per square inch for large copper bars ; 

that is, C= . or C= 1,000 A respectively, d being diameter in 

mils and A being area in square inches. The much lower current 


density in the latter case is due to (lie relatively smaller cooling 
surface of large conductors. 

Example : What is the resistance of and drop in 1,500 feet of 
No. 0000 (A. W. G.) copper wire at 30° C. carrying 105 amperes. 
The diameter obtained by measurement or from a wire table is 
46 inch or 460 mils. The resistance per mil-foot r being 10.35 
ohms at 20°, is 10 x. 004 = .04 higher at 30^; that is, 1.04 
X 10.35 == 10.76 ohms. Substituting these values in (15) 

Rf^ 1 500 X 10.76 -H 460^ = .0763 ohm. 

With a current of 105 the drop is 105 x .0763 ■■ 8 volts, and 
for a circuit of two wires each 1,500 feet long the resistance and 
drop are twice as great. This current would not heat the wire 
more than about 5°, being only one-half that allowed for rubber- 
covered wires of this size by the National Electrical Code and one- 
third the limit for weather-proof insulation, hence the temperature 
of the air must be 25°, since that of the wire was assumed to be 
30® C In this case the area of the wire is 460* =b 21 1,600 circu- 
lar mils, giving 211,600-*- 105 = 2,015 circular mils per ampere, 
which also shows that there is ample current capacity. 

The resistance of the cell or bath may be calculated by the ex- 

in which D is the distance between the electrodes in centimeters, 
S is the area of each in square centimeters, and p is the specific 
resistance of the liquid; that is, the resistance between the oppo 
site surfaces of a cubic centimeter of it. The measurements can 
also be expressed in inches and square inches, provided p is the 
resistance of a cubic inch, being .3937 as great. This formula 
assumes that the plates are perfectly parallel, and that the current 
flows directly from one to the other without spreading, and with- 
out any action on the sides or back of the plates. If the plates 
are not parallel, D should be the average distance between them. 
If the current spreads, owing to the fact that the cross-section 
of the bath is larger than the area of the plates, then we may 
assume 5 to be a mean between the two; but this is only an 
approximation, particularly if there is considerable difference be- 
tween the two areas, and the distance between the plates is also 


comparatively large. If the action may take place from the back 
and sides of the plate, then the exact resistance is usually very 
difficult to calculate; but with plates placed close together, not 
more than one-half inch apart, and being one foot or more on a 
side, or in that proportion, the resistance may be calculated by 
equation (i6) as being that of the liquid actually between the 
plates, since comparatively little current would flow outside of this 
volume, even though the bath extends somewhat beyond the edges 
of the plates, these conditions being very common. If a plate is 
placed midway between two others and the current flows in both 
directions from the middle plate, then the resistance on either side 
is found by equation (i6), and the combined resistance is one-half 
of that. In the same way, with two anodes arranged alternately 
between three cathodes, or vice versa, there are four compartments, 
and the combined resistance would be one-quarter of the resistance 
of each, and so on, for any number, if they are equal in dimensions. 

The specific resista?ice of electrolytes varies greatly with the chem- 
ical composition, the strength of the solution, and the tempera- 

As a general rule, acid and alkalies are better conductors than 
neutral solutions ; partly diluted solutions of acids or salts usually 
conduct better than concentrated solutions ; solutions of chlorides, 
bromides or iodides are better conductors than those of sulphates, 
nitrates and carbonates ; and fused salts conduct better than the 
corresponding solutions. 

Sulphuric acid, diluted with water to the density 1.25, has a 
specific resistance of .624 ohm per cubic centimeter at 16° C. ; 
this is about the minimum resistance not only of sulphuric acid, 
but of almost any solution at that temperature. 

A saturated solution of zinc sulphate at 10° C. (density =s 
1.422) has a specific resistance of 33.7 ohms ; and about half satu- 
rated (density = 1.27), the specific resistance is only 28.5, which is 
about the minimum value. A saturated solution of copper sul- 
phate at 10° C. (density = 1.205) has a specific resistance of 29.3, 
which, in the case of this salt, is the minimum. 

Temperature Coefficient of Electrolytes. — The resistance of solu- 
tions, as with all electrolytes, decreases rapidly with increase of 
temperature ; in this respect conductors of this class are similar to 
carbon and exactly opposite to the metals. The specific resistance 
of dilute sulphuric acid (density = 1.25), which was stated above 


to be .624 at 16° is 1.31 at oP and .358 at 28^ C. This corre- 
sponds to a negative temperature coefficient of about 3 per cent., 
which is nearly eight times as great as the positive temperature 
coefficient of copper. 

The conductivity of non- aqueous solutions, such as those of alcohol, 
ether, etc., have been investigated by W. Hampe* who stPtes that 
alcoholic solutions of cupric chloride, and of zinc or cadmium bro- 
mide or iodide, also zinc chloride or bromide dissolved in ether 
are all good conductors. Gore states as the result of his investi- 
gationsf that " alcoholic and ethereal solutions of metallic salts are 
much less easily electrolyzed than aqueous solutions of the same 
salts." The objects which might be attained by the use of such 
solutions are the lessening of the tendency to oxidation and sec- 
ondary reactions which are likely to occur in the presence of water 
as, for example, in the deposition of aluminum, magnesium, etc. 

Fused electrolytes are usually good conductors, for example, fused 
lead chloride has a conductivity almost five times as great as that 
of the maximum value for sulphuric acid. This figure is stated by 
F. Braun J who also gives the conductivity of various other fused 
salts. Gore, in his work on " The Electrolytic Separation of 
Metals" (London, 1890, p. 91), gives a long list of melted com- 
pounds which he has electrolyzed. W. Hampe§ also enumerates 
many salts which are " good electrolytes *' in the fused state. 

In recent years the Hall, Kleiner, Heroult and other processes 
have been applied successfully to the commercial production of 
aluminum on a large scale. The electrolyte employed in these 
cases is a fused compound of aluminum, such as cryolite. 

In the Hall process alumina is dissolved in a bath of melted cry- 
olite, the former being the actual electrolyte which is decomposed. 
This is precisely analogous to decomposing a copper compound 
dissolved in water except that by the use of the fused solvent the 
oxidation of the aluminum, which would occur if water were pres- 
sent, is avoided. 

Non-conducting coatings on electrodes often greatly increase the 
internal resistance of electrolytic cells. These insoluble coatings 
may in some cases almost prevent the passage of the current. They 

* Jour, Chem, Sec, March, 1888, p. 211. 

\ Proceed, Birmingham Phil Soc ^ Vol. V., Part II., p. 371. 

X t^^ig' '*** t CLIV., p. 161. 

§ Jour, Chem, Soc., March, 1888, p. 211. 


may be avoided either by using electrodes which are not affected 
by the action, by mechanically cleaning the plates or by adding 
to the electrolyte something which will dissolve off the coating. 

llu Counter E.M.F, in an electrolytic cell or the direct E.M.F. in 
a primary or secondary voltaic cell may be calculated from the 
energy involved or from the solution pressures according to the 
theory of ionization. The former method is the older and will be 
given first. 

A current in passing through an electrolytic cell produces a cer- 
tain amount of chemical action in decomposing the material pres- 
ent. The principle of the conservation of energy requres that the 
electrical energy thus consumed must be equal to the chemical 
work performed, assuming for the present that the electrical energy 
is completely converted into chemical energy. Now we know that 

Electrical Energy = Cet, (17) 

in which Ce is the product of the current and counter E.M.F.» 
that is, the number of watts used in doing the chemical work, and 
/ is the time in seconds. 

In chemistry the affinity or energy of combination of the various 
compounds may be measured by the heat which is liberated, when 
the combination takes, that is. 

Chemical Energy = wh, (18) 

w being the weight of the given metal or element in grams, and k 
being the number of heat units produced when one gram of this 
metal combines with the other constituent of the compound. 
The unit of heat is the gram calorie or heat required to raise one 
gram of water i®C. in temperature. Now from Faraday's law we 
know that the weight of metal deposited by a given current is 
equal to the product of the current, the electro-chemical equiva- 
lent of the metal and the time in seconds, hence 

w^ Czt, (19) 

which substituted in (18) gives 

Chemical Energy = Czht. (20) 

This chemical energy must, as already stated, equal the electri- 
cal energy (17), which liberates it, but since (20) gives the chem- 
ical energy in heat units, it is necessary, in order to equate the two 
values, to multiply the latter by the constant 4.18 to convert it 


into electrical units. This constant is the number of electrical 
units of energy (watt-seconds or joules) in one heat unit (gram- 

By combining equations (17) and (20) we have 

Cet^\.\%Czht, (21) 

and by cancelling out C and /, 

e^^AZzh, (22) 

which is the relation deduced by Lord Kelvin.* It may be used 
to calculate the counter E.M.F. set up by a given electrolyte when 
it is decomposed by a current. It also gives the direct E.M.P\ 
produced by a certain chemical combination, as in the case of a 
primary or secondary battery. 

Instead of stating the heat of combination in terms of one 
gram of the given metal, the more modern method is to express 
it in terms of one equivalent. For example, the chemical equiva- 
lent or combining weight of zinc is 32.45, hydrogen being taken 
as unity, and the heat of combination of 32.45 grams of zinc, with 
various other substances, are the values usually given in thermo- 
chemical tables. This is much more convenient and scientific, and 
enables direct comparisons and substitutions to be made, since 
each substance must always combine in exactly that proportion. 
In fact, by introducing this relation into equation (22), we can still 
further simplify it. Calling H the number of heat units produced 
by the combination of one equivalent Z of the given material, then 
the heat of combination per gram 

and substituting this value in equation (22) we obtain 

^ = 4.i8jr//; (23) 

The ratio -r^, which is the electrochemical equivalent divided by 

the chemical equivalent, is a simple numerical constant, since these 
two equivalents are the same thing expressed in different units. 
This ratio is always equal to .CXX)0I044, which, substituted in 
equation (23), reduces it to the simple form 

* Philosophical Magazine^ Dec, 185 1. 

VOL. XXII. — 10 


e = .0003436^^ (24), or ^ = - ^ . 

^ ' 22936 

For the latter we may take, without appreciable error, the fol- 
lowing very convenient formula : 

^= —, (25) 

In many cases the calculated E.M.F. (which may either be di- 
rect or counter) obtained by the use of equation (25) agrees almost 
exactly with that found by experiment, and in nearly all cases the 
disagreement is not more than a small fraction of a volt. 

The values of H for various combinations can be found in many 
electrical and chemical books. 

A paper on the " Possibilities and Limitations of Chemical Gen- 
erators of Electricity," by the writer ( Tm;/^. Amer. Inst. Elec, Eng.^ 
Vol. v., p. 277), gives results of combinations of thirteen of the 
most important metals with chlorine, bromine, and iodine, respec- 
tively, and the average difference between the calculated and deter- 
mined values is not more than about one-tenth of a volt. The 
calculated E.M.F. of Daniell and other primary batteries also agree 
well with the actual value. Nevertheless in some cases the dis- 
crepancies are considerable. They are due to the fact that all of 
the energy of combination may not be converted into electrical 
energy, some of it being converted into heat, which appears in the 
cell, hence equation (25) is not strictly true and it must be modi- 
fied as follows : 

e = r- (26) 

23000 ^ ^ ^ 

in which y is the amount that the actual E.M.F. falls short of that 
which would be obtained if the conversion of the energy of com- 
bination into electrical energy were complete. It is also possible 
for the action to absorb heat, in which case the E.M.F. would be 
higher. Therefore we may have 

^«- -±J. (27) 

23000 ^ '' 

The value of y may be deduced by considering the case of a 
reversible voltaic cell, which is caused to pass through a complete 
cycle of operations in the following manner : 

A secondary cell at an absolute temperature T'is charged with 


one coulomb, and, assuming the internal resistance to be negligible, 
the energy required in joules is numerically equal to e the E.M.F. 
The chemical action involves one electrochemical equivalent of 
the active materials since the quantity of electricity is one cou- 
lomb ; for example, one electrochemical equivalent of the positive 
metal would be reduced. 

The temperature of the cell is now raised to T+ dl and its 
E.M.F. becomes e + de. The action is then reversed, one cou- 
lomb being discharged and the cell is brought back exactly to its 
original chemical condition, and lastly it is cooled to its original 
temperature T, which completes the cycle. The electrical energy 
generated during the discharge is ^ + de. 

The increment de must be due to the conversion of a certain 
amount of heat into electrical energy since the temperature is the 
only condition which is different during the discharge. 

We know from the second law of thermodynamics that, when- 
ever heat is transformed into any other form of energy, the effi- 

ciency of conversion is equal to ^. 

In the present case a certain quantity of heat, which may be 
designated (in joules) as q^ is given to the cell and a certain 
amount of electrical energy is produced, hence : 

de dT 

Let us leave, for a moment, the case of the cell whose temper- 
ature is purposely changed by adding or subtracting external heat 
in order to study the effects, and return to the ordinary cell whose 
temperature changes are due to its own action. 

It was stated, in connection with equation (27), that when a cell 
absorbs heat it tends to increase the E.M.F., in which case 

^ = 2 + ^i (29) 

Q being the heat of combination in the cell, and q^ is the heat ab- 
sorbed, both expressed in joules per electrochemical equivalent • 

^1 = ^-5. (30) 

Now, since q^ is the heat absorbed by the action of the cell, it 
follows that it corresponds to q in equation (28), which is the ex- 


ternal heat given to the cell; therefore we may substitute the 
former for the latter in equation (28), and we have : 








T - 


This is called the equation of Helmholtz, and represents the 
amount that the E.M.F. may be less than or exceed that obtained 
by Kelvin's expression ; hence it should be substituted for y in 
equation (27), which then becomes : 

"" 23000 dT ' '^^^ 

,y, is the rate at which the E.M.F. varies with change of temper- 
ature ; hence it is the temperature coefficient of the cell, and it may 
be positive, negative or zero. If, in equation (32), 

e-Q>o, (34) 

the temperature coefficient is positive ; that is, the E.M.F. rises 
with the temperature ; but if 

^-Q<o, (35) 

the E.M.F. falls with increase to temperature; and if 

' = Q, (36) 

the E.M.F. is independent of the temperature, and Kelvin's expres- 
sion, equation (25), is exactly fulfilled. 

Hence, all that is necessary is to ascertain the effect of tempera- 
ture upon the E.M.F. of a cell, in order to determine whether its 
actual E.M.F. will be higher than, equal to, or less than, that which 
is equivalent to Q, the energy of combination. Confusion or error 
might be introduced, because of thermoelectric or Peltier effects ; 
but these should be guarded against. The correctness of equa- 
tions (32) and (33) were proved experimentally by Jahn,* who di- 
rectly determined the total heat produced, by placing the entire 
cell and its circuit within a Bunsen ice calorimeter, thus obtaining 
the value of Q, He also measured ^, the E.M.F., and then com- 

*lfui/, Anm., Vol. 28, pp. 21 and 491. 1886. 


pared the value e ^ Q, found in this way experimentally, with that 
obtained by determining the temperature coefficient and calcula- 
ting e — Qhy equation. (32). In most cases the results agree with 
remarkable closeness considering the difficulty of making such 
measurements. Taking, for example, a Daniell cell in which the 
reaction is CuSO, + iooH,0 + Zn = ZnSO, + iooH,0 + Cu, the 
E.M.F. was found to be 1.096 volts, which, at 23,050 gram 
calories per volt (the value adopted by Jahn), corresponds to 25,263 
calories; and e— Q was calori metrically determined to be + 208, 
whereas, by calculating it from the temperature coefficient ^ — (2 = 
+ 214. The agreement is almost perfect; and it may also be 
noted that the error in calculating the E.M.F. of this Daniell cell 
by (25) without considering (32) is less than one per cent. In the 
case, however, of a silver chloride cell in which the reaction is 
2AgCl + looHjO + Zn = ZnCI, + iooH,0 + 2Ag, the value of e 
is 1.031 volts s= 23,753 gram calories and ^—(2 = — 2,330 (ob- 
served) or — 2,574 (calculated). In this case the E.M.F. is almost 
exactly one-tenth of a volt less than given by equation (25). 

The explanation is that the physical work or heat of solution is 
much greater in the latter case; hence, the temperature coefficient 
and the correction required by the Helmholtz equation (32) are 
more important factors. 

In fact, if the action in a cell is purely chemical, and physical 
effects, such as heat of solution, production of gases, etc., are in- 
significant, then the temperature coefficient is correspondingly 
small, and the last term in (33) may be neglected; whereas, if the 
heat of solution is great or there is considerable generation of gas, 
then the temperature coefficient is usually large, and the E.M.F. 
may be 10 or perhaps 20 per cent, higher or lower than that given 
by equation (25). 

The general principle is that the energy due to true chemical 
combination is practically all convertible into electrical energy, 
whereas the energy depending upon physical action, such as solu- 
tion and change of state, is only partially convertible, being limited 
by the second law of thermodynamics as expressed in (28). The 
energy which can be transformed is called by Helmholtz " free,** 
whereas the remainder, which must appear as heat, is designated 
as " bound " energy. The electrolytic cell being perfectly analo- 
gous to the primary cell in all its actions ; what is true of the di- 
rect E.M.F.of the latter applies to the counter E.M.F. of the former. 


[CbDtribatioDS from the Havemeyer Labontories of Columbia University, No, 38.] 



Part II. Iron and Manganese Slags. 

The constituents determined in iron slags usually are silica, iron, 
alumina, lime, magnesia, phosphorus and manganese but arsenic, 
vanadium, chromium and tungsten occur in small quantities at 
times in the slags of today and methods for their estimation are in 
demand. A general scheme is given on the opposite page for the 
complete analysis of an iron or manganese slag to which the follow- 
ing notes refpr, followed by special methods for slags which vary 
in type and solubility. 

Note i. — If vanadic acid exists in the slag, it must be separated 
from the iron before an accurate estimation of the iron can be 
made ; it also is carried down with the titanium. In order to free 
the titanic oxide from any alumina, phosphoric acid, vanadic acid, 
ferric or ferrous oxide, fuse the ignited precipitate with sodium 
carbonate and a little sodium nitrate for one hour at a high heat, 
boil out the fusion with a solution of sodium carbonate and filter 
out residue, wash well with water containing some sodium car- 
bonate. Ignite residue of sodium titanate and fuse with sodium 
carbonate, treat the cooled mass in the crucible with a few drops 
of sulphuric acid and heat until fumes of SOj are evolved. Cool, 
dissolve in water, neutralize the free acid with ammonia, add 5 
grams sodium acetate and 10 c.c. of acetic acid 1.04 sp. gr. Boil 
for three minutes and filter the hydrated oxide of titanium, wash 
well, ignite and weigh as TiOj. 

If the vanadium and iron are to be thrown down from the sever- 
ally combined filtrates from hydrated titanic oxide, after oxidation 
of the iron, by ammonia, being careful to boil out excess ; then 
they are completely precipitated and the iron may be estimated 
as in note below. It is better to take new portions for the estima- 
tion of both vanadium and iron. 

Note 2. — To estimate the iron in the presence of vanadic acid 
take 0.5 gram of the slag and free from silica as usual ; .neutralize 
the solution with ammonia, and boil up with ammonium acetate. 
Repeat this precipitation. Dissolve precipitate in hydrochloric 




Fuse I gram of the finely divided slag with 5-6 grams of the mixed a 
solve fusion in a casserole with dilute hydrochloric acid and evaporate to 
hydrochloric acid, break up any lumps and repeat evaporation and dehydr; 
and three drops of sulphuric acid, heat to boiling and filter on an ashless | 
clay annealing cup in a mufHe furnace or in a weighed platinum crucible, 
residue may consist of barium sulphate, titanic oxide, alumina, etc. It is 1 
with water, filtering the residue and washing to free the residue from sulpl 
tested for barium by addition of sulphuric acid. Any precipitate of BaSO^ 
in water is acidified with dilute hydrochloric acid and added to the main 
arsenic or other metals of the fifth and sixth groups with H,S gas, and if 
potassium chlorate, boiling out all free chlorine. Divide the solution into 
and titanic acid and the remaining three-fifths for the estimation of other c 

To the portion representing two-fifths of the solution and equal to o.. 
ina passes into the filtrate, the ferric hydroxide (not completely if vanadic s 
behind. Dissolve precipitate in cold dilute sulphuric acid and neutralize d 
and one-sixth of the total volume of acetic acid, sp. gr. 1.04, pass H,S gas 
ten minutes. The hydrated oxide of titanium thrown down is filtered fn 
ignited, weighing as TiO, + P,0,. The ignited oxide is usually quite impi 
water, and the hydrated oxide of titanium again precipitated_as before, ig 


dorbonates of potassium and sodium and 0.5 gram of nitrate of potassium. Dis« 
tetolrynesSy heat at no® C. until moisture is expelled, take up with a few c.c. of 
.•hydntion. Take up with 50 c.c. of water and add 20 c.c. of cone, hydrochloric acid 
ilcss paper with aid of suction pump. Wash thoroughly with hot water, ignite in a 
dbk.. Weigh silica. Expel with HP + H,SO^. and determine silica by loss. The 
Itistested for barium sulphate by fusion with sodium carbonate, extracting the melt 
salpbtes. The residue is dissolved in a few drops of dilute hydrochloric acid and 
aSO^s filtered after boiling and weighed as BaSO^. The portion of the melt soluble 
main filtrate. At this point the main filtrate should be tested for the presence of 
od if present should be removed; after H,S treatment the solution is oxidized with 
intotro parts, one containing two-fifths oi the solution for the determination of iron 
Jief constituents. 

to 04 gram, add potassium hydrate in excess, warm, filter and wash well, the alum- 
adicadd is present) and hydrated oxide of titanium carrying phosphoric acid remain 
lire drop by drop with ammonia, leaving slightly acid, add 20 grams sodium acetate 
S gas to reduce the iron and when reduced heat the solution to boilmg, continue for 
ed ftoin the hot solution, washed with hot water containing a little acetic acid and 
iopune and should be fused with bisulphate of potassium, the melt dissolved in cold 
e, ignited and weighed. Add the filtrate to the former filtrate from hydrated oxide 


acid and add tartaric acid in sufficient quantity to prevent any pre* 

cipitation by ammonia ; reduce with HjS gas, make ammoniacal, 

and add ammonium sulphide and set aside in a warm place. The 

iron sulphide is filtered, washed and dissolved in hydrochloric 

acid, the solution oxidized with nitric acid, and filtered from free 

sulphur; the iron is then precipitated by ammonia and determined 

as FejO,. Or, the washed ferric hydroxide, after solution in dilute 

sulphuric acid, may, after reduction with HjS, expelling excess in 

atmosphere of carbonic acid, be determined by -KMnO* or 

- KjCraOj solution. (Compare Blum, Ztschr. anal. Chem., 1900, 


If vanadium exists in the slag and iron is to be determined in 
the portion taken for the titanium estimation, then the treatment 
with potassic hydrate must be omitted, as the iron is not com- 
pletely precipitated by potassium hydrate in the presence of van- 
adium. The titanium is separated as usual, and to the combined 
filtrates containing all of the iron (in a reduced state) tartaric acid 
is added in sufficient quantity to prevent the precipitation of iron 
by ammonia, which is added in slight excess, then HjS is passed 
in. The precipitated ferrous sulphide is filtered off, washed two 
or three times and dissolved in dilute HCI, evaporated with sul- 
phuric acid to fumes of SO3, diluted with water, reduced and titrated 

with — KMnO,. 
10 ^ 

The determination of iron in the slag in the presence of vanadium 
may be also made as follows : Evaporate i gram with sulphuric 
and hydrofluoric acids in a platinum dish until silica is expelled, 
then mix the residue with sodium carbonate and fuse, extract with 
water, filter and wash well, dissolve the insoluble residue in dilute 
sulphuric acid, reduce with HjS, boil out excess of gas in a cur- 

rent of carbon dioxide, and titrate the iron with — KMnO.. Some 

iron may pass into the aqueous extract, and must be recovered by 
acidification with sulphuric acid and precipitation of the iron with 
ammonia, and refusion of the precipitate with sodium carbonate, 
with treatment as above. 

Note 3. — In this portion sufficient iron must be present to carry 


all of the phosphoric acid, and, as in some slags the amount of iron 
is small and the amount of calcium phosphate large, it is necessary 
to add to the portion of three-fifths, before the basic acetate is made, 
a sufficient quantity of solution of ferric chloride (free from P), and 
in which the quantity of iron is known. This is deducted later in 
the calculation for the AljOg. 

Note 4. — In the 100 c.c. portion for FejO,, TiO^, AljO,, PjOj, 
CrjO,, V2O5, if vanadic acid is in the slag, the alumina and vanadic 
acid are not completely thrown down by ammonia, and some 
alumina and vanadic acid, and possibly some phosphoric acid, pass 
into the filtrate, and must be determined and the weights added to 
the united precipitate, giving the total Al^Og -f FejO, -f P^Oj 
+ TiOj -f V2O5. and from the figures obtained the results may be 
calculated. A trace of aluminium and considerable of the vanadic 
acid are in the filtrate. In such a case the succeeding method can 
be satisfactorily employed. 

If vanadic acid exists in the slag, the separation of AljO,, TiOj, 
Fe,0,. CrjOg, PjOg and VjOg is made as follows : 

To the 100 c.c. portion, acid with nitric acid, add tartaric acid 
to sufficient quantity to keep up the iron from precipitation later 
with ammonia, reduce with H^S gas (the iron is first reduced to 
the ferrous condition by hydrogen sulphide in acid solution, in 
order to prevent the precipitation of titanium, which is otherwise 
likely to happen), and add ammonia in slight excess, and then 
ammonium sulphide in excess ; set aside in a warm place, filter 
out ferrous sulphide, evaporate filtrate to dryness and fuse with 
sodium carbonate and nitrate, extract fusion with water. Sodium 
titanate remains behind and sodium chromat^, aluminate, phos- 
phate and vanadate go into solution. Repeat this treatment. 
Chromium may here be determined volumetrically. Filter and 
evaporate filtrate to dryness in a platinum dish with ammonium 
nitrate in sufficient amount to react with all the carbonate, and 
digest on bath until most of the ammonium carbonate is gone. 
This removes all of the PgOj and nearly all of the AIjO,. Decant 
and wash residue with weak ammonium nitrate solution, dissolve 
in nitric acid and estimate phosphoric acid, if desired, with ammo- 
nium molybdate. The alumina can be separated in the filtrate 
after removal by HjS of the molybdenum. 

The filtrate from the AI2O3 -f P2O5, containing vanadic acid, 
alumina and chromic acid, is concentrated if necessary, made 


ammoniacal and a current of HjS gas passed in. Precipitate 
= Al2(OH)8 + Cr2(0H)^, filter and dissolve ppt in hot nitric acid, 
evaporate nearly to dr>'ness, and heat with some strong nitric acid 
and potassium chlorate in crystals, and finally evaporate to dry- 
ness, to get rid of the acid. Dilute with cold water, add bicar- 
bonate of soda in slight excess; filter after several hours the 
separated aluminium hydrate. Chromium is thrown out of the 
filtrate by fresh ammonium sulphide, redissolved to free from 
alkali, reprecipitated and weighed. 

If vanadium is present, it is better to use the colorimetric method 
for the estimation of the chromium, as devised by Dr. W. F. Hille- 
brand. See /our. Am, Chem. Soc.^ 1898, Vol. 20, p. 454; Chem. 
News, 1898, jZ, 227, 239; Bull. U. S, Geol. Survey, No. 167, p. 37. 
First applied by L. de Koningh, Nederl. Tyds. Voor. Pharm. Chem, 
and Tax,, 1889. 

This method is largely taken from one derived by Dr. T. M. 
Chatard for the separation of the above constituents. See Am. 
Chem, /., 1891, 13. ic6; Bull, U. S. Geol, Survey, No. 78, p. 87; 
Chetn. News, 1 891, 63, 267. 


Ten grams of the finely pulverized slag are mixed with 50 
grams of sodium carbonate and six grams of sodium nitrate. The 
mixture is added in small portions to a platinum crucible and each 
portion fused before the next addition. When all the mixture is 
added, the whole mass is heated for one hour to a very high tem- 
perature by two lamps. The melt is digested in boiling 
water until the carbonates, etc., are dissolved and the ferric oxide 
disintegrated. A second fusion of the residue is necessary to sep- 
arate all of the vanadium. The aqueous extracts are combined 
and alcohol is added to reduce the manganese, after which the 
greater part of the alkali is neutralized with nitric acid leaving the 
solution just alkaline. Boil out carbonic acid. By this operation 
the alumina and the silica will be precipitated. These are filtered 
off and as the precipitate may retain some chromium and possibly 
some vanadium, as a precautionary measure, the silica and alumina 
precipitate should be evaporated with some sulphuric and hydro- 
fluoric acids, the residue fused with a little sodium carbonate and 

* Pope, Trans. Amer. Inst. M. E., 1899, 372. 


the water extract nearly neutralized with nitric acid, boiled for a 
short time, the filtrate being added to the main one. This filtrate 
is made slightly acid with nitric acid, and again made alkaline by 
the addition of a few drops of sodium carbonate, boiled for a few 
minutes and again filtered. To the alkaline liquid, barium nitrate 
is added. The precipitate, consisting of barium vanadate and car- 
bonate, also chromate, phosphate, if these elements are present, is 
collected on a filter paper and the paper with its contents is di- 
gested in a beaker with dilute sulphuric acid for some time. The 
solution is now filtered, and concentrated to about lOO cc, ren- 
dered alkaline with ammonia and heated to form ammonium met- 
avanadate. Solid ammonium chloride is added until it dissolves 
with difficulty and then 200 cc. of a mixture of alcohol and ether 
(1:1). The ammonium metavanadate now begins to crystallize 
out and after standing several hours in a cool place, may be filtered. 
The precipitate is washed with a strong solution of ammonium 
chloride which contains alcohol and is also alkaline with ammo- 
nia. Wash finally with the mixture of alcohol and ether. The 
dried precipitate is ignited in a porcelain crucible, two or three 
drops of nitric acid added, again ignited and weighed. The precip- 
itate may be impure with small amounts of phosphorus or chro- 
mium as well as tungstic acid. In order to obviate any such error 
the vanadic oxide is dissolved in sulphuric acid (1:1), reduced with 
sulphurous acid, after dilution. The V^Oj is reduced to V,0^ and 
the excess of SO, can be removed by boiling the acid solution in a 
current of carbon dioxide. The solution having a volume of about 


ICK) CC. is heated to 70^-80^ C. and titrated with — KMnO. so- 

' 100 * 

lution. Phosphoric acid does not affect the result, chromium is 
only oxidized at a higher temperature while tungstic oxide would 
remain undissolved when the vanadic oxide is treated with sul- 
phuric acid. 

Sulphur. — Take 2 grams slag, treat with aqua regia and evapo- 
rate to dryness and dehydrate silica ; take up with cone. HCl, boil 
five minutes and dilute with hot water, filter with aid of suction, wash 
well with hot water. Fuse residue with sodium carbonate, extract 
fusion with boiling water, and filter from insoluble residue, evapo- 
rate this filtrate to dryness and dehydrate silica at 1 10° C. Take 
up with hydrochloric acid and filter ; add this filtrate to the main 


filtrate from acid treatment of the slag ; nearly neutralize with am- 
monia leaving slightly acid ; heat to boiling and add 10 c.c. of boil* 
ing BaCi, solution (10% solution), boil half an hour, settle and filter 
on ashless paper. Ignite and weigh BaSO^. 

Wt. of BaSO, X .13734 = wt. of S. 

Potassium and Sodium. — Alkalies are determined as directed 
under lead slags. If the slag, however, is not soluble in acids, one 
gram is finely ground and intimately mixed with 8 grams of CaCO, 
(free from alkalies) and % gram of sublimed NH^Cl, C. P. and 
heated in platinum crucible for one hour at low red heat, care- 
fully avoiding any fusion of the mass, disintegrated by boiling 
water and filtered from the insoluble residue and the analysis pro- 
ceeded with as directed in alkalies estimation in lead slag, com- 
mencing with the addition of ammonium carbonate solution, etc. 

Tungstic Acid.— Weigh out 3 grams of the slag into a plati- 
mum dish, treat with hydrofluoric and sulphuric acids, evaporate 
to dryness, and fuse with 15 grams of sodium carbonate for five 
minutes, extract fusion with boiling water, filter from insoluble res- 
idue and neutralize the solution with dilute nitric acid, free from 
nitrous acid, making very faintly acid. Add very . cautiously a 
few drops of sodium carbonate solution in sufficient amount to 
make the solution neutral or faintly alkaline. This point can be 
determined by placing on a white plate some drops of methyl 
orange as an indicator. The tungstic acid is next precipitated by 
boiling and adding, drop by drop, with constant stirring, 3 c.c. of 
a saturated solution of mercurous nitrate in water (freshly pre- 
pared). The tungstic acid is precipitated as Hg,WO^ mixed with 
basic mercurous nitrate ; the liquid is boiled for a few minutes and 
allowed to digest on hot plate until the liquid is clear. 

The bulky precipate is filtered on an ashless paper and thoroughly 
washed with hot water; it is dried, ignited, and weighed when 
cold in a weighed platinum crucible. The residue consists of 
WO, + SiO,. The silica is removed by treatment of the oxides 
with HF and H,SO^, evaporation to dryness and final ignition. 
Weigh. The loss equals silica. The weight remaining is WO,. 

A direct estimation of the iron and vanadium* may be made by 
fusing one gram of slag with bisulphate of potash with solution of 
the melt in hot water containing some sulphuric acid, reduction of 

♦Wells and Mitchell, Jour, Amer. Chcm, Soc, 1895, '7» 7^* Chem, Ntws, 1896, 
73. "3. 


the iron and vanadium by H,S gas, the titanium not being re- 
duced, the excess of H^S removed by boiling the liquid in a cur- 
rent of carbon dioxide. The solution is cooled, the carbon dioxide 

still passing and the iron and vanadium are titrated by — KMnO^ 

solution at a temperature of about /o-So^C. The V^Oj is reduced 
to VjO^, and its action on KMnO^ is equivalent to two molecules 
of FeO. 

If the amount of permanganate used for titration of vanadium is 
so small as to throw doubt on its presence, it is necessary to apply 
a qualitative test and this is done by reducing the bulk to lO c.c, 
adding ammonia in excess and passing in H,S gas. Vanadium is 
indicated by a cherry-red color. 


In the examination of iron furnace slags and cinders five typical 
constitutions can be recognized, which conveniently can be divided 
into two classes : 

1. Slags insoluble in HCi. 

2. Slags soluble in HCI. 
The first class includes — 

Blast furnace slags, neutral double silicate of lime and alumina. 

Finery slag, ferrous silicate. 

Slag from acid Bessemer process, consisting of a double silicate 
of the protoxides of manganese and iron. 

The second class includes — 

Slag from the basic process, chiefly basic phosphate of lime. 

Tap cinder from puddling furnace, mainly a basic silicate of 
ferrous and ferric oxides. 

Owing to the difference in composition of these types of slags, 
it is not convenient to follow the general scheme at times, and the 
special methods may give greater satisfaction and be more speedy. 

Silica. — Fuse 0.500 gram with 5 grams mixed carbonates of po- 
tassium and sodium and 0.500 gram of sodium nitrate, extract fu- 
sion in porcelain casserole with dilute hydrochloric acid, evaporate 
to dryness and heat in air-bath at 1 10° C. until the odor of hydro- 
chloric acid has disappeared and the silica is dehydrated, take up 
the mass with a few c.c. of dilute HCI, evaporate and dry again at 
1 10^ C. take up with dilute HCI and water, boil, breaking up lumpy 


silica and filter on a small paper, wash well with hot water and oc- 
casionally with dilute HCI. Ignite residue and weigh the silica. 
Expel with HF + 3 drops H^SO^. Loss in weight equals silica. 
Calculate percentage. 

Alumina and Ferric Oxide. — To the filtrate from the silica add 
ammonia in slight excess, boil until no odor of free ammonia, 
filter on a fluted filter paper, wash well, then place paper and pre- 
cipitate in the beaker and dissolve in dilute hydrochloric acid, 
render again ammoniacal, boiling out excess and filter as before and 
wash thoroughly with boiling water ; this precipitation should be 
repeated if the salts from the fusion are not washed out by this 
time. Washing by decantation is sometimes possible and is an 
efficient way of quickly freeing the iron and aluminium hydroxides 
from the sodium and potassium salts of fusion. If much iron hy- 
droxide, dry the precipitate before ignition, and in any case finish 
the ignition with blast lamp. Weigh as A1,0, + Fe,0, + PjOj. 
If much manganese is present in slag, a basic acetate is made 
instead of an ammonia precipitation. 

Lime and Magnesia. — Boil down filtrates from the iron and 
alumina precipitations, add a few drops of ammonia and pass in 
H,S gas, if any manganese, nickel or zinc sulphides, filter out. To 
the filtrate add ammonium oxalate in excess, boil ; white precipi- 
tate is CaCjO^. Filter and wash well with hot water. Ignite and 
weigh as CaSO^. Calculate to CaO percentage or dissolve the 
precipitate in hot dilute sulphuric acid and titrate the oxalic acid 
with standardized KMnO solution. 

I c.c. KMnO. = .0028 gm. of CaO. 
10 * ^ 

To the filtrate from the lime, add hydro-di-sodium phosphate, keep 
cold and stir well, a white crystalline precipitate of MgNH^PO^ 
appears and is filtered off, washed with dilute ammonia, ignited 
and weighed as Mg^P^O,. Calculate percentage of MgO. 

Ferrous Oxide. — Fuse one gram of the slag with mixed car- 
bonates of potassium and sodium, extract with hot water and hy- 
drochloric acid, digest until the iron is all dissolved, add i c.c. of 
nitric acid and boil until complete oxidation of the iron has taken 
place, neutralize with ammonia and precipitate the iron with am- 
monium acetate, filter and wash several times with hot water, re- 


dissolve the iron hydrate in dilute HCI, neutralize, reduce and ti- 

trate with KjCr^O^ solution, i c.c.= .0056 gram Fe. Calculate 

percentage of Fe and also to Fe,0,, which latter figure deducted 
from combined weights of A1,0, + Fe,0, gives weight of A1,0,. 

Manganous Oxide. — To the filtrate from the acetate of iron, 
add I ex. bromine and a slight excess of ammonia ; the manga- 
nese is thrown down mixed with some alumina ; filter oiT pre- 
cipitate, and dissolve it in HCl, boil to small bulk and add excess 
of HNO,, 1.42 sp. gr., boil to low bulk to expel all traces of hy- 
drochloric acid, repeat treatment with HNO,-|- i gram of KCIO,, 
dilute with 75 c.c. hot water, and filter the hydrated oxides of 
manganese on asbestos, wash well until free from nitric acid and 
then to a known amount of ferrous sulphate in solution, after boil- 
ing, add precipitate, and when it has dissolved, titrate the excess 
of iron with a standard solution of K,Cr,0^ using potassium ferri- 
cyanide as an indicator. From the amount of iron oxidized by 
the manganese oxide the percentage of manganese can be calcu- 

Sulphur. — See foregoing scheme for its estimation. 

Phosphoric Acid (P^Oj). — A rapid estimation can be made by 
treating 10 grams of slag with strong nitric acid, filtering from res- 
idue which is fused with sodium carbonate and the melt extracted 
with water, filtered from residue and the filtrates combined, evapo- 
rated to dryness, and taken up with dilute nitric acid and the iron 
precipitated with ammonia, the ferric hydroxide and phosphate is 
dissolved in dilute nitric acid and the phosphorus precipitated by 
acid ammonium molybdate solution, by shaking for 10 mmutes, 
observing proper conditions as to temperature and acidity, it is 
then quickly filtered and titrated as in general scheme. 


Silica. — Fuse 0.5 gram of finely divided slag as in the case of 
blast furnace slag, extract with water and acidify with hydrochloric 
acid, evaporate to dryness, and dehydrate the silica. Take up 
with dilute hydrochloric acid, boil and filter off silica. Ignite and 
weigh as usual. 

Ferrous Oxide. — Oxidize and precipitate the iron in the 


filtrate from the silica, filter, wash well, and dissolve in hydro- 
chloric acid, reduce with stannous chloride and add mercuric 

chloride and preventive solution. Titrate as usual with — K,Cr,0^ 


Manganous oxide, phosphoric acid and sulphur are estimated 
as described under blast furnace slag. 

Lime. — One gram of the slag is fused with 8 grams of sodium 
carbonate, extracted with dilute hydrochloric acid, and evaporated 
to dryness and the dehydrated silica filtered o<T. The iron and 
alumina in the filtrate are twice precipitated with ammonia, and 
H,S is passed into the filtrate to separate manganese which sul- 
phide is filtered off. The lime in the filtrate is precipitated with an 
excess of ammonium oxalate, the solution boiled and the CaC,0^ 

filtered off, dissolved in dilute H,SO^ and titrated with — KMnO, 

* * 10 * 

solution. I c.c. =B .0028 gram CaO. The magnesia is thrown 

down in the filtrate from the lime in a cold solution by excess 

of hydro-di'Sodium phosphate. It is filtered and treated as usual, 

weighing as Mg^P^O,. 

Alumina. — ^The precipitate of hydroxides of iron and aluminium 
from the above determination is ignited and weighed as A1,0, + 
Fe,0, + PjOj and the weight of iron found by direct test, expressed 
as Fe,0,, deducted from the combined weights gives AI,0, -|- PjO^ 
by difference. This is of course assuming the absence of TiO, 
and also that the amount of P,Oj is small. The weight of P,Oj 
found elsewhere deducted from A1,0, + PjO^ gives Al,Oj by 


Silica.—Fuse 0.5 gram of the slag and determine the silica as 

Manganous Oxide. — Saturate the filtrate and washings from 
the silica with bromine and add slight excess of ammonia. Filter 
the precipitated manganese oxides and dissolve in dilute HCl, add 
5-10 c.c. of cone, sulphuric acid and evaporate to fumes of SO,. 
Dilute with water and add sodium carbonate until nearly neutral, 
and cream of zinc oxide until all of the iron is precipitated and 


the solution clears. Dilute to 500 ex. and take an aliquot portion 

of 100 c.c. and filter for titration with KMnO. solution, observ- 

10 * 

ing the usual precautions for Volhard's method. 

Lime and Magnesia. — Fuse 2 grams of slag and separate silica 
in the usual way. Dilute the filtrate to 200 c.c. and mix well. In 
100 c.c. the iron and alumina are twice precipitated with ammonia 
and after separation of the manganese as sulphide, the lime and 
magnesia are determined as usual. 

The precipitated iron and alumina is dissolved in dilute sul- 
phuric acid and boiled up with 10 grams of zinc (free from iron) 

until all of the iron is reduced and the iron is titrated with 


KMnO^ solution. A blank should be run on the same weight of 

zinc for the small traces of iron it may contain. 

In the remaining 100 c.c. the alumina is determined by precipi- 
tation of the iron and alumina by ammonia, boiling out excess, 
filtering and washing very thoroughly with hot water, igniting and 
weighing as Al^O, + Fe^Og. 

The precipitate is dissolved in cone. HCl and after reduction is 

N N 

titrated with KMnO, or K,Cr.O solution, or the precipi- 

10 * 10 * ^ ^ r r 

tate may be fused in a silver crucible with caustic potash (pure by 

baryta) and the ferric hydroxide estimated by weighing as Fe^O, or 

the alumina separated from the filtrate. 

If the iron has been estimated in the first 100 c.c. taken, this 

latter treatment is unnecessary, as the iron oxide may be deducted 

from the combined oxides, giving alumina by difference. 


The constituents of this cinder are chiefly ferrous and ferric oxides, 
silica, and phosphoric acid ; fusion is sometimes necessary but the 
cinder is largely soluble in acids. 

Silica. — Fuse 0.5 gm. cinder with 5 grams of acid sulphate of 
potash in a platinum crucible, extract fusion with water and a little 
sulphuric acid, filter off residue, wash well, ignite and weigh. 
Drive off silica with HF + H,SO . Loss in weight equals silica. 

Ferrous Oxide. — Oxidize the iron with i c.c. of nitric acid and 


precipitate with ammonia, filter and dissolve in sulphuric acid and 

reduce with zinc as before in acid Bessemer slag. Titrate with 


— KMnO solution. 


Phosphoric Acid. — ^This is determined in a separate portion of 
1-2 grams and, after separating silica, the iron is thrown out with 
ammonia, carrying all of the phosphorus. This precipitate is dis- 
solved in nitric acid and the phosphorus thrown out as phospho- 
ammonium molybdate and estimated in the usual way by weighing 
as Mg,P,0, or if the proper precautions have been observed as to 
temperature and degree of acidity, the phosphorus having been 
precipitated by acid ammonium molybdate solution, the Mo,0, 

may be titrated by KMnO^. 

•^ 20 * 

Sulphur is estimated as in blast furnace slag, manganous oxide as 
in blast furnace slag. 

Lime and magnesia are determined in a separate portion of one 
gram and after separating silica, alumina and iron, are precipitated 
as usual. 

Alumina is determined in the precipitate of iron and alumina 
obtained in the lime and magnesia determination. 


Silica. — One gram of slag is dissolved in aqua regia and the sil- 
ica dehydrated, filtered off and weighed. 

Phosphoric Acid. — The filtrate from the silica is evaporated to 
low bulk, excess of ammonia added and then made slightly acid 
with nitric acid. To the warm solution, about 70° C, add excess 
of acid ammonium molybdate solution and keep in warm place. 
Filter ofT yellow precipitate, wash with water containing nitric acid 
(about 10%) until ppt. is free from iron, dissolve in ammonia, add 
ammonium chloride and precipitate with ** magnesia mixture." 
Estimate the P^Oj in the usual manner. 

Lime and Magnesia. — These are determined in a solution of 
0.6 gram of slag from which the silica has been separated by 
evaporation to dryness. The A1,0,, PjO^. Fe,0, are separated by 
two precipitations by ammonia (sufficient iron must be present to 
ensure the precipitation of all the P,Oj. otherwise a known amount 

VOL. XXll.— XI. 


must be added) and the manganese separated by ammonium sul- 
phide, by fractional filtration, say 4CX) c.c. out of 500 c.c, and the 
lime and magnesia estimated by the usual precipitation as oxa- 
late and double phosphate respectively. 

FeO and FCgOj. — The total iron is first determined by titration 
with KMnO^ solution and another portion is taken for the estima- 
tion of the iron existing in the ferrous form, by decomposition in 
a sealed glass tube with sulphuric and hydrofluoric acid or by the 
decomposition of the slag by hydrofluoric acid and sulphuric acid 
in an atmosphere of carbon dioxide as described by Hillebraiid 
in Bulletin U. S. Geol. Survey, No. 176, p. 92. 

The presence of sulphides interferes with an accurate determin- 
ation of ferrous iron by any known method. 

Alumina can be determined in the precipitate of Al^O, + Fe,0, 
+ P^Oj, from the lime and magnesia portion. 

Manganous Oxide. — The manganese is determined as in Vol- 
hard*s method. 

Sulphur is determined also in the usual way. 

In basic slags chromium sometimes exists from the chromite 
bricks and as the percentage is small the method as described in 
the general scheme will serve for its estimation. 

Percentage Composition of Typical Iron Slags* 

Material. SiOg. CaO. MgO. Aip^. Fep,. FeO. ' MnO. , PjOj. S. Alkalies. 

Blast ' ^ '1 ' " I 

furnace 36.80 43.50 1.63 14.64 0-95 2.35 0.02 0.62' 1. 27 

slag. , 

^sfa^*^ 33-33 '-'9 050 5-75 5494 , 2.71 0.99 

Acid I 

Bessemer 47.27 1.23 0.61 3.45 '5-43 3189 o.oi I traces. 

slag. 1 

'^Cmder. 7-71 3-91 0-34 163 8.27 66.32' 1.29 8.07 1.78 1 

Basic I 

, 11.64 51.90 6.37 0.82 1.60 7.30 4.86 14.60 0.17 

* Taken largely from Percy's Metallurgy. 




South Dakota School of Mines. 

The work described in the following pages was performed with 
the object of establishing a satisfactory method of assay for the 
zincy precipitates obtained in the cyanide process, endeavoring to 
evolve a method at once accurate and short, which can be used in 
an ordinary custom assay office or mill. The difficulty of satisfac- 
torily and correctly assaying rich material containing much zinc, 
had been frequently noticed, and in view of the fact that the 
cyanide process is of widely increasing importance, and that the 
smaller plants but rarely possess arrangements for the refining of 
the precipitates, but dispose of them according to their assay value, 
a method giving rapid and accurate results would be of value. 

The precipitates used in the work were obtained from a 30-ton 
cyanide plant, using the MacArthur-Forest cyanide process on 
roasted chlorination tailings. The material was typical, containing 
42.3 % zinc in a finely divided state (some of it present as oxide), 
23.2 % SiO, (insol. residue), small quantities of Ca, Mg, Al, traces 
of Mn, Cu, Sb, and about 2,320 ounces of gold and 2,869 ounces 
of silver per ton. 

The sample used in the assays was obtained as follows : About 
thirty-five of the dried precipitates were thoroughly mixed on a 
large oil cloth, then spread until the mass had a uniform thickness 
of one-half inch. It was then marked oflfinto 2inch squares by a 
sharp spatula. All told there were about 250 2-inch squares. 
From each square, a small amount was then dipped, care being 
taken to go down to the bottom in each case, and to take the same 
quantity from each square. 

The amount thus taken, about one- fourth pound, was put into a 
glass-stoppered bottle. It is very difficult to get a uniform sample 
of material of this kind, so further care was bestowed on the prep- 
aration of the one-fourth pound taken out. The sample was put 
through a lOO-mesh sieve, and as but very little metallic zinc re- 
mained on the screen, this was also forced through. The sample 
was then thoroughly mixed for a long time. In taking out the 


amount used for an assay the procedure was as follows: The 
whole sample was spread out thin, divided into a large number of 
squares, and about one-half to three-fourth assay tons sampled out. 
This amount was then again spread out, divided into squares and 
the one-twentieth or one-tenth assay ton for the assay sampled out 
from this. This was repeated for every charge weighed out. All 
weighing was done on an analytical balance. 

Scorification Method. — Scorification was the first method tried, all 
the precautions recommended in the scorification assay of zincy 
ores being observed, such as a large proportion of lead to ore and 
low heat during scorification. The amount of precipitates taken 
was one-twentieth assay ton. See Table I. Charges number one and 
two and five and six were run with seventy grams of test lead and 
one gram of borax glass. Charges three and four were run with 
seventy grams of test lead and phe gram of borax glass, and fifteen 
grams of litharge as a cover. In every case the precipitates were 
mixed with half the lead placed in the bottom of the scorifier and 
covered with the rest of the lead. 

The slags and cupels were assayed in the usual way. 

Referring to Table I. it will be seen that the results are abnor- 
mally low when compared to the other methods. The slag correc- 
tions are very high. The slags obtained from the assay of the first 
slags undoubtedly still carried considerable values, but in referring 
to the corrections obtained from the crucible slags, in the same 
table, it can be seen that even if these corrections were made, the 
total result would be much too low. This points to an excessive 
loss by volatilization in the scorification method. The method 
was discarded as unsatisfactory. 

Crucible Meiliod. — The second method tried was the crucible 
method with the following charge : ^^^^ a. t. precipitates, i ^ a. t. 
litharge, l^ a. t. soda, ^ a. t. borax glass, 5 grams fluorspar, 2 
grams argol. 

The charge was made basic, so as to get a stony slag from which 
the button would separate, easily and cleanly. Table I., assays I. 
and II. show the result. The slag corrections are not high and re- 
sults also indicate volatilization. 

The following charge was then used, recommended by Wm. 
Magenau, E.M.* 

* Afiting' and Scientific P^ess^ April 28. i960 


y'^ assay ton precipitates, 70 grains litharge, 5 grams sodium 
carbonate (dry) i gram flour, silica 5 grams, 2 grams borax 

The results obtained by this charge are good but are apt to vary. 
Compare Table I., assays A, B, C, D, E, F. As will be seen by com- 
paring Table III., this method gives the highest average result, 
(corrected) on gold, but not the highest on silver. The uncor- 
rected results are higher in the combination methods. The loss 
in the slag is very variable, being very small in one case, and run- 
ning up to a considerable value in other cases. While the 
method in many cases gives the highest result it will again give a 
low result without any apparent reason, the conditions as regards 
charge and heat being in all cases the same. 

Combination Method. — The next method tried was the combina- 
tion wet and dry assay using nitric acid, y^ a. t of precipitates 
were treated with 20 cc. cone, nitric acid and 60 cc. of water. 
This was boiled gently for one hour, cooled and diluted to 1 00 
cc. with water. Then 75 cc. normal salt solution (5.4207 grams 
NaCl to the liter) were added, also 5 cc. cone. H,SO^ and 20 cc. 
of lead acetate solution. The precipitate was allowed to settle 
for an hour and a half, filtered, and washed to the point of the 
filter paper, dried and burned at a heat just high enough to ignite 
paper, and then scorified with 40 grams test lead. Buttons were 
cupelled in usual way. Table II., assays U,V, K and M will show 
the results obtained. The gold is far below that shown in crucible 
assays, even the corrected results. The silver is higher than in 
the crucible assays. This pointed so directly to the fact that gold 
passed into solution, that tests were made to demonstrate this. 
The filtrates from assays No. K and M having been carefully 
saved were treated as follows : They were evaporated nearly to 
dryness, expelling all of the nitric and nitrous acids, then were di- 
luted to ICXD cc. adding 2 cc. of cone, sulphuric acid and 10 cc. of 
lead acetate solution. Sulphuretted hydrogen was then passed 
through the solution for one-half hour. Some of the lead sulphate 
was changed to lead sulphide, the whole of the precipitate settling 
to the bottom readily. It was then filtered out, washed and treated 
in the same way as already described under the combination nitric 
acid method. The beads obtained were seemingly all gold and 
weighed 596.4 and 568.2 mgs. calculated to the assay ton. The 
above is the weight after parting, almost identically the same as 



'J9A|!S pu« p|o8 ptioi 

XvSSy p9)39JJ03 

vt ro m (*)^ ir. i/)M ei Qk»« n m 

♦ f*VO *o O 



pan pioif \K\o\ 
Xbs«v pd)3MM3ari 

M oo 00 00 c< ^ m« « c« o 


wm ^mm ««# *#N* VI -^ v«^^ V* %■* ^ y^-^ — • *# 

M Q QOO M<p lO^tomO ^M* 


paioauoo ai pioJB O) 

X»K\\% JO oii»H 

I^HVQOO «0000>O M tovO 
t«. ^ F. r«> F>«o 00 00 00 00 00 

« « « 

30 00 00 











o V 

a V s* 

Si" . 




"3 o^ 



-■SI! 8 . 
a Z. Au a 

3 V 9 B O 
o > o o *• 

< ^ e S. 

id . 

« 3 w 

o > 2 
< £ 2 o 5. 

ji a 9 - 

fc * o h 



' 6 . 

■ -o 


OS m ■♦ ■♦ «>5 »■ ovS>«nm*>^H 

• •■••••••>•••• 


mo ov 9«oo 00 n o t>«o fvG o t*« 


6 d d 6 6 6 6 6 6 6 6 6 6 6 

00 M 



SOS <*) fo r^sp 
H M H H M 

• • • 

OS <*) fo r^sp OOwOriHOO 

OsfoO r>.r*.mq q 

N c« M M e« M 

O 0\ H M 00 00 '^>C >C 00 M >o o o 
\0 O t^ tv%0 OOOOO OxOsO^^OvOn 


>0 I'to ^t^ moo 00 ov a« On Os Ov 

H lA ro O* » »«. r^ 





ci 00 d d d»oo ^'^^ ro rn r*« ^ o>od 

■♦ c*^ O m woo ■<• 


•5 00 »o>o O 0\ 








rii>.Mdddoo4-cJ ■♦\o too "♦ 
m «ovo r^oP»«tN>o«F.'* M 
vo ir> f, m «o r» 

q q q « 00 q vo 1- 00 q t^oo So 5> 
« Q 5"^ <> .5 « C'*^' od rn M o d 
m «o «o m »oo 


ri.»< t^r^OxMOO rors.o t^»-i if>.ro 
^o»« r^F»M r^o-oP*« 'oSs<*> 
vp «o ■♦■♦"♦ 1^ "♦ lAoO 00 r^oo 00 «0 

oO'^'mo ■♦« ciooeooo oo r^M 

r>oo w 

_ OvOO 00 
C4 »• M M 

tv M to 'i' f^ f^OO (^ Ov to d\ 
»oooo O O mr^N fnr'ipn 









«Os«oXK.^r^ e 



"2 8 




'Xvffsv JO J9quin|i4 

M ">«•<• o o « o « M e* onoo q « 

00 ♦ ir> N op f^o" N M CO rO Q ^ M 


00 cSOO ^0^^^ f>00 O r» r^ !>• 

o oo «o « f^^"^ f^. « ** ''loo 

too »oo' ft r'ioo eJ f^ ♦o W- ds « 
O « r^mKn N f*io (ioooo roM 

H « m 

♦ ui^-.«<CQUQWU. 

•uoii 'aiqiD **1<IP 


before parting, the only difference being that due to the silver in 
the test lead. 

The fact that gold goes into solution when treated with nitric 
acid, under certain conditions is well known.* It has been shown 
that this is due to the formation of nitrous acid f during the treat- 
ment, which readily acts on the gold. The conditions were very 
favorable for the action in this case, the gold being present in 
considerable quantity and in an extremely fine state of division. 
Whether the presence of finely divided zinc, more so than another 
base metal acted on by nitric acid, increases the formation of nitrous 
acid, and in this way the greater solution of the gold, the authors 
do not pretend to determine. 

By adding the amounts obtained from the filtrates the results in 
gold are brought up nearly to the average corrected assays of the 
other methods. (See Table II.) 

In order to overcome the difficulty of the gold going into solu- 
tion by the formation of nitrous acid, the fourth method, combina- 
tion wet and dry assay using sulphuric acid, was tried. The method 
of procedure was as follows : y^^^ a. t. of precipitates was treated 
with 20 cc. concentrated sulphuric acid and 60 cc. of water. 
This was boiled for one hour, cooled, diluted to lOO cc, 75 cc. 
normal salt solution, and 20 cc. of lead acetate solution added, the 
precipitate allowed to settle for an hour, then filtered, washed and 
dried. Care was taken to wash the precipitate into the point of 
the filter. The paper was burned off at a very low heat, and the 
residues scorified until half covered over in the scorifier, then 
poured and the button cupelled. Thirty to forty grams of lead 
were used. Table II., assays i to 8 inclusive, will show the result. 
As regards the gradual decrease in value, as shown in the column 
"corrected assay-total gold and silver," reference will be made 
later on in this paper. 

The results are good and quite uniform, as uniform as can be ex- 
pected with material of this kind. The gold is lower than in the 
crucible assay, while the silver is higher. The slag corrections in 
this method, as also in the nitric acid method, are small (relatively), 
as is to be expected. In order to see whether any gold passed into 
solution the filtrates from assays 3 and 4 were treated as described 

* T. K. Rose, Metallorgf of Gold, and references, pp. 478, 479. 

f W. R. Van Liew*s paper, E, and M, J.^ April 21 and April 28, 1900, pp. 479, 




'j9A[i« 01 p[o9 JO otie^ 

p« €1 — 00 5- O « Q 
00 00 00 F^ ^^0O 00 OB 

d c" d d d d d 

8 = 

00 oo 

d d 

-OO) jad 

S93uno Of '3j«Jt[g 

OJOJJ pai3A033J pto{^) 



C e8 V 


> - m 3v "♦nS-rwnn 1 onoox 

:= "*• »' d <-* -" d ».' «■ M -■ I M M d d 

^ I _ ^ 

•5 . O0O^f^ff^f^»<»<«, o-oB 00 1^ 

O d d d d d d d d d ' d d d d 

tj «s 

i . e« >c t^ <*-. c^'0 o o M ■<• f^ "♦ rnvc 



©■^ qoqwqi-qO'^HOOO'n 
C 6 6 6 6 6 6 6 6 6 6 6 6 6 6 






o t; 

be— o >, 
* t» *» rt 

t S 

E5i = 



> ^ 2"°" °o 2" 2*00 t>»oo r^oo 00 Qi r^vo 
— IF- ©■ ^©iOiCT>0>0»^0*0*0\&iONO 








•jaAjts pUB pioif [BJOJ^ 
Aussy pajDaojo^ 

'JdA|IS pUB pioJJ [BJOX 



a .t 

t 5 3 

O <'' w 











*; M 5» t^oo O" 5^ ?* 0\ •'lOO W O PO 
O &>o>avO\CT>&>pn r^oo Ov F» r^ 

O p O O 

wl O O N 

K-t>.e« d'od 6 into* 


PO O 00 

c^ to »/^ <7\ 

OQOpo•/^O^^u^l 00 

pj M rn f^ M pj ej fi (," I (j (1, m h 

^« q q\ ro N u^vo o o « 3- o ■♦ a> 
r^vd r*>« M ^M M rl dvd\t^d>e« 

2 00>0POO'^i^Q«00O0 



•o >^ o 
sqoo c?iwoo w^fnO ►« "^:^p^ row 

'TsO <>\6 f^t^ir)U-iAt^c r^ 1^00 


q po poso Of^o»^wi*twN ^00 

tT) jv u-i\o •• ir, ro M d-, d d- « r^vd 
000 r<^ N m m On On ■?^oo 5^ 00 ^ 



p^ ^ m o f^ moo 

lO o 
m r^oo 



* o ^:^ ON ■♦o OS in 

C "^ Q:<^ * 

O \rt xTiCO 00 


IT) to 

t». On P» 

\0 '-^ M 00 00 M N 

m rn m M 

« M N N 

Qs a ' 

N W « 


^ iv pn t"» ^ moo pn M 00 w N o 00 

p^N M ^^c« r^C t^w mn-O »^ 
00 00 00 00 oo 00 t^oo t^oo 00 oo 00 t^ 

•"C >c q q^o q « ^q »pim^*o 
>-« fi d 00 <•• d sd oo' 

vC vC N 1^00 M «« 00 

rr rr rr, fi ft m m r< 


"^ m ^00 ^ 
■c vo ''^ t^ 

►• « Ci H « 

XOTuy JO jaqu]n\ 

H N 




m •♦ in-O tvoo e»^Dr*»"^S 





under the nitric acid method, but only a trace of gold found. 
Whether the precipitation is complete by the method employed, is 
not known. The fact that the gold (corrected) obtained by the 
sulphuric acid method is somewhat lower than that obtained by 
the crucible method, may point to the fact that some gold goes 
into solution by the sulphuric acid treatment,* which cannot be re- 
covered by the method used. That it cannot be recovered by sul- 
phuretted hydrogen appears doubtful. 

The fine state of division of the zinc in the precipitate renders it 
very liable to oxidation. This oxidation of the zinc, causing an 
increase in weight, will cause changes in the gold and silver values. 
To test this point, after the first assays were made, the sample was 
spread out and allowed to remain in the air in a dry place for five 
days, when the second set of assays were made. It was then again 
exposed for one week and the third set of assays made. Set No. 

1 is represented by sulphuric acid method Nos. i and 2. Set No. 

2 by sulphuric acid method Nos. 3, 4. 5 and 6, and crucible method 
Nos. A, B, C and D. Set No. 3 is represented by sulphuric acid 
method 7 and 8, and crucible method Nos. E and F. The result is 
most clearly shown by the sulphuric acid method assay in Table II. 
column marked " corrected assays total gold and silver," the aver- 
ages being as follows : ist set, 5,245 oz.; 2d set, 5,184.3 oz.; 3d set, 
5,144.7 oz. 

These facts lead to the conclusion that the sample should be 
assayed at once after taking or kept in such a way that it will not 

Comparison of the CruciMe and Sulphuric Acid Method, — It will be 
noticed in looking at Table II., columns •* uncorrected assays total 
gold and silver',' and *• corrected assays total gold and silver," that 
the results check quite closely for the kind of material, while the 
gold and silver by themselves in but few cases check even reason- 
ably well. The results must be considered as sets (marked off by 
bracket) for the reason mentioned on the preceding page. This 
comparatively close checking of the total sum of gold and silver 
is not apparent in the crucible assay. The sulphuric acid method 
seems to give more uniform results than the crucible method, al- 
though the gold by the crucible method is higher. The difference 
is six ounces gold per ton, not great, at least not great enough 

♦T. K. Rote, Metallurgy of Gold, p. 11. 


tot warrant the assumption that the crucible method is superior 
The crucible method is apt to give erratic results, without any ap- 
parent reason (assays C and E, Table II.), and to get reliable re- 
sults more than two assays on a sample are necessary. The sul- 
phuric acid method, on the other hand, will give a reliable result on 
a duplication. 

The slag absorption in the crucible method is very variable, 
while that in the combination methods is first much smaller, and 
second fairly uniform. In fact for precipitates from a given mill, 
which do not vary widely a constant might be established with 
fair accuracy, or a correction made on one, which could safely be 
applied to all the assays. This cannot be done in the crucible 

The cupel absorption is about constant in all the methods. 
Slightly higher in the crucible, probably on account of some zinc 
still retained in the lead button. 

Conclusion and Remarks. 

Regarding the fact that the sum of the gold and silver, found 
by the sulphuric acid method, checks fairly closely in a given set 
of assays, while the gold and silver vary considerably, the writers 
wish to offer the following solution : The tailings from the treat- 
ment of which the precipitates were obtained vary considerably in 
value, particularly does the ratio of gold to silver vary in different 
lots. Hence the solution containing the gold and silver will also 
vary in gold and silver contents as it passes through the zinc boxes 
from day to day. The zinc boxes which are cleaned up but once 
a fortnight, would then contain a precipitate, consisting of an 
infinite number of small particles, many of which are made up of 
relatively different amounts of gold and silver. These particles 
are too small to be broken up by passing the ordinary mesh sieve, 
and pass into the sample unchanged in constttutton. By careful 
mixing the sample may be made to contain a given number of 
these particles per unit of volume, and if assay were made on units 
of volume the results of total gold and silver would not necessarily 
check, as on account of the variable ratio of gold to silver, the 
weight of the volume would vary ; but since the assays are made 
on a unit of weight the total weight of the particles is the same in 
each assay while the volumes would vary. This would explain 
why the results check fairly close in total gold and silver, but not 








'j»A|tff 01 pjofl JO ope^ 


00 I tV 






< — 


' rr^ 

lo O 

M <0 Ov^ 

M M M 


5; . . ■♦ •• fp o 

I "o >(l. 



M 00 


?■? S 

• • • « 

O *• en 

n tn m o 



3*- SI'S. I 'a 




'jdAijs puv p|o8 (BlOX 

'98«J9Ae pa)39Ji03UQ 
'J9AIIS pUB p[Oa \VlOJ^ 


fj mo 


!5 m ? m « 

0> ■<• N O 

vO 0(" i-i 
M f»» d ro 

• ■ • • 

trrr « 

• • •*■ • I 

tv o «« 

*• 3 «*) 
10 m ^ mt 


»c O 

0> ID N 10 

00 00 00 10 

N C* « Ci 

:? s 

m>0 2oo 

CI ci ! 




O m O 1*», 
• • • I 

»• 00 Ml «D 
N ■<• HOO 
00 t*«00 CT> 
« M « M I 

«A M 

X 0\ i^i K 

« « M M 



2§ . 


in gold and silver taken alone, as the particles are made up of differ- 
ent amounts of gold and silver. A column has been calculated in 

'• Au 
the tables, -^ ratio of gold to silver," which shows the vari- 
ation in the ratio. It will be seen that the ratio varies in the dif* 
ferent methods of assay, as is to be expected. In a given method 
of assay, however, it would seem as if it should be constant if the 
above premises are not true. 

It is of interest to note that relatively the loss of gold to silver is 
greatest in the scorification assay, and least in the crucible assay. 
The difficulty in getting an even sample with cyanide precipitates 
has been frequently noticed and commented upon ; the preceding 
explanation probably gives the reason. It is also claimed that the 
solubility of silver and gold by cyanide solution is relatively differ- 
ent at different temperatures.* In plants where the leaching vats 
are exposed to the varying temperatures of day and night which 
differ much in some localities, this fact will have the same effect 
on the gold and silver contents of the solution as a difference 
in the values of the ores. 

The losses in slags and cupels have been completely worked 
out, not to present anything new but to add to the literature of the 
subject. The loss in the cupel is fairly constant, being almost 
wholly determined by the physical nature of the cupel and the 
heat of cupellation. 

All beads of sufficient size were parted as in the bullion assay in 
the form of cornets. 

In conclusion, the authors would recommend as the method of 
assay for the precipitates, either the combination sulphuric acid 
method or the crucible method, preferably the former. The sul- 
phuric acid method to be performed as described in the preceding 
pages, using dilute acid I to 3 or i to 4, and always keeping the 
acid dilute. Nitric acid even in small quantities with the sulphuric 
acid is not to be used.f (Also see assays A and B, Table II.) 

The slags and cupels are best reassayed and the corrections 
added ; this can be modified as suggested under the combination 
sulphuric acid method. 

Every assay made has been placed in the tables. 

♦ T. K. Rose, " Metallurgy of Gold," p. 357, 1898. 
t T. K. Roic, •« Metallurgy of Gold," p. 1 1, 1898. 




The author desires to acknowledge his indebtedness to Admiral Melville, Engi- 
necr-in-Chief, U. S. Navy, through whose wide knowledge of the resistance and 
powering of ships and generous collaboration, it has been possible to present in this 
paper the results of the best modern practice. 

The determination of the amount of resistance which a pro- 
posed ship will meet in her passage, at wave-making speeds, 
through the water, forms a problem in the solution of which 
theory, unaided, fails. While, with familiar types, it is possible to 
estimate in fair approximation, this amount, the working formulae 
used for the calculations are but broad generalizations of theory, 
reinforced by empirical and experimental factors derived from past 
experience. Speed and power curves, drawn from the perform- 
ance of similar ships, are an invaluable aid in designing and tank- 
experiments, with a model of the proposed vessel, give some indi- 
cation as to her resistance in smooth water; but the final trial, at 
load draught on the open sea, is required for absolute determina- 

The conditions of the problem naturally exclude theory except 
in a broad way. A ship has not *• the form of least resistance," 
but the approach to it permitted by her load, stability, structural 
strength, and possible limitations as to length or draught. Fur- 
thermore, the hull whose shape has been moulded by these re- 
quirements and by the demand for a given maximum speed with 
economy of fuel, floats but partially immersed, in a fluid whose 
frictional qualities and surface-disturbance oppose to her motion 
an ever-changing resistance with each change in speed. 

The problem of ship resistance is, then, one whose widely vary- 
ing factors no absolute theory will ever fully cover; and yet, for a 
century and more, many able men have endeavored to formulate 
laws which would express mathematically their conception of the 
action of a mass of water when traversed by a ship-formed solid. 
The results, to this time, of the labors of these theoretical investi- 
gators with regard to wave-making resistance, have been long and 
complex formulae which have no practical value. Since the path 

♦Copyrighted by the School of Mines Quarterly. 


of a particle of water which passes within the range of influence of 
a ship's hull, is not fully known, the interactions of pressure between 
it and the hull cannot be determined with accuracy. Furthermore, 
the wave-form used in the most modern of these calculations, is 
admittedly assumed ; there is marked discrepancy in the views ad- 
vanced with regard to the final derivation of the energy expended 
in the maintenance of wave-systems ; and the formulae apply, at 
best, only to absolutely smooth water, a condition which is practi- 
cally non-existent with sea-going ships. 

In view of these limitations, the results of the labors of these 
investigators must be received tentatively by the engineer of ex- 
perience. The difficulty has been the relative lack of experiment, 
as compared with the vast volume of speculation and theory. It 
is a truism to say that mathematical processes are of fundamental 
importance to the engineer in the analysis and development of his 
work ; but, in the solution of complex physical problems and in | 
the estabishment of laws their functions follow, not precede, ex- 

Among the men whose rational or experimental researches have 
led to the widely accepted theories of to-day, there are two of 
foremost rank — Professor Rankine, who was the leader in the 
extended development of the stream-line theory as applied to ship- 
resistance, and Mr. William Froude, whose experiments on sur- 
face-friction and determination of the relation of model resistance 
to that of the corresponding vessels, marked advances of cardinal 

I. Analysis of Resistance. 

Total Resistance. — In a broad sense, ship-resistance has but 
two factors — the friction between the water and the wetted skin 
and the surface-disturbance of the former which begets waves. AJ! 
low speeds, in smooth water, the ship moves with ease, friction 
generating practically the only resisting force. At high velocities, 
however, there appears at her bow a wave-crest, from which trail off 
oblique, diverging waves. At the stern a similar crest exists, with 
a similar following system. Along the hull there is formed a troup 
of transverse waves with crests athwartships. She is followed by 
a ** wake " or current, constant in form for the same speed, although 
ever changing its constituent particles, from which wake extends 
forward an envelope of varying thickness over the immersed skin. 


Finally/and at the stern chiefly, a body of eddying or "dead 
water " exists, the size of which depends upon the formation of the 
stern. In the production of waves, in overcoming surface friction, 
and in the transport of the latter's partial derivative, the wake, re- 
sistance is met, work is done, and power is diverted from propul- 

Factors of Resistance. — Disregarding, for the present, the 
surface-disturbance and augment of resistance due to the screw- 
propeller, the ** tow-rope resistance,"/.^., that of the ship if moved 
by forces external to herself, may be divided broadly thus : 

1. Frictional Resistance, — Water has not perfect fluidity and the 
surface of the immersed skin is not theoretically smooth. The 
latter, therefore, meets resistance to its progress and power is 
expended in giving, to the contiguous fluid particles, force and 

2. Eddy Resistance. — As will be later shown, a sudden termina- 
tion of the easy curves of the '* run ** or •* after-body ** will permit 
a portion of the passing water to leave its, so-called, stream-line 
paths and to dissipate, in irregular, whirling motion of the parti- 
cles, pressure which should act upon the hull, at the run, in oppo- 
sition to that at the " entrance " or " fore-body." There is thus 
in the maintenance of the dead water, a loss of power, which, 
however, in modern ships of fine lines, is small. 

3. Warce resistance, — Waves are generated by an upheaval of 
the surface-water. The power thus expended is, in large degree, 
lost, since some of the waves so formed become wholly separated 
from the ship, bearing with them the energy spent in their pro- 

2. Stream Lines. 

I. A FrictwnUss Soltd in a Perfect Fluid. — The stream-line theorj 
refers primarily to the relative and steady motion of a perfect fluid 
and a frictionless solid deeply submerged therein. While water is 
not a perfect fluid, the wetted skin is not frictionless, and the hull 
is not submerged, but is only partially immersed, the application 
of the theory to the phenomena of ship-resistance meets wide ac- 

Consider first an example in full accord with the theory. /. e.y a 
frictionless ship-formed solid, as A-B, Fig. i , submerged in, and 
passing with uniform, horizontal, rectilinear motion through, a 



perfect, incompressible fluid, infinitely extended about it in ail 
directions, this " perfect fluid " being understood to be one without 
viscosity, whose particles glide over the surface of the solid or 
move among each other without frictional interference. 

Fu;. I. 

2. Stream Lines, — For clearness, regard the solid as at rest and 
the fluid as flowing past it, as above — which assumption makes 
no change in the relative conditions. Then, the paths of bodies or 
** streams " of particles, at a distance forward from the bow, will 
be parallel to each other and to the major axis of the solid. As 
the streams draw near the bow, they curve outward, their axial 
speed is checked, and their bounding lines separate. In receding 
from the bow, these lines approach each other, until, at the mid- 
ship section, the stream is narrowest. From that section stern- 
ward, the stream, reversing its former action, first widens, then 
contracts, and. at a distance from the solid, resumes its normal 
width and rectilinear flow. 

Now, two adjacent lines may be considered as forming a 
" stream-tube," in which the same particles, and no others, are con- 
fined and flow, while approaching, passing, and retreating from the 
solid. The assumed boundaries form thus a frictionless tube 
whose cross-section, before the stream comes within the influence 
of the solid, is uniform throughout, varies during that influence. 


and returns to its original uniformity when again beyond it. 
Further, as steady motion of the whole body of fluid is prescribed, 
such variation in section will produce, in inverse ratio, changes 
in the velocity of the stream. 

3. Stream Line Pressures and Velocities, — For such an imaginary 
tube, Bernoulli's theorem may be stated thus : 

*'In steady flow, without friction, the sum of the velocity-head, pres- 
sure-head and potential head, at any section of the pipe, is a constant 
quantity. " * 

In accordance with this statement, the equation for '^steady 
flow " may be written : 

v" p , 

1- <- + z^h, 

2g w ' 

in which, z/a velocity at any section,/ = internal fluid pressure 
per unit area at said section, ze/s weight of unit volume of fluid, 


z (potential head) » height of tube above a fixed plane, 

velocity-head) =■ height due to the velocity v, ~ (pressure-head) 

= height due to pressure /, h (total head) = constant, as above 
for each stream-tube. 

Inspection of this equation shows, that as horizontal flow is 
assumed, ir is a constant ; and since h also is a constant, that, when 
V increases, / must decrease and vice versa. At the bow and 
stern, the section of the steam-tube is large and, therefore, the ve- 
locity is low and the pressure in excess. At the midship section,^ 
these conditions are reversed and the pressure falls below the nor- 

4. Stream line Forces, — Horizontal forces, varying in intensity 
and direction, act, therefore, from the stream upon the solid. After 
leaving the latter, however, the stream resumes eventually its 
original velocity and direction. Its motion has been, for a time, 
modified, but its energy remains unchanged. Therefore, the op- 
posing horizontal forces, as at bow and stern, neutralize each 
other, and the fluid has no resultant action upon the solid. 

If then, as in the original statement of the case, the solid were 
in motion, under the given conditions, and the fluid were at rest, 
the former would have no resultant action upon the latter. 

* Mechanics of Engineeriiig, Church. 

VOL. XXII— 12. 


Furthermore, since the fluid is extended to an infinite distance 
above the solid, there would be no waves, and, since both solid 
and fluid are frictionless, there would be no skin-resistance and no 
wake. In other words, such a frictionless solid would pass, at 
uniform speed in a rectilinear path through a perfect fluid, without 
resistance, after motion has been established. 

5 . Partial Immersion, — Assume now that the solid moves, partially 
immersed, upon the bounding surface, as a water level, of a perfect 
fluid. Then, there exists no longer the superincumbent pressure 
of an infinitely extended fluid upon the stream-tubes ; and, in Mr. 
William Froude's words : 

"The existence of this surface cuts off the reactions of all those particles 
nvhich would have existed beyond the surface, had the fluid been un- 
limited alike in all directions. * * * By the absence of these reactions, 
(the stream -like motions which would have existed in the infinite fluid, are 
modified; and the differences of pressure involve corresponding local 
elevations of the surface of the water (or fluid) in the vicinity of the mov- 
ing body. And since, in consequence of the action of gravitation (the 
force which controls the surface), a water-protuberance seeks immediately 
•to disperse itself into the surrounding fluid in accordance with the laws 
of wave- motion, the local elevation partly discharges itself along the sur- 
face by waves which carry with them the amount of energy involved in 
their production.*' * 

In other words, the atmospheric pressure at the surface is con- 
stant and relatively ow ; the pressure in the stream-tubes is, at 
certain points, abnormal ; and that increased pressure seeks equi- 
librium by an upheaval of the fluid and the formation of wave- 
crests, primarily at bow and stern, with, conversely, a wave-trough 
at the midship section. 

Since the fluid is still assumed to be perfect, no skin-friction, as 
yet, exists. It will be seen, then, that wave-making resistance 
should be difTerentiated broadly from all other forces opposing a 
ship's progress. If now, water — which is not frictionless — be sub- 
stituted for the fluid, and the theoretrically smooth solid be re- 
placed by a hull with an immersed skin more or less rough, the 
ship's movement will produce surface friction, eddies and a wake 
will appear, and the conditions of practice will prevail. 

In this necessarily brief discussion of stream-lines, a horizontal 
flow for the latter, in passing the hull, has been assumed. This 

* Encyc. Brit., 9th Ed., Art " Shipbuilding." 


assumption has been made for simplicity only and is not in accord 
with the views advanced by Professor Rankine. The absence of 
exact knowledge as to the direction of that part of the flow con- 
tiguous to the ship, does not affect, however, the application of 
the broad principles given above as to the action of the stream, 
in view of the final and practically unchanged energy of the latter. 
The uncertainty as to this direction of flow illustrates the fact 
that the stream-line theory is a structure which has been reared 
by rational process upon the basis — for full sized hulls, at least — 
of a minimum of experiment. 

3. Frictional Resistance. 

/. Amount, — In 1871, by direction of the British admiralty, Mr. 
William Froude, carried out a series of experiments upon the re- 
sistance of H. M. S. Greyhound^ a screw sloop of 1,161 tons, the 
ship being towed in such a manner as to pass only through undis- 
turbed water, thus eliminating many sources of error. In accuracy, 
completeness and importance these experiments have remained 
unequalled in their class. In Fig. 2, which is taken from Mr. 
Froude's paper * on the subject, curves are shown whose abscissae 
are speeds in feet per minute and whose ordinates are the corre- 
sponding resistances in pounds. The curve A-A represents the 
total resistance of the ship, as measured by ordinates from the 
curve C-C, while the curve B-B shows her frictional resistance, as 
measured also from the curve C-C. The close approximation of 
the curves A-A and B-B at low speeds, indicates that the resist- 
ance, under such conditions, is almost wholly frictional ; while, at 
high speeds, it will be seen that although wave resistance is a large 
factor, the percentage of frictional resistance is still very consid- 

2. Character, — The power absorbed in frictional resistance is ex- 
pended in impressing on the particles of a belt of water adjacent 
to the wetted skin, a force in the direction of motion of the latter. 
In each unit of time, there is thus given a definite amount of new 
momentum to fresh particles, which momentum expends itself partly 
in frictional eddies within the belt thus affected and partly in 
advancing the belt with the skin, as seen in the wake. The consti- 
tution of the wake or current thus created, changes momentarily, 

* Tfant. inst. Nav. Arch.^ 1874. 











y^ft j^Hr-T I I I I I I ■ I ■ I I I I I I I I I I m-i 

v^#n ^^^^' I ' ' ' ' I ' ' ' 'V' ' ' ' ' ' ' "I 




Fig 2. 


new particles being constantly set in motion to replace those 
whose momentum is dispersed by communication with the mass of 
adjacent water astern. The weight of water thus transported by, 
and with, the ship in the wake and frictional belt has been esti- 
mated as reaching, at high speed, 20 per cent of the displacement. 
3. Factors of Frictional Resistancc-^For accurate and trustworthy 
analysis of the factors of frictional resistance, we are again indebted 
mainly to the researches of Mr. William Froude. In the Report of 
the British Association for 1874, he gives the results of experiments 
with a series of boards, 19 inches deep, -^^ inch thick, of varying 
length up to 50 feet, and covered with materials of differing frictional 
qualities, from tin-foil to coarse sand. While the boards were 
practically but thin planes, care was taken to provide each with a 
fine entrance and a fair run. The resistances of these prepared sur- 
faces, when towed lengthwise in a tank of fresh water, are given 
in the following tabular statement, as to which Mr. Froude ob- 
serves : 

** This (table) represents the resistance per square foot due to various 
lengths of surface of various qualities, when moving with a standard 
speed of 600 feet per minute, accompanied by figures denoting the power 
of the speed to which the resistance, if calculated for other speeds, must 
be taken as approximately proportional. Under the figure denoting the 
length of surface in each case, are three columns, A, B, C, which are 
referenced as follows : 

* ' (A) Power of speed to which resistance is approximately proportional. 

*' (B) Reftistance in pounds per square foot, of a surface the length of 
which is that specified in the heading — taken as the mean resistance for 
the whole length. 

*' (C) Resistance per square foot of unit of surface at the distance stern- 
ward from the cutwater, specified in the heading." 

Table I. 

I^ength of Surface, or distance from cutwater, in feet. 





of Surface. 

2 feet. 

8 feet. 

ao feet. 

50 feet. 




A i 

B 1 



B 1 



B • 




0.41 ' 





























I 90 



















f'ine sand. 













Medium sand. 













Coarse sand. 


1. 10 









4. Resume. — In summation of Froude's results, it will be ob- 
served that : — 

{a) The exponents given in column A show that friction resist- 
ance varies approximately as the square of the speed. Since, 
however, a varnished surface, 50 feet long, is the nearest approach, 
given by the table, to the surface and length of the wetted skin of 
a ship, the exponent for these conditions indicates that frictional 
resistance varies usually as the 1.83d power of the speed. 

(^) From the data of column B, it appears that the longer the 
surface, the less its average resistance per square foot. As to the 
value of increased length in thus reducing the mean resistance, 
Mr. Froude says : 

"The portion of surface that goes first in the line of motion, in ex- 
periencing resistance from the water, must in turn communicate motion 
to the water in the direction in which it is itself travelling. Conse- 
quently, the portion of surface which succeeds the first, will be rubbing, 
not against stationary water, but against water partially moving in its 
own direction and cannot, therefore, experience as much resistance from 

{c) With regard to surfaces exceeding 50 feet in length, Mr. 
Froude states : — 

** At a length of 50 feet, the decrease (with increasing length) of the 
friction, per square foot of every additional length is so small, that it will 
make no very great difference in our estimate of the total (frictional) re 
sistance of a surface 300 feet long, whether we assume such decrease to 
continue at the same rate throughout the last 250 feet of the surface, or 
to cease entirely after 50 feet." 

{d) It appears, then, that the frictional resistance of a ship de- 
pends upon the area, length, and quality of the wetted surface and 
varies as a certain power of the speed ; or 

R, =fAS\ 

in which, R^ = frictional resistance in pounds, A = area of wetted 
surface in square feet, 5= speed in knots per hour, n^ under ordi- 
nary circumstances, = 1.83, and/ is obtained by calculation from 
Froude's results as given in Table I. 

5. Tlie Wake, — ^The formation of this following current, while 
due very largely to surface-friction, is also affected, to some ex- 
tent, by eddy-resistance and at high speeds by wave-formation. 


The interaction between the hull and wake will be hereinafter 

4. Eddy Resistance. 

/. Cause, — Let the stern of the ship-formed solid shown in Fig. 
i» instead of tapering easily with the run, be ended abruptly as at 
C\ and let the solid be considered to be moving through water 
and to be partially immersed therein. The stream-line theory 
assumes a pressure above the normal to exist at both bow and 
stern. The formation of the stream-tube is due to the passage of 
the ship and causes an expenditure of energy at, and a resisting 
pressure upon, the entrance, which pressure, under perfect and 
theoretical conditions, is assumed to be balanced by the corre- 
sponding forward pressure in the stream at the run. Now, if the 
latter be terminated suddenly, as at C, the stream leaves the hull 
while forward pressure still exists within it, which pressure, instead 
of aiding to neutralize the entrance resistance, dissipates itself in 
the creation of a mass of eddying water directly astern, whose con- 
stantly changing particles whirl in involved paths and whose pres- 
sure is less than that of the normal stream- line at that point. The 
production of the "dead water" means, then, an expenditure of 
power uselessly in eddy formation, with, as a sequence, a decreased 
forward pressure on the run. 

2. Amount, — Mr. William Froude, on the basis of the Greyhound 
experiments, assumed eddy-resistance to be proportional to that 
due to friction and to be about 8 per cent, of the latter for ships of 
fair form. This percentage was, however, estimated, since this fac- 
tor of the total resistance was not determined separately. In mod- 
ern vessels of fine lines, the amount of resistance due to this cause 
is regarded as almost inappreciable, although past experience is 
heeded in the sternward tapering of such appendages as strut-bear- 
ings and the like. 

3. Effect, — The primary effect of eddy-making resistance is an 
increase in the power required for propulsion. When it exists in 
considerable amount, as in vessels with full lines aft, it has secon- 
dary consequences of even more importance, in that it affects the 
steering of the ship, since the rudder not only acts in eddying 
water but the irregular formation, disappearance, and change of lo- 
cation of the latter, from side to side of the stern, tend, in them- 
selves, to deflect the vessel from her course. 



5. Waves. 

I. Wave-motion is, in ap- 
pearance, the onward move- 
ment of great volumes of 
water, whose advance, ex- 
cepting in a confused sea, 
proceeds in long, straight 
ridges of the same height 
and equally spaced. As a 
mass, however, the fl uid 
does not travel ; that which 
is transmitted is simply the 
*• wave- form," with the con- 
sequent grouping, to a cer- 
tain depth of the particles. 
Each of the latter revolves 
in an orbit of limited ex- 
tent, whose center is near 
the position of repose of 
the particle. The locations 
which the latter momen- 
tarily occupy in their re- 
spective paths determine 
the distances from a given 
point of the nearest crest 
and trough at that instant. 
The top of a wave is called 
its crest; the trough is the 
hollow between two suc- 
cessive crests ; the Iteight is 
the distance, vertically, from 
trough to crest ; the length 
is the distance, in the line 
of advance, from trough to 
trough or crest to crest; 
the period is the time, in 
seconds, taken by a crest or 
hollow in traveling a wave- 

length; and the veloaty, or speed of advance of the wave-form, 





is the quotient, in feet per second, of the length divided by the 

2. Trochoidal Theory of Waves, — ^This theory is accepted as giv- 
ing the closest approximation to actual conditions which is capable 
of mathematical analysis. It assumes the profile to be trochoidal, 
y. ^., the curve described by a point moving uniformly in a circular 
path whose center travels uniformly in a straight line. Thus, in 
Fig. 3 » let the distance ^4^/^^^ = a wave-length = the circumference 
of the " rolling circle," A^B, which circle revolves uniformly, while 
its center moves with uniform speed in the line C-C, Then, if Pbe 
a point on the radius of the rolling circle, it will fulfill the condi- 
tions of motion, as above, and will describe the trochoidal wave- 
profile, P-P-P, shown in full lines. It is assumed, further, that, in a 
deep-sea wave, the particles of each layer revolve, with uniform an- 
gular velocity, in equal, vertical, circular orbits, whose centers are 
on the same horizontal line, and whose plane is perpendicular to 
the wave-crest, and that each particle completes one revolution 
during a wave-length. Referring to Fig. 3 : 

*' The upper plain wave-line marks the surface of a wave at the instant 
when the crest is passing below the point A^ and the two troughs below the 
points Aq and A^, The po'nts marked P, -P, P, etc., represent nine particles 
in that surface, which, when the water is still, are equidistant from each 
other and situated in the level line W- IF, directly below the nine points 
Af^fA^f etc. The orbits of those particles are represented by nine dotted 
circles, described about the centers marked C, 

"After the lapse of one eighth of a wave-period, each particle has 
moved through one-eighth of a revolution, or 45 ° of its orbit, as shown 
by the short, curved arrows. 1 he points of those arrows represent the 
new pos tions of the particles, and the dotted wave-line traversing those 
points shows the new position of the wave, which has travelled forward 
through one-eighth of a wave-length, its crest being now below the point 
A-, the following trough below A^, and the preceding trough below a 
point one-eighth of a wave-length in advance of A^. 

** The lowei plain wave-line represents the first position of a sub-surface 
of equal pressure, which, when the water is still, coincides with the level 
dotted line, w-w. The nine points marked Q represent nme particles in 
that surface ; their orbits are represented by small dotted circles ; and 
their motions, although performed in smaller orbits, exactly correspond 
and keep time with those of the particles marked P, which are respec- 
tively above them. 

** The slightly curved lines marked P-Q^ represent originally verlical 


columns of particles and show how those columns bend and sway with 
the wave-motion, so as at any given particle, to lean in the opposite di- 
rection to that in which the wave-surface slopes. The curved and dis- 
torted quadrilateral figures enclosed between them and the wave-lines, 
show the slopes successively assumed by a series of equal and originally 
rectangular b'ocks of water.'** 

It will be observed that the distance between the center-line of 
the orbits and the still-water level, the diameters of the orbits, and 
the consequent velocities of the particles are gradually reduced with 
descent. At a considerable depth, therefore, depending upon the 
wave-length, a sub-surface is reached where the size of the orbit 
is inappreciable and the water is practically undisturbed. The law 
governing this decrease is expressed approximately thus: — 

** The orbits (in diameter) and velocities of the particles of water are 
diminished by one-half for each additional depth below the surface of the 
water, equal to one-ninth of a wave-length.*' 

The arrows of direction marked on the orbits show that on the 
crest the particle is advancing with the wave-form, while in the 
trough it is retreating. The mechanical energy of the wave is 
divided into equal kinetic and potential factors, one-half consisting 
in the motion of the mass and the other in its elevation. 

3. Wave Formulce, — The formulae for deep-sea waves are as 
follows : 

Length of wave in feet = 5.123 x square of period in seconds ; 

Speed of wave in feet per second = 5.123 x period in seconds; 

Period oi wave in seconds = \/length x 5- 123; 

Energy of a wave-length, one foot broad, in foot-pounds = £ = 

8Z^//» (i —4.935 ,2 ), where //= height and L = length, both in 


Shallow Water Waves, — The investigations of Professor Rankine 
show that in shallow water the orbits of the wave-particles change 
from circular to oval forms with the major axis horizontal, since 
the lack of sufficient depth interferes with the maintenance of the 
deep-sea system and reduces both the length and velocity of the 
wave with regard to its period, this reduction depending upon the 
ratio of length to depth. The oval orbits become smaller and 
flatter with descent, until, at the bottom, the particles reciprocate 
in straight lines. Since the higher the speed of a given ship, the 

* Shipbuilding : Theoretical and Pracacal, Rankine tt al. 


greater the length of the attendant waves, the resistance due to 
shallow water increases very rapidly with the speed to be main, 

6. Wave-Resistance, 
I. Wave-Systems. — The wave-systems produced by the bow of 
a ship, when driven at high speed, are shown in Fig. 4.* The 
latter was prepared from exact measurements of the actual wave- 
features as seen during model experiments. For clearness, the 

vertical scale of the profile visible against the side of the model is 
twice the horizontal scale. With regard to this diagram, Mr. 
William Froude says : 

"The inevitably widening form of the ship at her entrance throws off 
on each side, a local, oblique wave of greater or less size, according to 
the speed and to the obtuseness of the wedge, aud these waves form them- 
selves {en ichelon) into a seiies of diverging crests. * * * These waves 
have peculiar properties. They retain their identical size for a very 
great distance with but little reduction in magnitude ; but the main point 
is that they become at once disassociated from the model, and, after be- 
comirg fully formed at the bow, they pass clear away into the distant 
water and produce no further effect on her resistance. 

Beside these diverging waves, there is produced by the motion of the 
model another notable series of waves which carry their crests transversely 
to her line of motion. * * * The wave is largest where its crest first appears 
at the bow and it reappears again and again, as we proceed sternward 
along the straight side of the model ; but with successively reduced di- 
mensions at each reappearance. That reduction arises thus: In pro- 

* T^BHtaetitMS Imtitntian Naval ArchUecti, 1877. 


portion as each individual wave has been longer in existence, its outer 
end has spread itself further into the undisturbed water on either side ; 
and, as the total energies of the wave remain the same, the local energy 
is less and less, and the water crest, as viewed against the side of the 
ship, is constantly diminishing." 

In Fig. 5,* there is shown, in plan, the wave-systems of the 
model corresponding to a ship, 333 feet long, at 18 knots speed, 
the crest of the first bow transverse wave being, for clearness, 
omitted. It will be seen that the two systems generated at the 
bow are repeated at the stern, the former, however, being more 
prominent The transverse and diverging wave-systems are il- 

lustrated also — the former more distinctly — in reproductions of 
photographs of a model of the proposed U. S. Battleship Georgia, 
under test in theU. S. Experimental Basin, and of the U. S. Bat- 
tleship liKua, steaming at full speed during her contract trial. 

From the foregoing, it will be observed, that, at low speed, wave- 
formation is practically absent, the resistance being almost wholly 
frictional. When the speed is increased, diverging waves appear 
at bow and stern, followed, as the speed grows, by the transverse 
system. The generally accepted theory as to the stream-line 
genesis of the parent-waves at bow and stern, has been described 

•RE. Fraude, Ttam. Jmt. Nav. Arch., 1881. 


in §2. From each of the two waves so formed, there^spring, at 
high speed, diverging and transversa systems. 

2. Power Expended in Overcoming Wave- Resistance, — ^With re- 
gard to the power expended in the maintenance of these systems, 
there is some conflict of opinion. The I.H.P, developed in an ac- 
tual ship, at a measured speed, can be calculated with accuracy 
and the power spent on engine-friction (§ 1 3, 14), on the lost work 
of the propellor (§15), and on the frictional component of hull re- 
sistance (§ 3), can be estimated in fair approximation. After de- 
ducting the three latter from the gross I.H.P., there is found the 
power expended on wave-systems proper, on " interference," as 
below, and on the interactions of hull and propellor (§ 16). Here, 
approximations to accuracy end and speculation begins. Mr. R. 
E. Froude, a high authority, says : 

<<The work a ship has to do in maintaining a wave- system, whether of 
transverse or diverging waves, is in effect equivalent to adding to the 
system one new wave for every wave-length travelled." 

The reasoning on which this conclusion is based, is as follows : 

**A system of deep-water waves does not travel as fast as the individual 
waves composing it. In fact, the energy represented by the wave-motions 
is transmitted from particle to particle, in the direction of travel of the 
wave by the mechanical conditions of wave-motion ; but this transmission 
is only effected at half the speed of the individual waves, so that, al- 
though, in reference to the panicles of water, the energy is being trans- 
mitted forward, yet in reference to the waves it is in effect draining 
backward from each wave into its successor. * * * Now, the wave- 
system as a whole can only travel as fast as the energy is transmitted. 
The speed of the system is, therefore, exactly half that of the waves, so 
that in a wave-system 100 yards long the wave, which is at one moment 
coming into existence as the hindermost crest by the time it has run 100 
yards, will have become the central wave; and, in aaother 100 yards of 
run will be disappearing at the leading end.""*" 

3. Intcfference, — The foregoing theories apply to both the bow 
and stern systems, acting independently. There remains to be 
considered the effect upon the latter of the train of transverse 
waves generated by the bow. It is evident that the length of the 
ship will affect materially the endurance and height of these waves 
by the time they reach the stern. Experimentsf conducted by 

*R. E; Froude, Trans, Inst, Nav, Arch.^ 1881. 
t Trans. Inst. Aav. Arck., 1877. 




^ *N «o ^ "^ «< 






Mr. William Froude bear directly upon this point. They were 
carried out with a series of models of identical entrance and run 
but with different lengths of parallel-sided " middle body," /". e,, the 
length through which the midship section is maintained. 

The frictional resistance, in each case, was calculated and de- 
ducted from the total, leaving a " residuary resistance " which re- 
presented the wave- and eddy-making factors. As shown in Fig. 
6, these residuary resistance in tons were laid out as the ordinates 
of curves of various speeds, corresponding, for full-sized ships, to 
those at which the models were towed. The abscissae were the 
lengths of parallel middle-body in feet. Each curve begins, at 
the right, with the residuary resistance of a i6o-ft. ship, having no 
parallel middle-body and is extended to include that of a 500* ft. 
ship, having a middle-body 340 feet long. The vertical series ot 
points at any ordinate shows the residuary resistance of a ship 
whose length is 160 feet, plus that of its parallel middle body, as 
marked at the base of the ordinate. 

At the high speeds in which wave-resistance becomes a very 
important factor the curves present a series of undulations, show 
ing that the gradual increase in the length of the middle body 
produces an alternate rise and fall of residuary resistance. It was 
found that these variations were due to the changes in the position 
of the after-body in respect of the wave-crests of the bow trans- 
verse system which reached it ; and that, when the crests of the 
latter coincided with those of the stern transverse waves, the resid- 
uary resistance was greatest, while, when a crest and a trough of 
the respective systems were coincident, the resistance was a mini- 

If two waves combine thus, the altitude of the resultant is, when 
the crests coincide, equal to the sum of the heights of the compo- 
nents ; and, when a crest and a trough meet, to their difference, 
/. Y., to the height of crest minus depth of trough, both being 
measured from the still- water level. Again, ** if the crests of one 
scries fall on tlie slopes of the oi1uf\ the resultant crest-position iKnll be 
a compromise heticeen the crest positions of the components^ though 
nearer to the larger of the tico.'* The degree of coincidence between 
the combining systems gives, then, a measure of the energy of the 
resultant system, which energy will be, in any case, represented by 
the square of the height of the resultant system. 



4. Entrance and Run. — Other experiments,* made also by Mr. 
William Froude, show generally the effect of certain variations in 
the form of the hull, while retaining the same displacement. The 
particulars of the ships corresponding to the four models tested are : 


Displacement, tons,.... 

Entrance, feet 

Middle body, feet 

Run, feet 

Total length, feet 

Breadth, extreme, feet 
Draught, mean, feet. 


Welted surface, sq. feet I 18,660. 


3,9«o. 1 




179-5 ! 








1795 1 




359- „ ' 




45.88 1 











These tests were made in order to show the relation between 
length and resistance in: 

{a) That form of hull in which a parallel middle body is inter- 
posed between two ends of greater or less fineness ; and 

[b) That type in which the whole length of the ship is utilized 
in fineness of form. 

Models A and D represented class (a) ; B and C, class {b). A 
differed from D and B from C in proportions and degree of fineness. 

Tests of the models were made at speeds corresponding to those 
from 9 to 20 knots for the ships, excepting model D, the trials of 
which were ended at 16 knots. The models of class {6) showed, 
in every case, less resistance than those of class (a), the difference 
being marked at high speeds. At low speeds, C having less wetted 
surface than B, had less resistance ; but, at high speeds, B, having 
a greater ratio of length to beam than C, met less wave and total 
resistance than the latter. 

These results indicate generally, where minimum resistance at 
high speeds is desired, that the parallel middle body should be 
omitted, leaving the entire length for entrance and run. This con- 
clusion seems natural, in view of the fact that the fore and after 
bodies are the wave-making features of the ship and that the 
greater length and finer form thus given them will affect materially 
the resistance and character of the wave generated in allowing a 
longer time at a given speed, for wave-formation. 

For the combined lengths of entrance and run appropriate to a 
given speed, Sir William White t gives, as a fair approximation 

♦ Trans, Inst, Nav, ArcA., 1876. 
f Naval Architecture, 5th ed., p. 459. 


under ordinary conditions, the following formula derived from the 
rules of Mr. Scott-Russell : 

in which Z., and Z, are the lengths in feet of the entrance and run 
respectively, and V'\s the speed in knots. In sea-going vessels of 
ordinary type, the lengths of entrance and run arc approximately 

With regard to the forms of these features of the hull, Mr. R. E. 
Froude states that " Experiments have shown that, as a rute, moder- 
ately U-shaped sections are good for the fore- body, and comparatively 
V-shaped sections for the after-bodyy * These results receive sup- 
port, in Mr. Froude's view, from considerations, in substance, as 
follows : 

Diverging waves are shorter than those of the transverse type. 
If the given areas of the various sections of fore or after-bodies be 
so disposed that the ratio of breadth to depth is relatively large 
(;. /., in V-shape), then, other things being equal, it is reasonable 
to infer that the tendency will be to increase the resistance due to 
diverging waves. Conversely, if the ratio of depth to breadth be 
increased, as in the U-section, the reverse effect will be produced. 
Again, the stern series of diverging waves is less marked and less 
formidable than the similar system at the bow. 

5. Resume. — The theories as to wave-making resistance, which, 
at present, meet general acceptance, may be summarized thus : 

{a) When a ship is driven at high speed there are generated, at 
both bow and stern, diverging and transverse wave-systems, the 
latter of which, under usual conditions, combine to form a single 
transverse train at the stern. 

(b) The diverging waves become at once disassociated from the 
ship; they are relatively short in the line of the crest and have a 
length and velocity agreeing approximately with that of ''deep- 
water waves traveling at the speed which the ship's speed would give, 
if resolved normally to the crest- ItnesJ* The velocity of the waves of 
the transverse system is that of the ship and their length is that of 
a deep-sea wave of that speed. 

(c) Power (at the rate of that required to form one new wave 
for each two wave-lengths travelled) is expended in maintaining : 
(i) the bow diverging system; (2) the bow transverse system; 

♦ Trant. Inst. Nav, Arch,, i88i. 
VOL. xxn.— 13. 


(3) the stern diverging system ; (4) the stern transverse system, 
as modified by that portion of the bow-transverse system which 
combines with it. 

{cL) The measure of the power required for maintenance, as 
above, is the energy of the waves created as the ship progresses. 
The general formula for such energy is given in § 5. 

[e) Extended experimentation is yet required to determine fully 
the laws governing the wave-making resistances of ships. The 
theory as to wave-formation and maintenance must be considered, 
therefore, as, in many respects, tentative. A formula expressing, 
with mathematical accuracy, wave- resistance, under varying condi- 
tions as to form and speed, cannot be written, excepting in very 
general terms, thus : 

R^ = >&5-, 

in which /?, = wave- resistance in pounds, 5 = speed in knots, m may 
have any value above zero and below six, and >& is a coefficient de- 
pendent upon the linear dimensions of the vessel. 

7. Total Resistance. 

The total resistance of a ship is the sum of the frictional, wave 
and eddy-making resistances; the latter is, in ships of good lines, 
very small and is included usually in that due to waves. By § 3 
and § 6, there may be written for the total resistance : 

R^R, + R^ ^fAS"" + >&5^ 

in which R^ total resistance, R^ = frictional resistance, and /?, = 
wave (and eddy) making resistance. The following estimates as 
to total resistance, will be considered as reasonable approximations 
only, based upon such records of performance as are available at 
this time. 

Tons. Knots. 
In battleships of I2,5cx> at i8, i?and I.H.P. vary as 5<*« and 5*-" respectively. 

« armored cruisers *• 14,000** 22, i? ** I.H.P. ** S^ ** 5* <* 

** protected cruisers ** 7,000 " 20, i? «« I.H.P. ♦« 5^-*» ** S^-^ 
"gunboats *« 1,500** 15, i? *• I.H.P. •* 5' w .1 ^ 

8. The Law of Comparison. 

I. Definition, — By the aid of this law, the resistance, in smooth 
water, of a proposed ship, can be ascertained, in approximation 
from that of its model. Mr. William Fronde, who, in 1870, first 
applied it to ship-resistance, stated the law thus : 


** If the ship be D times the dimension (as it is termed) of the .model, 
and if at the speecs V^, Fj, V^ ... the measured resistances of the model 

are R^, ^,. ^, ... then lor speeds Z>» X ^1, />* X ^j, Z^ X F, ... of 
the ship, the resista. ces will te />» X ^1 , Z>» X -^j , />» X R^ •• To 
thr speeds of model and ^hip thus related, it is convenient to apply the 
term, * Corresponding Speeds.' " 

It has been shown previously that, broadly speaking, the total 
resistance of a ship may be divided into frictional, eddy-making and 
wave-making factors. The method of calculating directly the first 
of thtse components has been indicated ; the second and third 
constitute the ** residuary resistance,** whose amount is deduced 
approximately by the law, as above, from model experiments. 

In a strict sense, the law of comparison applies only to wave- 
making resistance; and, further, its provisions are restricted to a 
vessel which is ** D times the dimension '* of the model tested, 
t. e.^ which, while larger or smaller than the model or ship with 
which comparison is made, is similar geometrically to either. The 
•• similar " ships, which come under this law are, then, only those 
which have, with respect to each other, the same ratio of length to 
breadth and to draught and the same measure of fineness; and 
their •* corresponding speeds" are those which are proportional to 
the square roots of their respective linear dimensions, or, in view 
of their geometrical similitude, to the square roots of the ratios of 
those dimensions, or of the lengths. 

2. DiTtvation, — According to accepted theory, waves are gener- 
ated by the differences of pressure in the " stream-tubes " formed 
by the passage of the ship ; and the law under consideration pro- 
ceeds, with regard to a vessel and its model, on the basis that, as 

"the originating forces are similar and travel at speeds proportional to 
the square root of their respective dimensions, the re-ultii«g forces (with 
rrgard to wave-making) being as the square of the speeds, will re such 
as to create wave-configurations precisely similar in every respect.*' 

i. e., configurations which, relatively to each other, are enlarged or 
diminished in the same ratio in all directions. Again, since the 
masses elevated in wave-making are similar, the resistance due to 
wave-formation will vary as the cube of their like dimensions, 
which variation will be in accord with the law as stated. 

Thus, as in Mr. Froude's example, if the waves surrounding a 
ship 160 feet long and moving at 10 knots' speed, are modeled, 
together with the ship, on any scale, the model will represent, as 


well, on one-half that scale, the surface about a " similar " ship, 320 
feet long and traveh'ng at 14.14 knots since 

D = W^ = 2 and Z?* x 10 = 2* x 10 = 14.14 ; 

or, again, on 16 times that scale, the wave systems about a model 
of the ship, 10 feet long and towed at 2.5 knots, since 

^ = TV(y = h ^"d Z?* X 10 = (^V)* X 10 = 2.5. 

The limits of space will not permit a mathematical deduction* 
of the Law of Comparison. In outline the process is as follows : 

The two ships, or the ship and its model, are geometrically simi- 
lar, the first being, in all directions, D times larger than the second. 
Assume them to be ship-formed solids, submerged under theoret- 
ical conditions as described in § 2. Then, for each, an equation for 
** stead/ flow " may be written ; and, from these equations it will be 
found that the velocity of the stream-lines for the first solid is, at 
all points-, D\ times that of the second, which factor is the ratio of 
•* corresponding speeds," the latter being, then, such speeds as will 
produce similar stream and wave-configurations. Again, the re- 
sistance due to a stream-hne, when the solid is partially immersed, 
may be measured by the summation of the products of the ele- 
mentary longitudinal components of the stream's pressure upon 
the solid and the corresponding elements of the latter*s immersed 
surface. Now this pressure is, for the magnified stream-lines of 
the ship, Z> times that for the model's stream-line. Again, the 
elementary areas of the ship's wetted skin are LP' times those of 
the model. Hence the product of these two factors will give, 
for the ship, a residuary resistance LP times that of the model at 
the corresponding speed. 

3. Limitations, — With regard to the Law of Comparison, Mr. 
Froude stated that : 

**It would be absolutely correct if the elementary resistances due to 
wave- making, to surface-friction and to the formation of dead-water 
eidies, constituted the entire resistance ; and if * * * it were strictly 
true of the latter two elements alike, that the resistance varies as the 
square of the speed and as the area of ihe surface on which it acts. With 
reference «o dead-water eddies, indeed this double proposition can be 
confidently accepted ; but the experiments on surface-friction * * * 
show that in regard to the latter element at least, the prop>sition does not 
express the exact truth.** 

* See ** Resistance and Propulsion of Ships," Durand. 


It has been shown previously that the skin-resistance of clean 
hulls varies, under usual conditions, not as the square, but as the 
1.83d powers, of the speed and also as a coefficient whose value 
is affected by the length of the ship or model. Hence, the law does 
not hold for this factor, which must be calculated separately. 
Furthermore, the reasoning on which, hydrodynamically, the law is 
based, assumes an ideal frictionless fluid with absolute continuity 
of flow, /. ^., with the absence of eddies or breaking waves. 

4. Application, — In Fig. 2. there is illustrated the application of 
Froude's method to the determination of the resistance of H. M. S. 
Greyhound from that of a model of the vessel. The abscissae of 
the various curves are speeds in feet per minute and their ordinates 
are resistances in pounds. The ship was 16 times the size of the 
model. Hence Z^= 16, Z> = 4, and Z^= 4,096. The "corre- 
sponding speed" of the ship is, then. 4 times, and its correspond- 
ing residuary resistance 4,096 times, that of the model. 

The figures marked just below the horizontal axis of coordi- 
nates, indicate speeds of model ; those immediately above it, cor- 
responding speeds of the ship. The scales to the right represent 
the ship's resistance in pounds, on the basis of 4,096 to i for the 
fresh water in which the model was tested and corrected for salt 
water in the ratio of the density of the two fluids. The ordinates 
of the curve A-A give the total resistances of the model, at vari- 
ous speeds, as recorded in towing experiments ; those of the curve 
B-B shows its frictional resistance, as calculated for these 
speeds; and the portions of the ordinates intercepted between the 
curves A-A and B^B represent its residuary resistances. 

The residuary resistance of the ship, at any speed within the 
range of the test, will be according to the law of comparison 4,096 
times that of the model, as shown by the portion of the corre- 
sponding ordinate lying between the curves A-A and B-B, if 
measured from the diagram, or it may be found directly by read- 
ings from the scales to the right. To ascertain the total resistance 
of the ship, its skin-friction at the various speeds, was calculated 
and the results set off downward from the curve B-B, giving 
thus the curve C-C, The total resistances of the ship corre- 
spond, then, to the portions of the ordinates between the curves 
A-A and C-C and its frictional resistances to the parts be- 
tween B-B and C-C. The portions between C-C and the 
horizontal axis of coordinates represent the excess of model- friction 
over that of the ship. 



5. Extension, — In practice, the Law of Comparison is extended 
to cover the total instead of the residuary resistance, to admit the 
substitution of the I.H.P. for the total resistance, and to compare 
ships whose dissimilarity is not excessive. The Extended Law 
will be considered in § 24. 

9. Resistance at Excessive Speeds. 

From § 6, we have, as the combined length of entrance (LJ and 
run (Z,) appropriate to a given speed : 

Zj + Z,= 0.937 F»; 



\ 0.937 / ' 

or, in general, assuming no parallel middle body and a total length, 

Again from § 8, we have for similar ships of lengths, L and Z', 
of speeds Fand v, and ratio of linear dimensions, /: 

where ^ = t/xly7l = the speed-length coeflficient or ratio. 

From the above formulae, it is apparent that, in vessels of normal 
type, the length is a factor of prime importance in determining the 
maximum speed at which the ship shall be driven. In vessels of 
excessive speed, such as torpedo-craft, there is a marked departure 
from these proportions, with, in consequence, a great increase in 
the resistance and in the I.H.P. required for propulsion, as will be 
apparent from the following data from the performance of vessels 
of the U. S. Navy. 


Battleship | 

Armored Cruiser..., 




















Per Too. 





From the above examples it will be seen that the I.H.P. required 
for small vessels of excessive speed, such as torpedo-craft, is about 
fourteen times that of large vessels of normal speed. 


With such undue powers and speeds, it is natural to expect va- 
riations from, or rather new developments in, the theories concern- 
ing wave-resistance. Under normal conditions the rate of the 
speed at which the resistance varies grows with increased speed. 
Sea-trials of torpedo-craft have shown that this holds with such 
vessels up to a certain " critical " speed, from which upward there 
is a decrease in this rate. Thus, M. Normand * found, with the 
torpedo-boat Forban of 125 tons, that the power of the speed at 
which the total resistance varied passed the square and rose stead- 
ily to higher powers until the vessel was steaming at 20 knots, 
when the rate began to fall and at 31 knots was still decreasing 
toward the square of the speed. These experiments locate the 
'•critical speed," /. /., that at which the resistance varies as the 
highest power of the speed, as, approximately, 

F= 9(Z))*. 

in which V^ speed in knots and D a displacement in tons. For a 
vessel of 500 tons, F=» 25.4 knots and for one of 8,000 tons, 40 
knots. It is apparent, that, in deciding upon the maximum speed 
for small, fast craft, the prudent designer will avoid the neighbor- 
hood of this point of greatest resistance and minimum utilization 
of the power expended. 

As the critical speed is passed, there is considerable alteration in 
trim and a bodily rising of the hull, due, M. Normand states, to 
" the vertical component of the direct resistance." In consequence 
of this rising, the reduction in the rate of resistance has been as- 
cribed to a lessening of the displacement and wetted surface. On 
the other hand, Mr. Sydney Barnaby contends : 

"There is doubtless a lifting above the surrounding water-level, due to 
the boat being upon the back of a wave which its bow frequently over- 
hangs in such a way as to be clear of the water altogether ; but it does 
not follow that there is a diminished displacement. It seems more prob- 
able that the improvement in performance is due to the subsidence of the 
wave-making element of resistance at a critical speed.** f 

The uncertainty as to the exact cause of this phenomenon, is in- 
dicated by Sir Nathaniel Barnaby, who says: 

"The longer and heavier the ship the higher is the speed at which na- 
ture begins to favor the engineer in his attempt to fly. In what way she 

*Jour, Am. Soc. Nav, Engs., IX., I. 

t Marine Propellers, Barnaby, 4th Ed., p. 105. 


makes this apparent charge in her methods it is not easy to explain. Sir 
William White says, ' The boat travels upon the back slope of a wave 
having the same speed as herself.' She is seen to rise in the water, the 
bow is eventually lifted out of it and the vessrl settles down to speeds 
gained with comparative ease under the new conditions." 

10. Resistance in Rough Water. 

The methods (§§ 3, 6) of estimating resistance which have been 
described, refer only to the conditions of smooth water. In a sea- 
way, a ship meets resistances which differ so widely from those 
opposing her motion in still water, that the designer must rely 
only upon judgment and experience in the determination of their 
possible magnitude. There is no process of calculation, exact or 
approximate, which will cover, with safety, the effects in infinite 
variety, of heavy seas upon the resistance of a proposed vessel. 

The ship, her wake, and the adjacent water unite to form a sys- 
tem, requiring, in smooth water, a definite amount of power for 
its maintenance. Any derangement of the conditions, as in pitch- 
ing or rolling, means a disturbance of the balance and an excess 
expenditure of energy. Furthermore, the constant change in the 
volume and form of the immersed hull, affects also the propelling 
mechanism, which is at a continual disadvantage in acting at an 
angle with the desired line of motion. This lb true especially of 
pitching, which not only, under all circumstances, increases resist- 
ance directly, but also, through the racing of the screws, sets a 
limit to speed in a heavy sea. 

As a result, sea-speeds and those indicated by model experi- 
ments or ascertained from still-water trials, differ; and the form of 
hull which is best adapted to the conditions of the latter, may 
be far from that suitable for the open ocean. For example, maxi- 
mum smooth- water speeds require great fineness of the ends of 
the hull ; and yet, if the latter are not sufficiently full, there will be 
undue pitching and increase of resistance at sea. In general, it 
may be said that length, size, weight, and consequent inertia, act 
to reduce resistance from irregular motion, especially when com- 
bined with steadiness and good freeboard. 

II. Resistance, Work, and Horse-Power. 

A general formula for the total (frictional, eddy, and wave) re- 
sistance has been given in § 7. The effective work of propulsion 


consists in driving the ship against that resistance through a given 
space. In the transformation of that work into tffectivc power, 
the element of time enters, the power being the work done during 
a given period. The units used are: The pound, the foot-pound, 
the horse-power = 33,000 foot-pounds per minute, and the knot 
= a speed of, approximately, 6,080 feet per hour. For a proposed 
ship, at a speed of 5 knots per hour, we may write (§ 7) : — 

Hull resistance (lbs.) ^^fAS"" + kS"^ ; 

Effective ivork per minute = {fAS"" + kS'^)iS X f- I ; 


Effective H.P (EHP,) = (/^5"+» + kS^-^') ^ ^^^- , 
•^ ^ / V ^60 X 33,000' 

which expression assumes the ** hull efficiency " (§ 16) to be unity. 

12. Indicated and Effective Hokse- power. 

The total work performed by any machine may be divided into 
two parts: the useful work, or that which produces motion 
against resistance external to the mechanism, and the lost work, or 
that which is consumed by the frictional or other detrimental re- 
sistances of the mechanism perse. If the consideration of the 
energy of propelling machinery begin at the cylinders, extend to 
and include the screw, and embrace the reactions between the hull 
and propeller, the total power developed or the total energy ex- 
pended, will be the indicated horse-power, as above ; the effective 
horse-power will represent the useful work; and the difference be- 
tween the two will be the work attendant upon, but ineffective in, 
propulsion. The total I. H.P. is expended, in detail, upon : 

{a) Engine Resistance, comprising the initial and load- frictions 
of the engine and shafting. 

{b) Lost Work of the Propeller,\tic\ud\tig its friction, the edgewise 
resistance of the blades, and the power expended in slip. 

{c) The Net Resistance of Hull and Screw Inter-actions (§ 16). 

{d) Hull Resistance, proper. 

(e) Pumps, — The total I.H.P. of a propelling engine should in- 
clude the power required to operate the air, circulating, and feed 
pumps, as this machinery — whether independent, or as was for- 



merly usual, forming an integral part of the main engine — is 
essential to the operation of the latter. 

13. Initial Friction. 

The initial friction is that of the unloaded engine, the term 
being used to designate the frictional resistance of the pistons, 
stuffing boxes and the various bearings and journals of the engine 
and shafting, when moving without external load. It is evident 
that this resistance will be a constant for each double stroke or 
revolution, whether the engine be unloaded or the reverse, and at 
whatever speed it may be driven. The initial friction is, hence, 
a function of the number of revolutions, and the power required to 
overcome it may be measured by an ascertained mean effective 
pressure (referred in stage-expansion engines to the low-pressure 
cylinder), which pressure will be approximately the same at all 

The amount of power thus required for a given ^engine varies 
with the type, the form of valve-gear, the standard of workman- 
ship, etc. It is obtainable with accuracy only by trial, either by 
indicating the engine at various speeds in the erecting shop and 
estimating the additional power required to operate the unloaded 


— ¥ 


6 a jc 

Fig 7. 


shafting; or, on shipboard, by disconnecting the propellor and 
measuring the power required to run the remaining mechanism 
without load. In the absence of such accurate tests various ap- 
proximate methods have been proposed for the estimate of the 
amount of initial friction, none of which may, however, be implic- 
itly trusted. 

Thus, Mr. William Froude constructed a curve of *• indicated 
thrust," as in Fig. 7, the abscissae of which are speeds in knots, 
and the ordinates are indicated thrusts in pounds, calculated from 
the expression : 

r ,. , ^ LH.P, X 33,000 

Induated Thrust = „ — „ -, 

P y. R 

in which Pas. pitch of propellor in feet and ^=» revolutions per 
minute. From the results of a progressive speed trial, the curve is 
laid out, its lower limit being at a point, as a, on the ordinate from 
the least speed reached. At very low speeds, Mr. Froude assumed 
hull resistance to be wholly frictional and to vary as the power 1.87 
of the speed ; the small residue of other resistances, including load- 
friction, was considered also to change at that rate, excepting the 
initial friction, which was taken as a constant. 

Under these conditions, if the element of initial friction be 
omitted, the curve, for low to zero speeds, from any point, as/, to 
the axis OY^ will be a parabola whose ordinates will be as the 
power 1.87 of their corresponding abscissae. Therefore, to con- 
struct the curve, draw through / the tangent /'-/" to the curve, 
a-b- H ; from /, let fall the ordinate p-h ; divide O-h into parts pro- 
portional to 0.87 and 1. 00; from the point of division, A', erect an 
ordinate cutting the tangent /'-/"; and through the point of inter- 
section draw the right line K-L parallel to the axis O-X. A' will 
then be the vertex of the required curve and the portions of the 
ordinates intercepted between O-Xand K-L will be the equivalent 
of the initial friction at the respective speeds. The curve of indi- 
cated thrust, laid out as above for the U. S. gunboat Yorktoivn is 
shown in Fig. 8. 

It will be observed that the expression, as above, for indicated 
thrust is, in some degree, arbitrary, since it makes no provision for 
slip, engine-friction, etc., being in fact, ''the thrust which the propellor 
would be exerting, if the force of t/ie steam wete employed wholly in cre- 
ating thrusty It has, however, a value relatively in this, that as 


Mr. Froudc explained, it is in constant proportion to the mean ef- 
fective pressure (referred to L. P. C>lindcr). Thus, the usual for- 
mula for I.H.P. and that for indicated thrust may be written: 

I.H.P.= m.e.p. -A Ry.C\ 

LH.P. = Ind. Thr. x R ■><. K , 
in which R= revolutions per minute and Cand K are constants. 
Since, in any specific case, I.H.P. and R will be the same in both 

formuItE, the unvarying relation between the indicated thrust and 
the referred mean effective pressure is apparent. 

In view of thi* ratio, analogous methods have been proposed, in 
the plotting of a curve of mean effective pressures on the revolu- 
tions as abscis-ije, either directly * or as a derived f curve from that 
of the I.H.P. constructed with similar coordinates. The distance 

• Resistance and Propolsion of Ship!, Durand, 

f Reiistance of Ships and Screw Propulsion. Taylor. 


to the origin from the intersection of this M.E.P. curve with the 
axis would then be, assuming accuracy, the referred M.E. pressure 
required to overcome initial friction. The objection to Froude's 
method, however, holds, in degree, with all others. Absolute 
knowledge of actual conditions ends with the last spot plotted 
from observation. The drawing of a tangent at, or near, the 
limit thus found or the prolongation, by judgment, of a direct or 
derived curve, has each its inherent element of uncertainty. 

Since, to overcome initial friction, there is required the same 
average M.E.P. per stroke, whether the number of revolutions per 
minute be small or large, it is evident that the percentage of the 
total I.H.P. thus expended will be much greater at low than at 
high speeds. Commodore Isherwood, in exhaustive experiments* 
conducted upon a 54-foot steam launch, found the initial friction to 
average about 8.5 per cent, of the maximum I.H.P. of the engine. 
In other experiments, this amount has fallen as low as 5 % and 
risen as high as 9 %. 

14. Friction of the Load. 

This term refers to an additional frictional resistance at the 
journals and bearing surfaces, caused by the load applied when 
the engine is overcoming external resistance, and dependent, there- 
fore, upon the horse-power which is equivalent to that load. The 
load-friction may be stated as a percentage of the total I.H.P. 
or of that power, less the percentage for initial friction. The data 
as to this subject are meagre, the obvious reason being that a trust- 
worthy determination of the value of the friction of the load 
could be obtained only by the interposition of dynamometrical 
appliances between the shafting and screw, in order to ascertain 
the ratio between the total I.H.P. and the power delivered to the 

An amount, long accepted for the load-friction and deter, 
mined originally by experiment, is 7.5 per cent, of the total l,H,?., 
neglecting, in the partial obscurity of the question, the small allow- 
ance for initial friction. With the separation of the initial from 
the load- friction and the depen !ency of the latter upon the load, 
experiments conducted by Prof. R. H. Thurston would seem 
to be not in accord — his conclusion being, that, for engines of the 

* Report of Secretary of the U. S. NaTy, 1874. 


type tested, •* the indicated power witfiaut load being (is) sensibly the 
measure of the wasted work of the engine, when in operation under 
load of whatever amount^* The engines tried were stationary, 
non-condensing, and of relatively small power. On the other 
hand Prof. Cecil H. Peabody f gives the equations : 

F^ aP^ + bP\ 

in which /'"«= total friction, aP^= initial friction, bP^ load fric- 
tion, P^ = normal net (brake) horse-power which the engine is 
designed to deliver, /^= net horse-power delivered by the engine 
at the time considered, and a and b are constants. From the re- 
sults of tests made by Walther-Meunier and Ludwig on a receiver 
compound, condensing engine of 290 I.H.P., Prof. Peabody gives 

I.H. P. = 0,07 P^ + 1 ,07 P, ( I ) 

Similarly, for a Corliss engine, 160 I.H.P., condensing, tested by 
Delafond at Creusot : 

I.H.P, = 0.062P -h I . I ;/", (2) 

In (i), a = 0.07 ; b = 0.07 ; 

" (2), ^^ = o.o6; ^ = 0.17. 

15. Lost Work of the Propeller. 

After the power required for initial and load frictions has been 
deducted, the remainder is the I.H. P. delivered to the propeller. 
In the operation of the latter, waste work occurs principally from : 

I. Resistance of t/ie screw, (nctional as to its surface and edge- 
wise as to the passage of the blades through the water. The first 
of these resistances is the same in kind as that of the wetted skin 
of the hull and may be calculated from Froude's experimental re- 
sults as to the friction of planes (§ 3). The total horse-power 
thus absorbed may be found from, (a), the area, in square feet, of 
the helicoidai surface of both sides of the blades ; (b) the speed, 
in feet per second, of that surface in its helical path ; (c) a fric- 
tional factor from Froude's results, altered to suit the speed of the 
helicoidai surface. For accurate results, the surface should be 
considered in sections. 

*TraHS. Am. 6V. islav. Aw/f, Vols. VIIL, IX., X. 
f Thermodynamics of the Steam Engine, Peabody. 


The heady or edgewise, resistance cannot be calculated. There 
are, however, limited data available as to the total resistance of 
screws. Isherwood, in the experiments noted previously, found 
the increase of hull-resistance to range fron* 8.56 to 19.51 per 
cent., with various propellers, when the launch was towed at 7 
knots with the screw disconnected and revolving. Again, the 
triple-screw engines proposed recently for U.S. armored cruisers of 
I3,6(X> tons, 23,cxxD I.H.P., and 22 knots, maximum, were planned 
to develop one-half the power, as above, in the central engine; 
and it was estimated, that at 10 knots, with the wing engines only 
in operation, 200 I.H.P. would be required to overcome the drag 
of the middle screw when revolving idly. The high circumferential 
speeds of propeller, now usual and growing, in torpedo-craft espe- 
cially, make screw-resistance a matter of importance. 

2. Loss due to the Propelting Medmm, — The propeller, when 
driving a ship, may be compared, very generally, to a short length 
of screw-bolt, of small body and deep multiple thread, working in 
a somewhat yielding nut. As the bolt advances, against a head- 
resistance, the " nut" recedes to some extent, /. e,, there is forced 
backward from the screw an approximately columnar volume of 
water. If it be assumed that the water be replaced by media 
which grow gradually less mobile and more viscous until solidity 
is reached, then, for the same head-resistance, the volume of fluid 
driven aft per revolution, would decrease, until, with solidity, this 
amount would become zero and there would be no recession of the 
virtual nut. 

The work done by the assumed fractional bolt will consist, then, 
while fluidity prevails, of: {a) the ** useful" portion which over- 
comes head-resistance, and (^) the "lost" or wasted portion which 
moves backward the mass of fluid, as above. With solidity, {b) 
becomes zero, and, disregarding engine- friction and sere w- resist- 
ance, Froude's formula for indicated thrust applies with accuracy. 
It is apparent that the element {b), as above, while *• lost," is at- 
tendant essentially upon propulsion in a mobile medium, since it 
is the reaction of the mass driven backward which gives the for- 
ward thrust and which is, in all cases, equal to that thrust, u e„ to 
the resistance of the ship. Since, then, a screw propels by forcing 
water rearward, there is, inevitably, work lost in its operation equal 
to the energy of the water thus displaced, and, therefore, the prod- 
uct of the weight and velocity of this columnar stream bears a 


definite relation to the propelling force. This sternward velocity 
is known as the real slip. If a screw had no slip, there would be 
no sternward column, no reaction, no forward thrust, and no pro- 
pelling force. 

i6. Interaction of Hull and Propeller. 

A ship, when towed, has a definite resistance, which in smooth 
water, is practically invariable for the same speed and condition 
of skin. Similarly, a propeller, if driving a ship at such a dis- 
tance from her hull as to be uninfluenced by the latter, would de- 
velop, in still water, the thrust and slip which should characterize 
\\.,perse, at a given speed. When, however, as under the condi- 
tions of practice, the ship is moved by the propeller and the latter 
revolves in water whose motion is affected by that of the ship, in- 
teractions occur between the hull and the screw which change 
both resistance and thrust. To the wake (§ 3) and the augment 
of resistance (or thrust deduction) these effects are due. 

I. The wake, or current following the ship, depends upon her 
form, skin and speed ; it is produced by {a) hull friction, owing 
to which the ship transports with her a volume of water, not only 
astern but covering in varying thickness, her immersed skin ; (p) 
stream line action, which, according to generally accepted theory, 
generates following waves, especially at the stern, which waves 
may be modified by "interference" (§ 6). A wave gives a for- 
ward pressure on the screw if the latter be beneath a crest and 
sternward, if beneath a trough ; {c) eddy nsista?ice, if of ap- 
preciable amount, increases the volume of the wake. 

These actions result in the formation of a current, which, at 
the surface and near the stern, has its maximum speed, the latter 
being, in some cases, fully one-half that of the ship. In directions 
sternward and downward, however, this velocity decreases rap- 
idly, thus giving a current of widely varying speed, the average of 
the latter being taken under ordinary conditions, as 10 per cent, 
of that of the ship. Model experiments indicate that the wake, in 
its effect upon the propeller, may be treated as a forward current 
of uniform speed. The real slip (§15) has been defined as the 
velocity of the fluid column forced sternward by the screw. It is 
obvious that when a propeller acts in a forward stream like the 
wake, the slip and wake velocities will oppose eaqh other and 
therefore, that the observed or apparent slip will be less than the 


real slip by the amount of the wake-speed. Thus, let Vw^ speed 
of screw » pitch x revolutions and 5 and v a speeds of ship and 
wake, respectively. Then will 

Apptrent Real. 

Slip, -F-5, ^V-S + v; 


% of K ■» j^^ ' 100, ■« rr 100. 

The final effect of the average wake is, then, through its effect 
upon the propeller, to increase the speed of the ship. In other 
words, the power expended originally in wake-formation is parti- 
ally regained by the operation of the screw within the limits of this 
current. The amount of this recovery will depend upon the 
strength of the wake and upon the location of the screw or screws 
with regard to it. Its effective action, under these conditions, ex- 
pressed in terms of the ship's speed, ranges from 5 per cent, to 30 
per cent, of that speed, the lowest value being for long and 
lean twin-screw vessels whose wake is naturally small and whose 
propellers work near the boundaries of the stream, while the high- 
est percentage applies to a bluff vessel whose single screw revolves 
where wake-speed is a maximum. Triple-screw ships, especially 
when the central engine develops a large proportion of the power, 
should have a decided advantage over twin-screw vessels in this re- 

2. The Augment of Resistance (Thfust Deduction) is the increase 
of hull-resistance due to the operation of the screw at the stern. 
When a ship is towed, the natural tendency of the water is to close 
in at the stern (§ 2,6) and form a wave-crest there, thus producing 
a forward pressure in opposition to that on the forebody. A 
screw, however, propels both by rearward thrust and by suction 
with the forward faces of its blades, the latter action being 
increased, when the pitch is relatively great, by the excessive race- 
rotation which prevails. The effect is to withdraw the water in 
advance of the screw and from beneath the stern, thus decreasing 
the forward pressure on the latter and causing a virtual increase in 

The amount of this augment depends upon the lines of the stern, 
the speed, and the proportions and locations of the screws. Its 
value ranges from 5 per cent, to 40 per cent, of the ship's net re- 
sistance, under practically the same conditions as to hull and pro- 

VOL XXII — 14. 


pellers as were noted previously for the wake-gain. Torpedo-craft, 
in exceptional cases, have shown a zero or negative wake near the 
screws, and, similarly,^ thrust deduction considerably less than the 
minimum as above, is not uncommon with such vessels. While 
an increase in the wake means usually a growth in the wake- gain, 
it implies also an increase in the power expended in the generation 
of the former and in the thrust deduction as well. The aim should 
be, therefore, rather the utilization of the normal wake than its en- 

In considering the relative values of the wake-gain and thrust- 
deduction, Mr. R. E. Froude* compares an actual vessel with an as- 
sumed " phantom ship " (one considered as capable of resisting the 
thrust of the screw but creating no disturbance of the water), the 
I.H.P., thrust, and number of revolutions being the same in each case. 
Since the phantom ship will have no augment of resistance, these 
conditions can be secured only by assuming her speed as reduced, 
but number of revolutions maintained, until by increased slip, the 
thrust of the actual ship is reached. We have : 

V^=^ Speed, actual ship ; 

R a Net Resistance (total minus augment), actual ship ; 

7^=3 Thrust {R plus augment), actual ship ; 

E.H.P. = if X F, actual ship ; 

Propulsive coefficient = ^ ' ' ' , actual ship. 

V = Speed, phantom ship ; 
T = Thrust, phantom ship ; 
E.H.P. =3 T^x v^. T.H.P. (thrust horse-power), phantom ship; 

T H P 
Propulsive coefficient = y' p', phantom ship. 

Then the ratio of the efficiency of an actual ship to that effi- 
ciency in water undisturbed by wake-gain or thrust deduction is : 

„ „ „ . E.H.P. T.H.P. E.H.P. RV R V 

Hull efficiency = ^^-^ - j^^ = ^^-^ = _ == _ X -. 

Concerning these two final factors of hull efficiency, Mr. Froude 

**One (^RjT) represents the amount by which resistance is less than 
thrust, is theref re, termed the "thrust deduction" factor, and implies 

» Trans, Jnst, Nav. Arch,, 1 886. 


20 1 

loss in efficiency. The other ( V\v) represents the excess of speed of ship 
over speed of screw through the water it works in, is therefore, termed 
the ** wake-factor " and implies gain in efficiency * * * They appear 
to be, to a great extent, inter dependent, the conditions which increase 
the wake-gain tend naturally to increase the thrust-deduction loss and 
the product, the *' hull efficiency," is, therefore, roughly speaking, con- 
stant. It is, in fact, roughly equal to unity in all cases ' 

17. I.H.P. FOk AiK, Circulating and Feed Pumps. 

For various reasons, the tendency in modern designs is to ac- 
tuate some or all of the pumps independently of the main engines. 
In warships of large powers and displacements, the latter is practi- 
cally the invariable rule, and in the merchant marine, especially in 
swift vessels, it prevails to a large extent. The air pumps of gun- 
boats and torpedo-craft are, howevtr, frequently driven by link and 
lever connections from the main engines. Recent improvements in 
design and construction have made possible a marked reduction in 
the power required for this auxiliary mechanism. The following 
table gives results from U. S. vessels, as obtained on official trials, 
arranged in the order of date of the latter : 

Table II. — I. H. P. for Pumps. 




B. S. Oregon. 
Ar. Cr. Brooklyn. 
B. S. Iowa. 
B. S. Kearsarge. 
B. S. Kentucky. 


1 1,9 ^3- 13 
11,787 86 




147.07 1.33 



Per cent averages. 




Air Pumps. 





14. 3« 




Circ. Pumps. 









Feed Pumps. 











18. The Propulsive Coefficient. 

The ratio between the power actually utilized in propulsion (i. ^., 
in overcoming the net resistance of the hull) and the total power 
expended, is a measure, at a given .speed, of the efficiencies of en- 
gine, screw, and hull. In assigning an average value to this ratio 

I. e. 

Propulsive Coefficent = E.H.P. -t- I.H.P. 

estimates have differed considerably, owing to the lack of ex- 
perimental data available. In resume of the preceding discussion 
of the elements of the lost work, there may be noted, as to : 


1. Initial Friction. — In view of the improvements in workman- 
ship and in the materials and methods of lubrication, and of the 
further reduction of friction attained by the use of lightly sprung 
piston-rings of small face, of metallic packing for all rods, of anti- 
friction metals in all bearings, of piston -valves, and of valve-gear- 
ing of simple type, the initial friction may be estimated safely as 
requiring not more than 6 per cent, of the maximum I.H.P. in en- 
gines of medium or large power. 

2. Load Friction, — While the results of Professor Thurston's 
experiments noted in § 14 cannot be disregarded, it will be ob- 
served that they were conducted upon a limited number of small 
engines, in which much difficulty might have been met in register- 
ing minute quantities of frictional power ; and further, that the re- 
sults given for the work lost in friction are in excess of the power 
usually assigned to initial friction alone. The conditions which 
govern the reduction of initial friction apply also to that of the 
load and such data as are available justify a similar esti- 
mate of the latter as 6 per cent, of the total I.H.P., making, for the 
entire engine-friction, 12 per cent, of, and at, the maximum I.H.P., 
in engines of medium and large powers. 

3. Propeller Resistance. — This in modern screws of high speed, 
cannot be taken as less than from 15 to 20 per cent, of the total 
I.H.P. at maximum speed. 

4. Slip. — It is difficult to separate the work lost in slip from 
the portion regained from the wake or that spent in thrust-deduc- 
tion. Actual average cases have shown wake and thrust- deduc- 
tion factors approximately equal and model experiment seems to 
confirm these results. With ships of ordinary type at usual speeds, 
they may, in a general estimate, be considered as neutralizing each 
other and be, therefore, disregarded. 

If the aggregate power expended in engine friction and screw- 
resistance be deducted from the I.H.P. of the main engines, the re- 
mainder will be the power spent on the net resistance of hull and 
screw interactions, in slip, and in propulsion. Considering the 
former to be combined or omitted, as above, the work of slip wil 
be, approximately, the product of this remainder of the I.H.P. by 
the speed of the slip expressed as a fraction of the axial speed of 
the screw. 

Isherwood,* in the trials of the yacht Yosemite (445 tons, 1,400 

^ fournal Am, Soc, Nav. Eng*s, VII., 4. 


I.H.P.), using this method, found at a speed of 15.6 knots and a 
slip of 20.34 per cent, the work lost in slip to be 17.67 per cent, 
of the net horse-power, 1. ^., the I.H.P. less initial friction. In 
average performance, the work of slip will range from 1 5 to 20 
per cent, of the I.H.P. 

5. Losses due to Fornix Location^ and Excessive Speed of Screw, — 
Minor losses due to these causes sometimes occur, which may be 
but mentioned here. For example, a screw with a high pitch- 
ratio will lose efficiency in the undue rotary motion given the race. 
Again, excessive inclination of the shaft from the horizontal or to 
the median vertical plane will aflect its useful thrust. Further, if 
the blades have not sufficient immersion and cut the surface, they 
will draw air down, churn foam, and waste power to a large extent. 
Finally, at very high velocities, " cavitation " may appear, 1. ^., the 
formation of cavities forward of the screw, owing to the inability of 
the water to flow sufficiently fast to follow up the blades. The 
total amount of the lost work of the propeller is such that the effi- 
ciency of a well-designed screw, working under the best attainable 
conditions, seldom reaches 70 per cent. 

6. Resume, — In view of the above considerations, the expen- 
diture of the total I.H.P. at maximum speed may be analysed in 
general, thus : 

Work of initial friction, 6 per cent, of I.H.P. 

«• load friction, 6 " " 

" propeller renstance, 15 <* '* 

" slip, 17 " " 

" hull propulsion, 55 " •* 

" pumps, I ^ " 
Total, lob 

giving a 

Propulsive coefficient ^^.H.Y. -*- I.H.P. » 55/100 » 0.55, 
which coefficient is based on a hull efficiency approximating to 
unity and holds only for the maximum speed of well-formed ves- 
sels. In blufT ships, its value may fall to 0.45 or less, while under 
favorable circumstances, it has exceeded that given above. In 
torpedo- craft, a coefficient of 0.60 or more has been reached. 
There appears to be no marked difference between the coefficients, 
under similar conditions, for single and twin screw vessels. 

At low speed and I.H.P., the only percentage in the above table 
which remains constant is that for the load friction, which is taken 
as 6 per cent, of the power developed. The percentages for pro- 


peller-resistance, slip, etc., will be changed, the most noteworthy 
variation being in that for initial friction which is expressed, as is 
customary although scarcely logical, in terms of the maximum 
I.H.P., the power thus assigned remaining constant at all speeds. 
The result is, as the speed is reduced, the excessive increase of the 
proportion of the power spent in friction, as is indicated, very ap- 
proximately, in Fig. 7, and is shown clearly in Fig. 8. Again, 
from the progressive trials of the U. S. Cruiser Chicago^ we have : 
Power developed at 15.5 knots, sa 4,550 I.H.P. 

Power absorbed in initial friction =5 4,550 x .06 = 273 *• 
Power developed at 7 knots, = 540 " 

Per cent, of power at / knots for initial friction = (273/540) x 
100= 50.55. 

It is apparent, that, at speeds below the normal, small values of 
the propulsive coefficient will obtain. 

19. Relations of Wetted Surface, Displacement, and Mid- 
ship Section. 

The surface, volume and midship section of the immersed por- 
tion of a hull, may be taken as roughly proportional to those of 
the circumscribing parallelopipedon. Consider, therefore, two 
similar parallelopipedons, floating awash, whose lengths, breadths 
and depths are, respectively, Z, B^ //, and I y, L, I x B^ I y. H, I 
being the ratio of the linear dimensions, i. ^., **D " as given in the 
Law of Comparison, § 8. Let a^ v and m be, respectively, the area 
of wetted surface, volume and sectional area of the former solid 
and A, Fand J/ represent similar data for the latter. Then will 
A^ l^ X a, u e., the areas of the wetted surfaces of similar ships 
are proportional to the square of the ratio of their linear dimen- 
sions or to the squares of said dimensions. Again, V^ /' x z/« 
Now, taking 35 cubic feet of sea water to the ton, the respective 
displacements will be 

V V 

— = Z> and — = rf.-. D^l^xd, 

35 35 

L e., the displacements of similar ships are proportional to the cube 
of the ratio of their linear dimensions or to the cube of said di- 
mensions. Again, 



I. e,, the areas of the wetted surfaces of similar ships are propor- 
tional to the two-thirds power of their displacements. Finally* 
calling J/ and m the respective immersed midship sections, 

m a \ a / 

t. £., in similar ships the areas of immersed midship sections 
(§ 20) are proportional to the square of the ratio of the linear 
dimensions, to the squares of said dimensions, to the areas of wet- 
ted surface and to the two-thirds power of the displacements. It 
will be understood that this deduction and those which precede 
apply only to similar ships. 

For example, consider two similar vessels, one having linear di- 
mensions, in all respects, four times those of the other. Then /« 
4 and 

Area of wetted surface otlarger « 4* times that of smaller ; 

Displacement of larger = 4* times that of smaller ; 

Area of midship section of larger = 4* times that of smaller ; 

20. Coefficients of Form. 

Coefficients, based on various relations of linear dimensions 
areas, and displacement, are used in comparing the immersed form 
of a ship with those of other vessels. The factors of these coef- 
ficients refer always to the hull at, or below, the water-line. They 

(a) the length, Z, in feet, usually on the load water line, some- 
times extreme, and properly the average immersed length ; {b) the 
extreme breadth, B, in feet, not moulded, but to outer surface of 
plating ; {c) the mean draught, W, in feet, to top of keel ; {d) the 
displacement, D, in tons, expressed in terms of 35 cubic feet of sea- 
water to the ton; (e) the area, in square feet, of the immersed 
midship section or maximum transverse section btlow the water- 
line; (/) the area, in square feet, of the water-plane on load 
water-line. These coefficients are : 

1 . 714^ Coefficient of Fineness of Displacement or '^ Block *' Coeffi- 
cient^ which is the ratio of the volume of the displacement to that 
of the circumscribing parallelopipedon, or 

-Px 35 


While this coefficient is of much service in comparing vessels 
known to be wholly or approximately similar, it is lacking in this, 
that, for the same linear dimensions and displacement, there are 
an indefinite number of hulls, varying indefinitely in the contour 
of the water-lines, which have the same block coefficient. The 
values of the coefficient, for all classes of vessels from torpedo-craft 
to large cargo-steamers, range from 0.37 to 0.78, a common value 
being 0.60. 

2. The Coeffiaent of Fineness of Midship Section, which is the ratio 
between the area of immersed midship section and that of the cir- 
cumscribing rectangle, or 

area of immersed midship section 

By. W • 

This coefficient, for all classes of vessels, ranges in value from 0.68 
to 0.95, a common value being 0.90. 

3. The Cylindrical Coefficient of Fineness of Displacement^ which 
is the ratio between the volume of the immersed hull and that of 
a cylinder whose length is that of the ship and whose sectional 
area is equal to that of the immersed midship section, or 

-Px 3 5 

(area immersed midship section x I^' 

If two vessels have the same midship section and length but differ- 
ent water-lines, their displacements will vary directly as this coeffi- 
cient. It is hence a trustworthy measure of fineness of water- 
lines. Its values, in steamers of all classes, range from 0.54 to 
0.83, a common value being 0.66. 

4. The Coefficient of Fineness of Load Water-Plane^ which is the 

ratio of the area of that water-plane to that of its circumscribing 

rectangle, or 

area of water-plane 

The values of this coefficient range generally from 0.65 to 0.85 , a 
common value being 0.75. 

21. I.H.P. BY Displacement and Midship Section Formuue, 

These formulae are : 

y>ix 5» 

LH.P^ ^ 





in which, D^ss. displacement in tons, M^ss. area of immersed midship 
section in square feet, and CTand /fare the displacement and mid- 
ship section coefficients, respectively. With regard to these for- 
mulae, the following considerations will be noted : 

1. Origin — They were introduced at a time when, owing to 
the moderate speeds which only were attainable, the resistance 
was very largely fnctional, and, therefore r§ 3), varied as the area 
of the wetted surface and as the square of the speed, the power (§11) 
varying as the cube of the latter. Accordingly the formulae contain 
the factor 5* and also the factors D^ and M^ the two latter being 
(§19), in similar ships, proportional to the area of the wetted sur- 
face. While ihe midship section formula was not constructed 
with these assumptions in vie^r, it, nevertheless, fulfils the latter, 
within the limits in which the midship section is proportional to 
the wetted surface. 

2. Identity, — The formulae may be written : 


from which it is apparent that C has a constant relation to K. If 
the coefficients obtained from the progressive trial of a ship be 
plotted on the speeds as abscissae, one curve will serve, therefore, 
for both Cand A', the scale of the ordinates differing, for each coef- 
ficient, in accordance with the ratio between the latter. 

3. Values of tlie Coefficient. — With regard to either coefficient, 

ii may be said that its value is the same only for similar ships at 

corresponding speeds; for dissimilar vessels and for the same ship 

at different speeds, it varies. The curve of coefficients, for a ship 

of usual type, passes through three stages — a rapid, diagonal rise, 

a gentle slope to and from a maximum, and a gradual descent. 

We have : 




and from § 18, 



J fj p ^ i^,n,ir, 

propulsive coefficient ' 

K^ M y, ~p~j-rp ^ propulsive coeffiaent. 


from which it will be seen that K varies directly as the propulsive 
coefficient and inversely as the E.H.P. 

At low speeds, where wave resistance is slight, E.H.P. varies as 
5', and hence, if the propulsive coefficient were equal to unity, the 
curve would become a straight and horizontal line. The latter co- 
efficient is, however, always a fraction, whose value is, at minimum 
speed, very small, but increases rapidly, especially in high-speed 
engines of short stroke. In the first stage, or sudden ascent of the 
curve, this coefficient grows more swiftly than the E.H.P., with, 
therefore, rising values of K, 

In the second stage wave- resistance becomes an important fac- 
tor, the E.H.P. enlarges quickly, the still expanding propulsive co- 
efficient is overtaken and the highest point of the curve is reached. 
Finally, in the third stage, as maximum speed is approached, the 
then slow increase in the coefficient is more than matched by the 
growth in the total resistance and E.H.P., the latter dominates, and 
the curve gradually descends. In vessels of excessive speed, as 
torpedo-craft, there is a fourth or supplementary stage, due to set forth in § 9, in which the rate of the E.H.P. and of the 
corresponding I.H.P. attains a maximum, while the curve reaches a 
minimum and, thereafter, steadily ascends. 

4. Coefficients of Similar Vessels. — Consider two similar ships of 
different sizes at corresponding speeds. Let nt, 5, i.h.p, and k^ and 
M, Sl^, (§ 19), I.H,P. and K, be the area of immersed midship sec- 
tion, speed, indicated horse-power, and midship section coefficient 
of the small and large vessels respectively. Then : 

k^mx .-7— r and Ar=iWxTTjro» 


k _m I.H.P, I 

K" M ^ lApT ^ T^' 

By § 19. J/= m X l\ and, by § 24. I.H.P. = i.A.p. x /*, there- 

/> ^ 72 ^ ^ X /I '^ I . ' . ^ = A,, 

/. e., for similar ships, at corresponding speeds, the values of the 
midship-section coefficients are identical. This reasoning and 
conclusion apply also to the displacement coefficient. 

This property of these coefficients is utilized, for similar ships. 


in obtaining the required I.H.P. of a proposed vessel from the 
curve of coefficients of another whose performance is known. 
The coefficient for a given speed of the latter will be also the co- 
efficient for the corresponding speed of the former. From these 
particulars, the curve of coefficients of the proposed vessel may 
be constructed and its I.H.P. curve derived therefrom ; or points 
on the new I.H.P. curve may be obtained directly by substituting 
the coefficient and its corresponding speed in either of the two 
I.H.P. formulae under consideration. 

5. Limitations of the Fotmulce. — From the preceding, it is appar- 
ent, that, in a strict sense these formulae apply only to similar 
ships, at corresponding speeds, with the same frictional quality of 
immersed surface, the same efficiency of machinery, and within 
that range of speeds only, in which the total resistance is propor- 
tional to the area of wetted surface and to the square of the speed, 
u e.^ in which wave- resistance has but a minor part. Beyond these 
limits, the formulae become largely arbitrary, of no absolute value, 
and of service relatively only as measures of I.H.P., when, in the 
selection of proper coefficients, judgment and experience dictate 
suitable allowances for conditions which are not within their scope. 
Despite these restrictions, they have, when skilfully handled, their 
place in comparisons and in the approximate calculations of pre- 
liminary designs. Of the two. the displacement formula furnishes 
the nearest approach to the true resistance, in comparing some- 
what dissimilar ships. 

22. I.H.P. From Area of Wetted Surface. 

In 1880, the late Dr. A. C. Kirk brought before the Institution 
of Naval Architects, a method of analyzing the form of a vessel, 
of ascertaining readily and in close approximation, her area of 
wetted surface, and of computing the required I.H.P. from the 
latter. This process, known as ** Kirk's Analysis," is as follows : 

I. Analysis of Form and Computation of Wetted Sutface, — 
As shown in Fig. 9, the immersed hull is reduced to, and repre- 
sented on, a convenient scale, by a *• block-model" of uniform 
draught, having a parallel middle body, symmetrical and equal tri- 
angular fore and after bodies, and dimensions as follows : 

Length = AB = length of ship. 

Draught =r EC^ draught of ship, mean. 

Breadth = EK^=s area of immersed midship section of ship -4- 
mean draught, EC, 



Displacement = that of ship ; therefore 

AH^ GB = (displacement in tons x 35) -5- area of immersed 
midship section of ship. 

Length of fore body = length of after body = BH ^ AB — AH* 

BE^BK^ {EH^ + i5i7^)i . 

The area of wetted surface of the block model is that of its sides 
and bottom. For example, consider a proposed ship with data as 
follows: Displacement, 850 tons; length, 180 feet; breadth, ex- 
treme, 32 feet; mean draught, 11.5 feet; area of immersed mid- 
ship section, 275 square feet. Then : 

f . 

AH^ (850 X 35) -^ 275 -= 108 , 

BH^ 180— 108= 72'; 

EK^27S-^ 1.15= 24'; 

£^5 = (T2»+y2»)*=73'; and 
Area of bottom = 108 X 24 = 2,592 sq. ft. 

" sides =B 2 X II 5 X 36= 828 

" entrance and run = 4 x 11.5 X 73 =» 3.358 

«« " wetted surface, total = 6,778 





Fig. 9. 

The surface thus obtained exceeds that of the actual ship by 2, 
3 to 5, and 8 per cents , respectively, for full, ordinary, and very 
fine hulls. Since the error is small and in excess, the surface given 
by the block model is usually taken without deduction. 

2. Determination of I,H,P. — With regard to this, Dr. Kirk as- 
sumed the resistance to vary as the surface and as the square of 
the speed; that the I.H.P. would vary, therefore, as the speed 



cubed and that 5 I.H.P. per 100 square feet of wetted surface would 
drive, at 10 knots, a hull of usual form and efficiency. For fine 
vessels of maximum efficiency, the estimate was reduced to 4 I.H.P. 


S' f JO J J JJ^ /J j4 ^^ 

Fig. 10. 

and, for bluff ships, it was increased to 6 I.H.P. per 100 square feet 
of surface. Expressing the above mathematically, we have : 

LH.P. ms6 X A X S* X .00001. 


Comparing this with 


LH,P, = ^ X Z^« X 5» 

LH,R^^xMx S\ 

it is apparent that the th ^ ormulae are fundamentally the same, 
since D^ and M are, in similar ships, proportional to the wetted 
surface A ; and b (the LH,P, estimate per lOO square feet of sur- 
face) corresponds with Cand AT. Dr. Kirk's formula is, therefore, 
subject to the same limitations, and has, in general, the same char- 
acteristics, as the displacement and midship section formulae. It is, > 
however, superior in this, that, instead of quantities proportional 
to the wetted surface, the latter is, itself, employed. The formula i 
was not designed originally to cover other than moderate speeds. ' 

3. Rate Curves, — In using Kirk's Method, the values of the co- 
efficient b, for various speeds, may be plotted on speeds as ab- 
scissae, as " Rate Curves," giving thus, like the curves of coefficients 
previously described, a ready means of comparison. A curve so 
constructed is shown in Fig. 10, the data being as follows : 

Single-screw Cargo Steamer. 

Dimensions of Hull 340'X42'X29' 

Draught 15' lo*' 

Displacement 4f5<x> Tons 

Midship Area 604 sq. ft 

Length of Entrance.. , 79' 7^ 

Angle of Entrance 14®, 9' 

Surface 20,850 sq. ft 

Propeller-blades Tip flush with water. 

23. I.H.P. From Model Experiments. 

This method of obtaining, approximately, the resistance of a 
proposed ship, at a given speed, has been described generally in § 8. 
There is reproduced herein a photograph of the interior of the United 
States Experimental Basin at Washington, D. C, showing a model 
in process of testing. The water-surface of the bisin is 470 feet long 
and 43 feet wide, with a depth, from top of coping to bottom of 
basin, of 14 feet, 8 mches. Wooden models are u.sed, the standard 
length being 20 feet. The towing, dynamometrical, recording, 


and brake apparatus, are of the most modern type. The driving 
power of all mechanism is electricity. 

Recapitulating, with regard to model experiments, we note : 

1. Conditions. — The immersed hull of the model must be, in all 
respects, geometrically similar to that of the ship and their dis- 
placements must be proportional. If / be the ratio of like linear 
dimensions of ship and model, then comparison can only be made 
between resistances at a given speed 5 of model and the corre- 
sponding speed 5 X /* of the ship. Again, if, at the speed 5. the 
resistance of the model be r, then at the speed 5 x /*, the resist- 
ance of the ship will be r x /*, less the correction, noted below, 
due to friction. 

Further, if the model be tested alone, the resistance measured 
will be net^ i. ^., that of the hull only. If hull and screw interac- 
tions are to be included, a model propeller, supported and revolved 
by independent mechanism and developing a thrust equal to the 
total resistance, must follow immediately astern of the model, ad- 
vancing with the latter, but wholly disconnected therefrom. 

2. Determination of Hull Resistance. — The total resistance of the 
model is measured, as above, by trial. Now, owing to diflferences 
in length, speed and quality of surface, its frictional resistance and 
that of the ship will not be the same per unit of area of wetted 
surface ; and further, frictional and wave-resistance do not vary at 
the same rate at various speeds. The former must therefore be 
treated independently. 

From the model's total resistance, its frictional component is 
subtracted, the latter being calculated by the formula (§ 3), 

R, ^fAS\ 

The remainder multiplied by /• will be the residuary (wave and 
eddy) resistance of the ship. The frictional resistance of the latter, 
calculated by the formula, as above, and added to this residuary 
resistance, gives the total resistance of the ship at the speed 


In using the formula for frictional resistance, the coefficients/ 
and exponent n may be selected, with due regard to the conditions, 
from Tables III. and IV., both of which are derived from Froude*s 
experiments upon planes (§ 3\ the former referring to the original 
tests in fresh water while the latter i> modified for ship-surfaces in 
salt water. Th^ value of the wetted surface. A, may be obtained 
from Kirk's Analysis or from formulae such as : 



B\^1 * 

5=5,[i+(/-o.2)(^) ]; 



in which 5 = wetted surface, S^ = reduced wetted surface (the 
longitudinal integration of the rectified lengths of sections), / == 
prismatic coefficient, Zr = length, ^= beam, /f= draught, jD = 
displacement, and AT is a coefficient whose value is usually 15.6, 
but which changes thus : 

As ^/-A^ varies from 2.0 to 3.5, 
A" varies from 15.63 to 15.89. 

Table III. 

Frictional Exponent and Coefficient. (Froude.)— Fresh Water. 

of Sur&ce. 





Fine sand 

Medium sand 
Coarse sand... 

3 fe«t. 








Length of Surface. 

8 feet. 






2. 00 
















50 feet. 








Table IV. 

Frictional Exponent and CoEFFiaENT. (Froude.) — Salt Water. 

Length In ft. 



Length in ft. 



































































♦ Durand, Trans, Soc. Nov, Arch, and Mar. Eng*s^ "895. 
f Taylor, TVans, Soc, Naw, Arch, and Mar, En^*i, 1^3. 


3. Determination of I.H.P. — The total resistance of the ship R^ 
in pounds, at the speed S x /*, having been thus determined, we 
have by § 11: 

Hull resistance {lbs,) = R ; 

r, rrr^ I^ X Si^ X 6.O8O 

E.H.P. =x —. ; 

60 X 33,cxx> 
and by § i8 : 

/ // p „ JTlf^'fl 

Propulsive coefficient^ 

the value of the latter being assumed. 

4. Resume, — ^The fundamental truth of the law of comparison, 
as applied to the resistance of ships by William Froude, has been 
so fully demonstrated, during the last twenty-five years, as to leave 
no doubt with regard to the importance and value of its principles. 
In respect of model experiments, conducted in accordance with 
this law, it may be said that they form an efTective, and, in some 
cases, essential, aid in designing. In fact, without the information 
thus obtained, it would be a matter of extreme difficulty to make 
a reasonable estimate of the power required to drive proposed ships 
of abnormal form, such as the Russian Popoffka^ and Uvadia and 
the U. S. Ram Katahdin, 

In dealing, however, with a proposed vessel of known type, 
model experiments hold a minor place, as compared with the actual 
performances of similar, or approximately similar, ships. Thus, one 
of the most eminent and successful of the shipbuilding firms of 
Great Britain, which has had, for eighteen years, an experimental 
tank, has made it also an invariable rule to hold progressive trials 
of all the vessels built at its works. Again, the number of such 
tanks, public or private, is still exceedingly small, and but limited 
information as to the laborious and costly tests conducted therein 
has been made public ; yet the governments or individuals who 
have carried on this work, have had no monopoly of the successful 
ships of the world's fleets. 

Comparisons by model-experiment, in the present stage of de- 
velopment of the system, are subject to error, of more or less 
moment, as follows : 

{li) Scale of the Experiment. — The ratio between the estimated re- 
sistance of the ship and that recorded in a model experiment is so 

VOU 3CXU.— 15. 


great, that an error in the latter is magnified by thousands in a 
total resistance which, for even a large ship at high speed, is, rela- 
tively, but small. It is apparent, that, for trustworthy results, only 
very delicate mechanism, calibrated with great accuracy, and 
operated with much care, will serve. 

{b) Frictionai Resistance, — Froude's experiments as to this factor 
of the total resistance, were conducted upon plane surfaces, mov- 
ing at uniform speed throughout, and of a length which did not 
exceed 50 feet. While his results are beyond question, the esti- 
mate of the frictionai resistance of a hull therefrom is open to 
some doubt, since her immersed surface is not plane; its speed, ow- 
ing to the obliquity to the line of motion, is not uniform throughout; 
and the results of experiments on surfaces of 50 feet are applied to 
ships whose length may exceed 600 ft. 

[c) Viscosity of the Mediitm, — The Law of Comparison assumes a 
perfect fluid without viscosity. Water approximates closely to, but 
is not, such a fluid. While the influence of such viscosity as exists, 
may, doubtless, be neglected with ships, its effect upon small 
models, at low speed especially, should be appreciable. 

24. The Extended Law of Comparison. 

The Law of Comparison (§ 8), strictly construed, applies only to 
the wave and eddy-making resistances of ships. Under these re- 
strictions it is difficult of, and limited in, application. In practice, 
when used for estimating the I.H.P. for a proposed vessel, its pro- 
visions are modified by assumption that : 

(a) It applies to the total resistance including, therefore, that 
due to friction. 

{p) The I.H.P. is proportional to the total resistance. 

(^r) Ships whose dissimilarity is not excessive, may be compared 
by the extended law without unreasonable error. 

The justice, from a practical view- point of these assumptions, 
depends upon the judgment exercised as to the ships and speeds 
which are compared. Consider two vessels not precisely similar 
and of diflferent displacements. Let I.H.P. and tJt.p. be the pow- 
ers, S and s the correspondmg speeds, R and r the total resistances 
at said speeds, C and c the propulsive coefficients, iV=6.o8o -4- 
(60 X 33,000), and /= ratio of linear dimensions. Then, under the 
extended law : 


IH.P. R S N c ^ ^ ^1* ^Ji ,i 
t,n.p. r s N C r s r s 

The second member, as above, is simply the mathematical state- 
ment, in detail, of the first. If rt can be shown to be equal for all 
practical purposes to the final member, the justice of the extended 
law is apparent. This equality necessitates the regioval of the fac- 
tors involving N, C and c, and the replacement of the expressions 
including Ry r, S and s, by their equivalents under the extended 
law. The numerical factor NIN is equal to unity and disappears. 
Consider : 

1. The PropulsTve Coefficients, ?i^ included in the tdiCiot, c/C, If 
Css f, this expression vanishes. It has been shown (§ 18) that the 
value of these coefficients depends, primarily, on engine-friction, 
propeller-resistance, slip, and hull and screw interactions ; and, 
secondly, upon the speed of the engine relatively to its maximum. 
Equality between them will, then, require that the engines and pro- 
pellers compared shall be generally similar in type and efficiency; 
and, further, that, at the " corresponding " speeds used, that of each 
engine shall be relatively the same, with regard to its maximum. 
While it is not always possible to obtain a close approximation to 
these conditions, comparisons should be avoided between engines 
of long stroke, driving a large propellor at low velocity, and engines 
of high rotational speed with small screws; and,Jurther, between a 
speed much below the normal for a small ship and a corresponding 
speed which is close to the normal for another of large displace- 
ment. With due care in selection, however, the error may be kept 
small and in excess, thus giving a margin of safety in the estimate* 
and the coefficients may be considered, therefore, as practically 

2. Frictional Resistance, — The transformation, as above, of the 
expressions referring to total resistance and speed, involves the ex- 
tension of the law of comparison to skin- friction. If this be per- 
missible, then the ratio of the frictional resistances of the two ships 
will be (§§3, 19): 

r,^ y ^ a ^ V ^ d^ ^ s" " d^ ^ ^ " ^ ^ ' 

i. e.f if the Law of Comparison is to be extended to this element of 


the total resistance, — ^ must = /'; and, to obtain this equality, 

the exponent n must = 2, owing to the value (f) of the displacement 
power. Under perfect conditions, the value of n is not 2 but 1.83. 
This difference in the values may mean, in certain cases, a very con- 
siderable error in that of the total resistance; but any fouling of 
the skin, incident to service, will decrease the difference and cause 
the value of the exponent to approach 2. As a rule, the error is suf- 
ficiently small to be within the limits of a closely approximate esti- 
mate of I. H. P. for a proposed ship and the transformations referred 
to are permissible in practice. 

3. Hull Dissimilarity. — The amount of this dissimilarity which is 
allowable is a question to be answered by the judgment and ex- 
perience of the designer, since undue differences will require a 
modification of the results obtained under the Extended Law. It 
may be observed, as to this, that, as pointed out in § 20, the block 
coefficient is not a measure of the fineness of hulls, which are not 
known, from other sources, to be similar. 

4. Resume, — It will be seen that, if the Extended Law be ap- 
plied skillfully, its inherent error is practically negligible, and that 
when estimating the performance of a larger, from that of a 
smaller, vessel, the discrepancies due to (1) and (2), as above, are 
in excess and give thus a margin of safety. Assuming the equality 
of the coefficients of propulsion and the extension of the Law of 
Comparison to skin-resistance, we have, for two approximately 
similar ships of different displacements: 

By§ 8. 

By § 19. r s 
R^Pxr and 5=/*XJ.*. 

I.H.P. rP sli ,^ 

--r— = — X — = /*. 
t.n,/f, r s 



By §8, 

and by § 19, 



- (f ) •■ 


5 iD\^ 

The latter equations express the conditions of the Extended 
Law in terms of I.H.P. displacement, and speed, thus giving forms 
of ready application to the data of any specific case. 

25. I.H.P. BY Speed and Power Curves. 

These curves give the most trustworthy means of powering a 
proposed ship. Consider : 

I. Construction, — The curve is constructed on speeds as abscis- 
sae with I.H.P. as ordinates. That shown in Fig. 8 is plotted 
from the progressive trials of the U. S. Gunboat Yorktown at speeds 
of 4.4, 10.6, 14.75, and 16.6 knots. Now if, instead of one vessel 
driven successively at various speeds, there had been four ships, 
identical in every respect, each running at a different one of these 
four speeds, it is apparent that precisely the same curve could have 
been laid down from their performance and that it would have 
served to show the I.H.P. of any one of the ships at any speed be- 
tween 4.4 and 16.6 knots. 

Again, assume the four ships, A, B, C, D, to be still exactly 
similar but of various displacements and that a progressive trial be 
had with each. The results would give four difTerent curves, each 
representing the performance of the hull-type on a different dis- 
placement; and, from each of these, there might be derived, by the 
concluding formulae of § 24. the curve given by the type at any 
one of the other three displacements, or, in fact, by a similar ship 
of any displacement. For example, from the curves of B, Cand D^ 
there might be deduced curves corresponding with that given by 
At and the new curves, B' , C, D', would be identical with that of 
A, were it not for the small but inherent error of the Extended Law. 

Finally, if the I.H.P. of various approximately similar ships of 
various displacements at various speeds be known, and it be desired 


to estimate the I.H.P, for the general type at a new displacement, 
there may be deduced, by the formulae as above, from the results 
of each trial, the data lor a series of points on a curve representing 
the probable performance of the average type on the new displace- 

2. Use. — The practical use of the derived curve is illustrated by 
Fig. II, which shows that drawn during the design of U. S. gun- 
boats, Nos. S and 6, now the Machias and Castine, which were 
sister ships of 1,050 tons, 2,ooD I.H.P. and 15.75 knots maximum 

speed. The centers of the blank circles mark the position of 
points of the derived curve deduced by the Extended Law from the 
known performance of approximately similar vessels of various dis- 
placements reduced to 1,050 tons as a standard. Naturally, owing 
to the approximate nature of the law, both in its origin and exten- 
sion, these "spots" did not all fall upon the fair curve, which, 


drawn through or between them, would represent the average per- 
formance of the hull-type at this displacement. The close agree 
ment between the upper limit of the curve and the results obtained 
on the contract trial will be observed. The curve drawn to ascer- 
tain the required power for the U. S. battleships Kearsarge and 
Kentucky^ recently added to the fleet, gave even better prediction 
of the results finally attained. While the required speed of these 
battleships was 16 knots, they developed 16.816 and 16.897 knots, 
respectively, on their contract trials. 

These differences indicate an essential principle of successful 
design, namely, to add the power required for at least half a 
knot more than the desired maximum speed, no matter how ac- 
curate the estimate for the I.H.P. of the latter may appear to be. 
A ship and her machinery are not a vast integer formed, as in 
a mould, at one mighty operation. On the contrary, the com- 
pleted vessel is a complex structure of many thousand parts, built 
by hundreds of men. The human factor enters at every stage, 
and on her speed trial, she represents, not what her designer laid 
down, but the better or worse approximation which remains after 
the balance of error has been struck. In effect, there are, there- 
fore, no such structures as precisely " similar " ships and any rea- 
soning based upon exact similitude is essentially that of the doc- 
trinaire. No two vessels of the same lines, and with the same 
engines and screws, have, with the same I.H.P., exactly the same 
speed. It is, in fact, impossible to construct two ships, identical 
in form, although they may be built side by side, from the same 
drawings, at the same time, by the same mechanics. The difference 
is especially marked when sister ships are constructed by different 
builders. The following extracts from the data of contract trials, 
illustrate this : 

Vessel. Displacement. { I.H.P. 

B. S. Indiana I 10,225 | 9,545.16 

B. S. Massachusetts 1 10,265 10,177.21 

B.S.Oregon | 10,250 | 10,992.77 




The two former of these sister ships were built at the same 
works, the latter 'at another. The results of their trials show that 
it is not reasonable to expect that any ship shall make exactly her 
intended speed ; and a designer of experience will select, on his 
curve-sheet, a point rather worse than the average, giving her a 


margin of about half a knot. A 19-knot ship would thus be pow- 
ered for 19.5 knots. If, in the end, the balance of error be against 
her, she will make at least her contract-speed ; and, if it shall favor 
her, she may reach 20 knots. 

3. Resume, — The speed and power curve, properly constructed 
and used with judgment, gives unquestionably the most accurate 
method of estimating the I.H.P. required for a proposed vessel. 
Every speed-trial plotted for it represents a test, on the open sea, 
of a ship, her engines, and propellers. It is, in no sense, an ex- 
periment upon any scale, small or large, and its results require 
neither transformation nor correction for any element of size or re- 
sistance. The record is one of actual performance, giving unas- 
sailable facts upon which, in full confidence, prediction may be 
based as to the probable qualities of approximately similar vessels. 





Heat Engine Cycles Analyzed. 

Part I, 

Prime movers are useful when they produce motion in required 
directions against resistances. Nearly all our machines which in 
general constitute the resistances to prime movers are designed to 
be operated through an applied forceful rotary motion ; therefore the 
prime movers that are to be of most service to us in our ordinary 
working operations must develop forced rotary motion. By far 
the largest number of these rotary motion prime movers come un- 
der the head of Heat Engines. These heat engines may be di- 
vided into two classes : 

(a) Those that do work by utilizing the expansion of a sub- 
stance when changing from the liquid to the gaseous state. 

(b) Those that do work by utilizing the expansion of a perfect 
gas, this expansion being caused in some mysterious way by ab- 
sorption of heat. 

The engines of class a usually consist of two parts, a part for 
the production of the vapor, and a part for the utilization of this 
vapor, converting an increase of volume into a forced rotary mo- 
tion, in ordinary language, of a boiler and an engine proper. The 
amount of work that can be done with a given amount of heat by 
a prime mover of this class, is definitely known within certain lim- 
its, when we know how much liquid can be converted into vapor 
by this heat, and the relative specific volumes of the liquid and re- 
sulting vapor. It therefore depends chiefly on the liquid chosen, 
and, of course, on the mechanical efficiency of the system as a 
convertor, or, as we may say, on the design of the machine. 

The cycle of operations is: (I.) Add heat to liquid and produce 
vapor. (II.) Allow vapor to expand to as low a pressure as pos- 
sible, and then discharge it either as vapor or as reconverted liquid. 

Thiscycle is unchangable except in incidental details. On the con- 
trary, however, when we employ a perfect gas to which to apply our 
heat, and whose expansion gives us our work, we may have a large 
range of different cycles or series of operations that may be per- 
formed on or by the gas in question. The amount of work done 
by our expanding gas due to the initial application of a given 


amount of heat will depend on the manner of heating, method of 
expansion, ultimate disposition of the gas, and, of course, on the 
mechanical efficiency of the machine for performing the operations 
desired, and will depend not at all on the gas chosen. In short, 
the varying amounts of work that may be done will depend solely 
on the cycle itself. It is therefore evident that there is consider- 
able importance in knowing just how the cycle can effect this 
change of ultimate useful work for given heat supplied. 

In the actual application of any cyclic principles we find various 
other questions beside the ultimate useful work that demand at- 
tention and study. For example, one cycle requires a larger vol- 
ume of gas to do same work as another ; a larger engine is therefore 
necessary; some cycles operate under higher temperatures than 
others ; some through wider ranges of temperature and pressure. 
Many other questions might be cited, but enough are given to show 
that it is necessary that we study the cycles as such, and obtain a 
statement of every question in terms of the cycle, before we begin 
the consideration of the mechanical difficulties involved in its car- 
rying out. 

It is possible to cause a similar mass of perfect gas to pass 
through each of the cycles, and obtain an equation for every vari- 
able entering into the cycle in terms of the initial conditions and 
the quantity of heat supplied. For example we can write 

For cycle I. Efficiency = E =^fj {Hfi^) 
For cycle II. E=fj,{H^C") 

For cycle III. E=f,„ ^Hf") 

For cycle «, E=f^ {Hf^) 

where H^ is heat supplied, and C a constant. 

We thus get a series of curves of efficiency, one for each cycle, 
in terms of the same variable, and get exact relations of the cycles 
regarding efficiency at a glance. Instead of efficiency we might 
have chosen the final volumes or the maximum temperatures. 

It is the object of this paper to consider the various cycles as 
above outlined and cause one pound of air to pass through each 
of the cycles under ideal conditions, and to determine every cyclic 
variable in terms of H^, and arbitrary initial conditions. To pass 
from ideal conditions to practical ones we need only apply a cor- 
rection factor. 

In what follows we shall not consider how the heat is applied or 
abstracted, the mechanisms involved, nor the practicability of the 



The following cycles will be considered 



FIG. 1. 

FIG. 2. 

Let Fig. I be a P.V. and Fig. 2 be a 0<p diagram for the cycle. 
Then we have : 

From B to C. Addition of heat isometrically from atmospheric 

From C to D. Adiabatic expansion to atmospheric pressure. 
From D to B. Cooling at atmospheric pressure. 





V. 4>. 

FIG. 3. PIG 4. 

We have : 

From B to C. Addition of heat isometrically from atmospheric 

From C and D. Adiabatic expansion to point above atmos- 
pheric pressure. 

From D and E. Cooling isometrically to atmospheric pressure. 

From E to B. Cooling at atmospheric pressure. 








X ^^y:. 


V * 

FIG. 6. FIG. 6. 

We have : 

From B to C. Addition of heat isometrically from atmospheric 

From C to D, Adiabatic expansion to below atmospheric pres- 

From D to E. Cooling isothermally to atmospheric pressure. 

From E to B, Cooling at atmospheric pressure. 



fk;. 7. 

FIG. 8. 

We have : 

From B to C, Addition of heat isothermally from atmospheric 

From C to D, Adiabatic expansion to pressure below atmos- 
phere such that we get, 

From D to B. Cooling isothermally to original volume and at- 
mospheric pressure. 



V <t> 

FIG. ». FIG. 10. 

We have : 

From A to B. Adiabatic compression from atmospheric pressure. 

From B to C. Addition of heat isometrically. 

From C to D, Adiabatic expansion to atmospheric pressure. 

From D to A. Cooling at atmospheric pressure. 









V 4> 

no. It FIG. 12. 

We have : 

From A to B. Adiabatic compression from atmospheric pres- 

From B to C. Addition of heat isometrically. 

From C to D. Adiabatic expansion to pressure above atmos- 

From D to E. Cooling isometrically to atmosphere. 

From E to A. Cooling at atmospheric pressure. 




V. * 

FIG. 13. PIO. 14. 

We have : 

From A to B. Adiabatic compression from atmospheric pres- 

From B to C, Addition of heat isometrically. 

From C to D, Adiabatic expansion to pressure below atmos- 

From D to E, Cooling isothermally to atmospheric pressure. 

From EX.O A, Cooling at atmospheric pressure. 


FIO. 1 5. 


We have : 

From A to B. Adiabatic compressioa from atmospheric pressure. 

From B to C, Addition of beat isometrically. 

From C to D, Adiabatic expansion to pressure below atmos- 
phere such that we get, 

F'rom D to A, Cooling isothermally to original volume and at- 
mospheric pressure. 




T\a. 17. 

FIO. i& 

We have : 

From A to B. Adiabatic compression from atmospheric pressure. 

From B to C. Addition of lieat isopiestically. 

From C to D. Adiabatic expansion to atmospheric pressure. 

From D to A, Cooling at atmospheric pressure. 











— — -^'""^"'^ 


no. 19. 

FIG. ao- 

We have : 

From A to B. Adiabatic compression from atmospheric pressure. 
From B to C, Addition of heat isopiestically. 
From C to D, Adiabatic expansion to pressure above atmos- 

From D to E. Cooling isometrically to atmospheric pressure. 
From jfc to A, Cooling at atmospheric pressure. 





• ^^^ 






__— ---^^ 

V 4> 

FIG. 21. FIG. 22. 

Wc have : 

From A to B. Adiabatic compression from atmospheric pressure. 

From B to C. Addition of heat isopiestically. 

From Cto /?. Adiabatic expansion to pressure below atmosphere. 

From D to E, Cooling isothermally to atmospheric pressure. 

From ^to A. Coolincr at atmospheric pressure. 









V • 4> 

FIG. 33. FIG 94. 

We have: 

From A to B. Adiabatic compression from atmospheric pres- 

From B to C, Addition of heat isopiestically. 

From C\o D. Adiabatic expansion to pressure below atmos- 
phere such that we get, 

From D to A, Cooling isothemally to original volume and at- 
mospheric pressure. 







FIG. 25. 

FIG. 26. 

We have : 

From A to B. Adiabatic compression from atmospheric pressure. 

From B to C Addition of heat isothermally. 

From C to D, Adiabatic expansion to atmospheric pressure. 

From E to A, Cooling at atmospheric pressure. 







FIG. 27. 

FIG. 2 8. 

We have: 

From A to B. Adiabatic compression from atmospheric pressure* 
From B to C. Addition of heat isothermally. 
From C to D. Adiabatic expansion to pressure above atmos- 
From D to E. Cooling isometrically to atmospheric pressure* 
From Eto A. CooUng at atmospheric pressure. 

TOU XXII.-— x6. 





FIG. 29. FIG. 30. 

We have : 

From A to B, Adiabatic compression from atmospheric pressure. 

From B to C, Addition of heat isothermally. 

From Cto D. Adiabatic expansion to pressure below atmosphere, 

From D to is. Cooling isothermally to atmospheric pressure. 

From E to A, Cooling at atmospheric pressure. 

CYCLE nrc. 


PIG. 31. FIG. 3 a. 

We have : 

From A to B, Adiabatic compression from atmospheric pressure 

From B to C. Addition of heat isothermally. 

From C to D, Adiabatic expansion to pressure below atmos- 
phere such that we get, 

From D to A, Cooling isothermally to original volume and at- 
mospheric pressure. 




FIG. 33. 


PJO. 34. 

We have : 

From A to B, Adiabatic compression from atmospheric pres- 

From B to C Addition of heat at variable pvd. 

From C to D, Adiabatic expansion to atmospheric pressure. 

From D to A, Cooling at atmospheric pressure. 

Cycles v., A, B and C may have the same modification on Cycle 
V. as II. A^ B and Chave on III., for example. 

The Atmospheric or Vacuum Cycles. 

Here all the cyclic operations take place at or below atmos- 
pheric pressure. 



FIG. 35. 

FIG. 36 

We have: 

From A to B, Addition of heat at atmospheric pressure. 

From B to C Cooling isometrically. 

From C to A. Adiabatic compression. 




FIG. 37. 

FIG. 38. 

We have : 

From A to B. Addition of heat at atmospheric pressure. 

From B to C, Adiabatic expansion. 

From C to D, Cooling isopiestically. 

From D to A, Adiabatic compression. 

CYCLE vn. 

V. 4> 

FIG. 39. FIG. 40. 

We have: 

From A to B. Addition of heat at atmospheric pressure. 
From B to C, Adiabatic compression to such a pressure, that 
we get 

From C to D, Isothermal compression to original state. 




FIG. 41. FIG. 42. 

We have : 

From A to B. Addition of heat at atmospheric pressure. 

From B to C, Adiabatic expansion. 

From C to D, Cooling isometrically. 

From D \o A, Compression adiabatically. 


We might have many modifications of these but as a discussion 
of the type throws sufficient light on the variations considering 
the importance of the cycles, these modifications will not de dis- 

Cycle I. 

Fig. I. Fig. a. 

Let H^ be the heat added from B to C 

Let £7 be the specific heat of gas at constant volume, and here 
assumed constant for simplification. It is probably a variable, but 
so assuming it gives unmanageable formulae. A correction may 
afterward be applied, if desired. C^asheat to raise one pound 
gas to 1° F. at constant volume. 

Let z/j be the volume of the gases at point B of the diagram, 
u ^., before heating and expressed in cubic feet. 

Let/j be the corresponding pressure in pounds per square foot. 

Let 7j be the corresponding temperature in absolute degrees 

Then will the increase in temperature be given by 

^« ^^ Hi 


T,= T,+ "^\ (I) 

Since volume is constant from B to C, 


From (I) 

Since this quantity 


I + 

will enter into many of our equations, let us denote it by 

I + > 7- = -^ 


P, = P„^- (2) 

The adiabatic relation 




But/ =/, by hypothesis, hence 

^.=^.(^;)^=nW- (3) 

Another adiabatic relation gives 
remembering /^ = /j and substituting the value of 7^ 

(I vrzl ^-Y+Y 1 

-^) r == T;^ Y = 7,^r. (4) 

Let H^ be the heat discharged. Then 


Where C = specific heat at constant pressure and assumed con- 
stant. Hence substituting 

The work done in heat units will be 

W^H,-H, (6) 

^//^-CJ^ixLi). (7) 

And in foot pounds 

W = /lH,-C^T,{xl-i)l 

This work of expansion could have been obtained by tempera- 
tures and by integration as well. 
We have 



c,_c.+ x^ 

••• ^' C,T, - CJ, - CJ, + (7.7, - i [(^^^•) 7; + (A^^) 7;] 
We know also 



A = A- 

in heat units. This second term is the area of the rectangle be- 
tween \\ . _ , and ■; ^* and lying below atmos. 
( / = atmosphere f^v=v^ ^ ^ 

phere is not available for work. 

By integration W ^ J '^ pdv » area between expansion curve 
and axis of volumes. The expansion is adiabatic. 

^^ ^ Ju ^ r- lb"- I J.; 

I -r I -r ■ 


c — c 

^ C 

=/(7^(7^— 7 J in foot pounds. 
Subtracting the rectangle p^v^ — v^ we get 

in foot pounds, or in heat units 


as before. 

Before going farther let us apply a test to each'of the states B^ 

C, D from the law of perfect gases 

An r, 

~ = A 

^ h 

X " T.x -X~ 


Pj>i_Pt,K{X)y _p^v,_ 

T ~ I ~ T' 

' nX)y '" 

hence these are identities as they should be. 

Denote the volume swept through or volume range by R^. 

then will 


K=v^-v,-=v^-v,=v,[Xy-\-]. (8) 

Whence mean eflfective pressure 

M.E.p. = .;^-/.'?^r_'-,^'(^'r-». „) 


The entropy range is given by 

f. - *'» = C, log. ^' = C, log. X. (I I) 

Mean effective temperature 

The temperature range 

Rr=T^-T, = ^^. (13) 

The pressure range 

Whence we may write a mean effective volume 

MEV=-^ = /^^'-^'^'^'^'-')^ (15) 

ie,= 7;-7;=7;(x-i). (i6) 

These results are here tabulated for reference and comparison 
with what follows : 

We might take the formulae* derived for mean effective tempera- 
ture, but as these were the results of a comparison of cycles, none 
of which ran below atmospheric pressure, it would be better to 
take another standard here. Let us take arbitrarily as the mean 
effective temperature one-half the sum of the mean temperature 
of heat addition and the mean temperature of heat abstraction. 

Cycle I. X^ i + ^ ^* 

Formula Reduced to Initial. 
Sjmibol. Formula as Derived. Conditions. 

A Arbitrary p^ 


^ " ^ 


A^A A^6 

R R 



^, n «'» 

T, 7;+^' T,X 



«'^ ^x:y ""X' 


V r^v 

H, CIT,-'Q CTlXy - 1 ) 

^ /(^.-^,) /{II,-CT(A:i-i)} 

"i^ School of Mines Quarterly, XXL, 4, 1900. 


Formula Reduced to Initial 
Symbol. Formula as Derived. ConditioDS. 

W . C,UXy- I) 

H\ ^\ 

Ry^ ^d - ^. <^y - I) 

M.E.P yf J ^ H,- c^ux^ - .) . 

v,{Xy - I) 

K A - A Pl^ - ' ) 


W ^ I H, - C,T,{X'y - i) 

jy fiSX-i\ ) 

^* C,\og,^ C,\og,X 




*\ /?♦ / ^\ c.iog^sr 

Rt T,-T, T,{X-i). 

Cycle I. A. 

Fig. 3. Fig. 4. 
We have as in Cycle I. for point C. 

«'. = «'» (0 

A=A^ (2) 

T. = T,X (3) 



Then from the adiabatic relation 



^- = ^.(^)'- (4) 

7-.= 7:(;^)T. (5) 


Substituting, values of /^ and 7^ in (4) and (5) we get 

If we write 



V, = z^,(^«)^" (8) 


7;-7;xrV---^»<^">. (9) 

I'. = «'tf " J'»(^«)^ (jo) 

t; - 7; ";• = 7;(x«)^. (II) 

Let us apply the perfect gas law to the points -5, C, -O and E. 

A^6 r, 


A^c_A^^6_ r, 

'P ^ 1 1 ^^ ''^» 



Heat is abstracted in two parts, the first at constant volume 
from D to E and the second at constant atmospheric pressure from 
Eto B, 




- cjix4 [^ - 1] + ^p'^^ V^''^" - A ^'^^ 

The work done in foot-pounds is 

W^J{H,-H,) (13) 

.-. w = j{h, - cjixnf ( ^ - i) + c;r,\{x4 - 1]). 


.-. jE= I — 

cjix4{^-^ - i) + <r,7;[(x«)r _ I] 



The mean effective pressure 

M.E.P. = 




K='^a-'^i = ^',i{Xny- I]- 





H, - C,T,{Xnf{^- - i) + CJ,\(,Xn)y - i] 




M.E.V. = 


= / 

^, - C,T,{Xnj'(^^ - I ) + C;7;[(A'«)r _ I] 

PIX- I) 
As before, the entropy range is 





Taking the mean effective temperature as the mean of the aver- 
age heating temperature and the average cooling temperature 




CJ,{Xn)y[[- I ) + C^T,i{Xn)y- i] 

The temperature range is 

R,= T^-T,= T,{X-i). 
The pressure range is 





M.E.V. = 





H, - CMXn)y[\ - I ) - c;r, \_{x4 - i] 


Pix- I) 

Tabulating these results we get for 

Cycle I. A. 

Symbol. Formala as first derived. Formula reduced to initial condition. 

/^ . . . . Arbitrary p^ 

Vf^ . , . , 






A f 



^.•■^.(■ + c1i) ^'-^ 



Symbol. .Fonnula as first derived. 
A Pc>Pd>Ph' • 




Formula reduced to initial condition. 















^2 ■ ■ C,{T,-T,)+C,{T-T,)=C,T,iXnji(l-i)-C^T,[(X»)y-i} 

H,-CJ,iXn)y(l-i)-CJ,[{Xn)y-i-] \ 


• • • • A 

//, H^ 





^» ^»[(-^«)^-] 

ff-CJ,{Xnr{^-- i) - C,T,[{Xny- i] 


R^ C^\og,^ C,\og,X 


.E.T. = \ ( 




H,+C,T,{Xni>\^^- y)-CJ,\_{x4y-i^ 

A -A A(^- 

IV J ^, - CMXn)y{^^ - i) - C;7; [iXnj- I] 

= J 

lYA . lit, V**««*-y •••/ 



t;-^; 7;(^-i). 



Cycle I. B. 
Fia 5. Fio. 6. 

As the operations up to the point C, i. e., after addition of heat 
are the same as in Cycle I., we may assume these results : 

7: = 

Choose/^ so that 
Expansion CD gives 




m'? - T^x- 

\ If A';' 




From the isothermal relation along DE, 




"•(a)'' ""»^'(a)'' "'^''"^ '^ ""a- ^5^ 


-*• — •'d 




/, = /j by hypothesis 

.' .V. 







Applying the perfect gas law to the various points 




A^. A^»^ A^, 


A^d A^'*('-^«)^ A^ 


7; {" ( jr«)v 

-,= ''* = -/? 




A^. „ 


= R 


— 1 

~ J'. 



Heat is abstracted in two parts, first a part isothermally and 
second a part at atmospheric pressure. That part abstracted 
isothermally is extremely difficult to calculate without the aid of 
the 0^ diagram and its relations. 

The entropy range along BC has been found to be 

^* = f. - f» - c, log. p = c; log.x. (I I) 


Now it is evidently the same so far as entropy range is con- 
cerned whether we cool at constant pressure from EtoBor heat 
isopiestically from B to £, thus 


9e-9h^ ^ploge^- (12) 

Hence the entropy range for the isothermal operation will be 

given by 

94-9.= C, log.^- C; log^ ^* (13) 

= C. log.^- C^ log. [x\ (^^)"' ]. (14) 

This latter isothermal change taking place at temperature 7^ = 
7^ the heat of cooling will be given by 


Hence the total heat abstracted is 



f^^-CJ,7,-T,)+T,\c,\os,X- C^\ogJ^^ (IS) 


C, log. ^'^^^' = C; log, A-> + C^ log, «v-' 

= (f.iog.A'+((r.-Qiog.«. 





— I 

c ^C 




CJ, [ ^-^^ - I J + \-^- log. «^--^^ (16) 

The work in foot-pounds is 


£^— I — ^^ 

.-.£■= I — 



^.= ^,-n = ^»[(^«)^-i]- 


.-. M.E.P. =/ 


^._c;r.p;>^-,] + .^'t^«^^og.« 

rC^— G, 


Rp=' Pc— Pd"^ Ph^ — Pa- 



.-. M.E.V.=/ 


The mean effective temperature being the mean of the heating 
and cooh'ng means is given by 


where R^ is same as in previous cycle. 

VOL. xxn.— 17. 





B . 

Tabulating : 

Cycle I. B. 

Sjmibol. Formula as first derived. Formula reduced to initial conditions. 

p^ Atmosphere Atmospheric p^ 

V, Arbitrary z/. 


R R 


^ ^6 ^6 

^» • • • • ^» ( ' + ct) ^^^ 


Pi Pt>pH>0 A >/d> O 

T. T l^"! y 




PJ « 

P. A / 




A " 


T. T, 



*i m •••••• 

C,{T, - 7;) + 7; [ (f. log,^- <r, log, ^] 

fv. Aii,-^^) 

{Xn)l ,T T.iXn)^ 

/H'.-^.^^^r-O- -?-'°^.«^^ 

I ^ 


+ ^» -„- log. n<^o^ 








{■!'.- V,) «'»[(^«)* - I] 




I 1 

H,-CJ\ J _,J_7;N^log.« 



Pc — Pd Ph^ — Pd 


= / 


» 6. 

log. v-' 



. Ju>. la*******!' jp I 

[(-, . .] 

2 log^xl ^> + ^^^» I ^'"^^ - • I + "^" ■' >°g.»''-"^ 



R T,-T, r\.-&\ 

Cycle I. C. 

Fig. 7. Fra. 8. 

Assume ail results to point C from Cycle I. 

A=A^ (I) 

». = «'» (2) 

T, - T,X (3) 

From the adiabatic CD, 

This adiabatic must meet the isothermal from B in point D^ 

v.--.% (S) 

Equate (4) and (5) 

tc IC ^. * \ 

V, (A)v= ., 


/, ' V, 


•••A = A^*^ = A: (6) 

This is the pressure at which the isothermal through B will 
meet the adiabatic through C. Its corresponding volume is 

A -i- 

«'d - ^» — X = ^»'^''"' (7) 

7; - r,. (8) 

The heat abstracted by the isothermal cooling is found as before 

from Ofp relation. 

9c- n^^ ^. lo&e / = ^. l^^g.^ (9) 


The work done in foot-pounds is 

w^Aff, = ^.) -7(^1 - T,c, log. JO. 

The efBc ency is 
The volume range is 




v,{Xy-i - I) 

The pressure range is 


\ x-i-y 

The entropy range was found R^ = C, log. A' hence 
M.E.T. - i (^' + ^') - ^^^ ,;,g^ ^(AT. + r.C.log.^) 












^r=7:-r,= 7;(^-i). 

Cycle I. C. 

Formula •* first derived. ^<*«"^ J^^wtS *^'*^ 

■ /j Atmospheric Atmospheric /^ 


«/j Arbitrary 

7^ A«'» 

^» '/? •■ 




















Fonnula as first derived. 

Formula reduced to initia] 





^. ^»0+A) 




v,Xy-' v,Xy-' 


T,C, log. J<= 

T,C, \oz.X 

J{H,-H^ J{H,-T,C,\og,X) 


I — 


v,{Xy-^ - I) 




vlX'r'^ - I) 

\ ^Y-l/ 


pA^— \\ 


C,\og' C^\o%,X 


Mrioe A'+ ^») 

...T,-T,...: UX-x) 



The Metallurgy of Gold. A Practical Treatise on the Metallurgical 
Treatment of Gold- Bearing Ores, Including the Assaying, Melting 
and Refining of Gold. By M. Eissler. Fifth Edition. London, 
Croslcy Lockwood and Son; New York, D. Van Nostrand Co. 317 
Illustrations. Pp. 638. Price, $7.50. 

The first edition of this work, which appeared in 1888, comprised but 
300 pages, considerably smaller than those of the present volume. 
Shortly after the appearance cf the third edition, in 1891, the success 
attending the introduction of the cyanide process in the South African 
gold fields and elsewhere, urgently demanded for this process an im- 
portant place in any treatise on the metallurgy of gold. This need was 
met, in a measure, by the publication of the fourth edition. 

But perhaps the most marked improvements in the work have been 
made in this last edition, of July, 1900,' resulting from revision, rear- 
rangement and enlargement, together with better typography, quality of 
paper and iliustratio? s. The presentation of the subject matter is much 
more satisfactory. Concerning the milling, and especially the cyaniding 
of gold ores, the author is enabled to speak with authority by reason of 
exceptional opportunity for observation, and experience gained, during 
an extended professional visit to the Witwatersrand, a field in which, at 
the outbreak of the Boer war, seme of the most advanced practice was to 
be found. 

In his rearrangement at the text Mr. Eissler has adopted the following 
plan of subdivision : Part I. Introductory : being mainly a brief re- 
view of the physical and geological features of a few of the gold-produc- 
ing districts; chemical examination of 01 es; methods and appliances 
used in hydraulicking and dredging for gold. Part II. Milling of Gold 
Ores : stamp batten ies, plate and pan amalgamation ; the working of the 
gold mill, with examples; other crushing appliances; statistics and ex- 
amples of milling costs. Part 111. 7he Treatment of Gcld'Bearing 
Ores : concentration ; methods of roasting various refractory ores ; 
roasting and smelting of pyritic ores. Part IV. The Hydro Metallurgy 
of Auriferous Ores : the Plattner process of chloriuation, lixiviation and 
precipitation ; later cMorination processes ; electro- metallurgy of precious 
metals. Part V The Cyanide Process of Gold Extraction: erection 
of a cyanide plant ; synopsis of process ; conditions, results and costs of 
plant and treatment ; Siemens-Halske cyanide process ; examples of op- 
eration at various works ; chemistry of the process. Part VI. The Test- 
ing and Refining of G* Id; melting and assaying; cupelling, parting and 
refining gold bullion. This arrangement is calculated materially to as- 
sist the reader in attacking the mass of information contained in the book. 

In treating of the application of cyanide solutions it would seem as 
though the description of the chemistry of the process should h aV e pre- 
ceded the discussion of erection and operation of plant. It may be said, 
also, that the importance of the subject of causes of imperfect extraction, 
both as regards defective condition of the solvent and the possible pres- 


qnce of interfering substances, warrants a fuller presentation than that 
given on pages 512 and 563. 

On page 180, and again on pages 5 jy and 538, the author touches upon 
an interesting subject, viz : the adoption of dry- instead of wet-crushing, in 
preparing the ore for subsequent metallurgical treatment. This subject 
might well have been expanded somewhat in view of the attention given 
to it within the past few years, in connection with cyaniding ard chlori- 
nation, by engineers in Australia, New Zealand and South Africa. Dry- 
crushing may, indeed, often be unavoidable under such conditions as 
prevail in western Australia. But even in South Africa, where hitherto 
the water question has not been a specially serious one, attention has 
lately been directed to the possibility of improving the present methods 
of working certain kinds of ore, by introducing rolls and ball mills, or 
other pulverizers, followed immediately by cyaniding the entire product, 
instead of wet-5 tamping, succeeded by the three processes of plate amal- 
gamation, cyaniding of sands and slime treatment. The question of out- 
put is an important one, but it is a significant fact that, not long since, 
during a run of 23,000 tons of ore at the mine of the Luipaards Vlei 
Estate, the output per horse-power, crushed dry, was greater than that 
usually accepted on the Rand for wet- crushing. 

In Chapter V. of Part I., the reader might desire fuller information as 
to the modes of saving gold by hydraulicking and dredging, although, 
strictly speaking, these subjects do not come under the head of metallur- 
gical treatment. The author has omitted, in Part I., to refer to the devel- 
opments in Alaska during the past few years, and, in connection with the 
tellurides of gold, to the occurrence and treatment of these ores in the 
Cripple Creek district of Colorado. 

The chapters on hydro-metallurgy are well arranged and, with those 
on the cyanide process, are among the best in the book. The matter re- 
lating to the roasting and smelting of pyritic ores is also good, though the 
author has perhaps given to some of the older furnaces more than their 
due share of attention, and has omitted to describe one or two types of 
roasting furnace (such as the Pearce Turret) which are doing good work 
in this country. 

Within the limits of a single volume, large as is the one before us, it is 
impracticable to present fully all parts of the subject, as, for example, in 
notiqg minor variations in modes of treatment, which may result from 
slight differences in mineralogical constitution of the ores, or differences 
in local conditions. Realizing this, it might be useful to have appended 
footnotes of reference to articles of special value contained in monographs 
and the transactions of technical societies ; or, a similar result could be 
reached by giving after each chapter a bibliography, comprising a list of 
the more important articles bearing up(>n the subject. Upon the whole 
Mr. Eissler's work covt rs quite satisfactorily the broad field of gold metal- 
lurgy, and in its new form will be appeciated by all who are interested in 
the industry. 

R. P. 

The Coal and Metal Miners^ Pocketbook of Principles, Rules, Formulas 
and Tables, sptcially compiled and prepared for the convenient use of 
mine officials, mining engineers, and students preparing themselves for 
certificates of competency as mine inspectors or mine foremen. Sixth 


edition; revised and enlarged with original matter. Scranton, Pa., 
The Colliery Engineer Co., 1900. Pp. 637* including 110 pp. of 
tables. Price I3.00. 

The first edition of this work appeared in 1890. Although somewhat 
expanded in 1893 and subsequently » no very radical alterations were made 
until the appearance of the prejent edition. In issuing this, better paper 
and typography have been employed, and the subject matter, aside from 
the copious tables and glossary, has been increased in volume from 319 
to 4S« pages. 

Among the sections which are practically new are those on electricity 
and electrical plant, mine haulage, ore-dressing and the preparation of 
coal, and drilling and blasting, with special reference to tunnehng and 
shaft-sinking. The principles of electrical engineering and the applica- 
tion of electricity to mine work are well presented in considerable detail, 
and are accompanied by a number of good illustrations. Thirty-three 
pages are given to this subject. The part devoted to electric signaling 
for mines is also worthy of commendation. Under the head of ore- 
dressing will be found a very serviceable 25 -page outline of the subject, 
with illustrations comprising descriptions of crushers, rolls, crushing 
mills, stamps, sizing and classifying apparatus, jigs, etc. Some omissions 
are to be noted here, however, such as with regard to concentrating tables 
of the endless belt and other types ; also the treatment of slimes on tables 
of various kinds. 

The section on the ventilation of mines has been rewritten and greatly 
improved, both in quality and gei eral arrangement, and a good outline 
description is given of mechanical ventilators, with principles of con- 
structim. The revision and enlargement of the subjects of prospecting, 
opening of mines and methods of working, is also noticeable. Finally, 
the tables of logarithms and trigonometric functions are improved by re- 
arrangement and condensation. 

Throughout the book a more logical order and arrangement of the text 
has been observed, so that it is no longer necessary, as formerly, in some 
cases, to re'er to several different parts of the work in looking up a given 
subject. We think this handbook takes rank with the very best works of 
the kind, and is superior to most of them. Upon the whole, it may be 
cons dered as almost a new work, and it cannot fail to meet with very 
general appreciation. 

R. P. 

Chemistry — Its Evolution and Achievements, By Ferdinand G. Wiech- 
MAN»r, Ph.D. New York, Wm. R. Jenkins, 1899. i2mo, pp. vii, 
+ 176. Price, $1.00. 

This is a sketch of the growth of chemistry as a science, from the early 
ages when knowledge of the properties of matter was very rudimentary 
and ungeneralized, through the development of theories and discovery of 
laws governing chemical phenomena to the great achievements of to day. 

The vast amount of work included in the upbuilding of the chemistry 
of the present is treated in this brief volume in masterly fashion, for no 
important link in the chain of development has been slighted, and an ad- 
mirable sense of proportion amongst the various discoveries in the science 
is evinced throughout. Thus it is told in most interesting manner, how 


the love of jewels, precious metals and the artificial enhancement of per- 
sonal beauty — feminine beauty — ^all contributed their quota to the knowl- 
edge of certain chemical prop>erties of the substance involved. Some 
empirical knowledge of metallurgy, ceramic making, dyeing and fermen- 
tation was early acquired by man, and many s()ecu)attons upon the nature 
of mafer were indulged in by the philosophers of the cultured nations of 
antiquity long before the more ambitious -alchemists labored to transmute 
base metal into gold or discover the elixir of life. 

The influence of the Arabians in the seventh and eighth centuries is 
dwelt upon in spreading a knowledge of alchemy through Europe, and the 
next step of note is the impetus given to the subject in the sixteenth cen- 
tury through the application of the known facts of alchemy to the needs 
of medicine. Here extended observations and experiments served to 
place the subject on a surer footing, and th - transition from fantastic 
alchemy to rational chemistry had begun. 

Bacon's deductive method of reas mrn< and Boyle's contributions to 
experimental data were potent factors in the seventeenth century in ad- 
vancing the science. Scheele and Priestley did much for chemistry by 
their work, notably in the discovery of oxygen, by which so great a field for 
exploration was opened up in 17 74. Then, in 803, Dalton propounded his 
Atomic theory, which, though so low o be accepted, was the foundation 
of modem chemistry. The many discoveries and advances in chemical 
theory following Dalton's discovery are here most interestingh discussed, 
as well as chemit^al nomenclature and the hterature of the science. 

A careful history of analytical chemisiry's development follows, along 
with a discussion of each of the great departments of applied chemistry 
that displays most scholarly research and a broad grasp of the subject. 
More than half the book is devoted to the achievements of modern chem- 
istry, both in respect to the scientific theories and the practical benefits 
which have accrued from its wonderful advance. 

The book is very useful to all who are interested in the progress of the 
day, and will be found most suggestive to those who follow the subject 
more closely, for it abounds in references, so useful to the earnest student. 
It really seems as though nothing of importance has been overlooked. 

The style is buoyant and easy, the whole appearing as a continuous 
narrative, presenting a sketch which is in itself complete, yet inviting one 
to examine more fully the vistas whose recesses contain such wonders and 
whose exploration is such a delight. 

W. D. H. 

A School Chemistry. By John Waddell, Ph.D., D.Sc. New York, 
The Macmillan Co., 1900. Pp. 278. 

This text-book is intended for use in high schooh and in elementary 
classes in co leges, and is in many respecis an excellent introduction to 
the study of chemistry. It is particularly well adapted to the use of stu- 
dents preparing for technical or scientific schools, where just such knowl- 
edt^e as is given in this book is required and where the instruction 
begi ts with the metallic elements. For students who are not going 
further in chemistry the book has a serious defect — too little space is 
given to the metals, 45 pages compared with 326 on the non-metallic 
element i. 


The information is decidedly up to date; the recent discovery of 
argon in the atmosphere being prominently mentioned as well as the pro- 
duction of acetylene from calcium carbide. The quantitative relations 
of chemical combination and reaction are emphasized both theoretically 
and experimentally, a decided advance over most elementary text-books. 

Many chemical reactions are illustrated by manufacturing processes 
which tend to stimulate the interest of the student and point out the 
practical application of the science. 

The book is of convenient size, well printed, illustrated and bound. 

E. H. M. 

Topographic Surveyings Including Geographic, Exploratory and Military 
Mapping, with Hints on Camping, Emergency Surgery and Photo- 
graphy. By Herbert M Wilson, Geographer, Ur-ited States Geolog- 
ical Survey; Member American Society of Civil Engineers. Illustra- 
ted by 18 engraved colored plates and 181 half-tone plates and cuts, 
including two double page plates. New York, Wiley & Sons. Pp. 
XXX + 900. 8vo, cloth. 1 3. 50. 

This comprehensive work consists essentially of three separate books or 
treatises: Topographic Surveying, Geudetic Surveying, and Practical 

The general subject of Topographic Surveying is further subdivided 
into three parts: Plane Surveying, Hypsometric Surveying, and Map 
Construction, with a preliminary discussion of the relations and distinc- 
tions bet«veen the essentially arbitrary classification of topographic, geo- 
graphic, and exploratory surveys. 

In like manner the ge- eral subject of Geodetic Surveying is treated 
under the headings of Terrestrial Geodesy and Astronomical Geodesy. 

The divisions and subdivisions under the main headinidS into which the 
general subject has been divided, are clearly defined, and the topic treated 
in each article is indicated by a few weli-chosen words, printed in heavier 
type. The article representing the topic is again divided into paragraphs, 
or sections, which are further clearly defined by having the significant 
words indicating the subject matter treated in that paragraph printed in 
italics. Thus the reader, seeking for some special information, may pass 
easily and rapidly from chapter to article, and from article to paragraph, 
by merely glancing at the distinctive headings, which appeal easily and 
instantly to the eye. 

** The methods elaborated are chiefly those which have been developed 
in recent years by the great government organizations ; " but the author 
has "endeavored to go beyond these, and, guided by personal experience, 
to adapt them to the most detailed topographic, as well as to the crudest 
exploratory surveys." The reputation and experience of the author in 
these great government surveys entitle him to speak with authority. 

The reader will at once notice the absence of ** any detailed descrip- 
tion of those instruments or methods which are e aborated in works on 
general surveying. The volume is devoted practically to higher survey- 
ing, and presupposes a knowledge of all the more elementary branches." 

When it has been found necessary to treat subjects which are essenti- 
ally elementary, in order that all the facts of a given topic may be suit- 
ably grouped, to insure clearness and conciseness, such subjects have been 


briefly described ; and, consistently with the author's expressed purpose, 
but little space is devoted to the elaboration of the mathematical proc- 
esses by which the various formulae have been derived, since their devel- 
opment in detail is to be found in most of the well known works on gen- 
eral surveying. All the essentials, however, have been iiitroduced. 

The author departs in many ways from the usual procedure adopted in 
text books on surveying. "Instead of describing the instruments and 
their uses independently, each is described in that portion of the text in 
which its development in field surveying is most prominently mentioned." 
Thus the subject of the engineer's transit, to which several pages of de- 
tailed description are devoted in most treatises on general surveying, here 
receives brief treatment in the chapter devoted to Traverse Instruments 
and their Methods. The author states, **The transit is an instrument 
but little used by topographic surveyors, and is so commonly employed 
in ordinary surveying, and described in text-books and catalogues, that 
its description will not be elaborated here." 

Again, instead of bringing the tables together at the end of the vol- 
ume, each is inserted in that portion of the text which relates to its use. 
Both the advantages and disadvantages of this method are obvious. 

The subject of map construction is clearly and logically treated, and 
very fully illustrated. Tables for the projection of maps are also included, 
and their uses explained. 

Chapter VI., on lopographic Forms, and what may, in a sense, be con- 
sidered its complement, namely Chapter XX., on Topographic Drawing, 
are among the best portions of this work. The subject of conventional 
signs is treated with unusual fullness and is profusely illustrated, and a 
complete glossary of terms is given, together with several pages of refer- 
ence works. 

Those portions of the book which deal with the astronomic and geo- 
detic features of surveying, are carefully arranged and well treated, and 
a list of reference works is appended. 

An interesting portion of the book is Part VII., which is devoted to 
Camping, Emergen* y Surgery and Photography. This will doubtless 
prove useful lo tlwjse who are responsible for ihe organization and well- 
being of campm^ and exploratory parties. 

In conclu» ion, the writer c onsiders the book before him a distinctly 
noteworthy work, giving evidence through^>ut that the author has drawn 
largely from his own varied experience for his material and methods. It 
is rich in suggestions, and full of practical hints. The numerous illus- 
tranons are not of the conventional tvpe usually to be found in works on 
surveying, but are quite distinctive and up-to-date, and are taken, in 
nearly everv ca^e, from actual practice Much valuable material, and 
many beautiful illus rations and maps, have been taken from the published 
works «>f the United Siatt s Geo^o^ical Survey. 

Fmally, the element of cost of the various kinds of surveys, and where 
possible, of the different portions of surveys, form an unusual and valu- 
able »ea^ure of this work which will appeal to engineers and surveyors. 

On the whole, the write considers this book valuable as a work of 

A. B. 

/^^^^ Jena Normal 

C. A. F. Xahlbaum's 

CheHically Pura Organic 
iMrgailc Chemlcth. 

Chas. Schleicher & 



tStli St and 3nl Avs., New York. 

Chemicals t Chemical Apparatus 


SCHOTT & GENOSSEN, Manufacturers of the famous Jena Glass 

C. A. F. KAHLBAUM, Chemicals. 


CARL ZEISS, Microscopes. 

FRANZ SCHMIDT & HAENSCH, PoUriscopes and Spectroscopes. 

DESMOUTIS & CO., Purest Hammered Platinum. 


GREINER & FRIEDRICH'S, Fine Stopcock Ware. 

MAX KOHL, Physical Apparatus. 

TENDER, Reagent Bottles with Black Indelible Letters. 

KAVALIER'S Famous Bohemian Glassware. 

The tbove Grms >re knowD to fumuh the bat gocxli in their rapectire linet 
We arc their repreBentadTci for Ihc United State* and Canada and carry a coniider 
able Mock of tbew goodi on band. Kohl'a Apparaloi and Slender's Bottles are 
nported to order ooly . 

Jena Glass the Olass of the Fnture 


Book Binders 

6 and 7 EAST 16TH ST., 



Books bound in any style. 
Single volumes or in quantities. 

Woodbridge School, 


School of Mines Preparatory School. 

417 Madison Avenue, 

Between 48th and 49th Streets, NEW YORK CITY 

Twentieth Year Begins October ist, igoi. 

THE school is well equipped with physical and chemical laboratories, in which the 
students are required to perform a complete set of experiments illustrative of their 

recitations in physics and general chemistry. A special labc^atory is devoted to 
qualitative chemical analysis for advanced students. 

Five hundred Students of Columbia School of Mines have been instructed in the 
NVoodbridge School. Also a large number have been prep&red for Massachusetts In- 
stitute of Technology, Stevens Institute, Sheffield Scientific School, Lawrence Sden- 
ific School, Troy Polytechnic Institute, Cornell University, and the Classical, Medical, 
and Law Departments of Harvard, Yale, Columbia and Princeton. 

A summer schiX)! for students who have failed in June begins August 13th smd 
coaches men for the Fall examinations. All classes are limited to five. CoUege men 
are coached in Freshman and Sophomore Mathematics and Quantitative Analysis. 

An advanced course for older students prepares them to enter the second yeir class 
Last June one of our students received his degree in Electrical Engineering in 
three years. 


J ^ I 6~^0' II 

Vol. XXII. No. 3. APRIL. i9ot 






A. J. MOSES, Prof, of Mineralogy. JOS. 8TRUTHBRS, Ph.D.» 

J. P. KEMP, Prof, of Geology. Lecturer in Metallurgy. 

R. PEBLB, Adj. Prof. Mining. I. H. W00L80N, Instructor in 

A. D. P. HAMLIN, Mechanical Engineering. 

Adj. Prof, of Architecture. W. H. PRBBDMAN. B.E., 

R. B. MAYBR, Instructor in Drawing. 8. O. MILLER, Assistant in 

B. WALLER, Analytical Chemist. Mechanical Engineering 

Manafflriff Editor, R. E. MAYER, 
Business Manager, 8. O. MILLER. 


Antipyrin and its Derivatives. By David C. Eceles, B. S -^59 

A Laboratory Classifier. By Henry S. Munroe 303 

A Laboratory Slime Table. By Henry S. Munroe 306 

The Serpentines of Manhattan Island and Vicinity and their Accom- 
panying Minerals. By D. H. Newland 307 

The Siluric Fauna near Batesville, Arkansas. By Gilbert Van Ingen 318 
A Method of Cyclic Anal3rsis of Heat Engines. By Charles E. Lucke, 

M. S 329 

Experiments with Commercial *' Dry Cells '* for Secondary Standards 

of E. M.F. By Henry St. John Hyde, Ph.B. A.M 366 

Discussion of Paper of Mr. Scherr on Reduction Roasting. Its value 
for Arsenic Expulsion from Copper Ores and Mattes. By 

H. M. Howe 381 

Charles Hill Bergen ; 382 

Alumni and University News 384 

Book Reviews '. 389 



Registered at the New York Post Office as Second Class Matter. 

All Rcraittancee should be made payable to Order of "The School of Mines Quarterly. 



Tool and Die Steel of the Best Quality, Sheet Steel, 
Spring and Machinery Steel. Steel for Bdge-Toola and 
Hardware of Every Description. Polished Drill Roda 
and Pine Wire. Steel FoiKinga, Coiled Springs, etc. 

WcncTO OIBcc and WHehouic, Eaitern Omca and WaiebouM 

M and 68 So. OUnton Street, 480 Fearl Street, 


16,000,000 BARRELS 








Cmit, Angle, Cktck, Ctbis, V, Sa/iiy, He., are mnnraclured 
of the besi (team metal and are ful'y guaranieed. If jon 
want tbe be$i, ask ^our dealer for valvts mai-ii/actmrtd hy 
TA Jtukini Broihirt. Rcniember all genuine are ttanped with 
Trade Mack like cat 

JENKINS BROS., New York, Boston, Phflailalph(«,Ch>c«oo 




Vol. XXII. APRIL, 1901. No. 3. 

[Contributions from the Havemeyer Laboratories, No. 40.] 



The Introduction of Antipyrin. 

The year 1883 marked the opening of an important era in or- 
ganic chemistry. Dr. Ludwig Knorr, of Wiirzburg University, 
while working upon phenylhydrazine condensations, produced a 
body which he called di-methyl-oxy-chinizin. Later this proved 
to be phenyl -di-methyl-pyrazolon, now known in medicine under 
the name of antipyrin. The discovery of this body opened up an 
entirely new field for organic research work and since that time 
several hundred pyrazolon bodies have been produced. 

Antipyrin, in 1884, was investigated physiologically by Knorr 
and Filehne and its antipyretic power discovered. From 1887 
to 1889 many favorable reports were printed by such European 
medical authorities as Cesari, Carrara, Mendel, Bonne, Iwanoff 
and Sawadowski. In the United States, about the same time, Drs. 
Hare and Gleason, of Philadelphia ; Dr. Park, of Buffalo, and 
others, after clinical tests, pronounced it a very useful remedy. 
Every physician of standing in the United States has now probably 
prescribed it for some one of its uses, but as an antipyretic it 
finds its most important application. It is often used as a styptic, 
as a local anaesthetic solution, with or without cocaine, and as an 
anti-rheumatic combined with salicylates. 

The chief ill effect charged against this remedy by clinicians has 

VOL. XXII —18. 250 


been its weakening of the power of the heart. The same objection 
has been made against all such synthetic drugs as acetanilid, 
phenacetine and exalgin. Dr. McCall Anderson, in the British 
Medical Journal for December, 1894, says in this connection " that 
out of 189 observers who report on antipyrin, no fewer than 138 
(or 73 %) have never observed any ill effects at all worth mention- 
ing. It may be given in safety in large doses, even in the case of 
children, although the initial dose must be small and it must be 
slowly and cautiously increased, the patient always being carefully 

Antipyrin is now recognized in the British Pharmacopoeia under 
the name of " Phenazone," and is therefore given place in the dis- 
pensatories under this title. Pharmacists find it incompatible with 
many ingredients used in prescriptions where antipyrin is indi- 
cated. Carbolic acid, chloral hydrate, phenyl urethane, /9 naphthol, 
or resorcin, when triturated with it form oleaginous liquids. Pow- 
ders containing it and sodium salicylate have to be wrapped in wax 
paper to keep them from deliquescing. Tannin and all tinctures 
containing an appreciable quantity of tannic acid are precipitated 
by antipyrin solutions. With tincture of iodine and compound 
solution of iodine it forms brown precipitates, and with sweet spirit 
of nitre or concentrated ethyl nitrite, it gives a green solution or 
precipitate according to the amount present. When used with 
quinine or caffeine salts it increases their solubility. 

The first patent on antipyrin was taken out in Germany on 
July 22, 1883, the number being 26,429. Later two other processes 
were patented as numbers 32,277, November 25, 1884, and 40,- 
377, November 4, 1886. Shortly after the first grant, Knorr re- 
ceived his patent rights in the United States. According to our 
law, these became void when the time limit upon the first German 
patent had passed. 

There are three distinct methods known for the preparation of 
antipyrin, each of which Knorr patented. The specifications of 
his first patent cover the best of these, and is described as the con- 
densation of phenylhydrazine and aceto-acetic ethyl ester, followed 
by methylation or an addition of the methyl radical to the prod- 
uct. The other two processes are less direct but the danger of 
their being used by other manufacturers led to their being pat- 
ented by Knorr. The second process is the condensation of 
symmetrical methyl-phenylhydrazine with aceto-acetic ester to 


directly form antipyrin. The third method is at present valuable 
only from a chemical standpoint, but it was patented as a protec- 
tive measure and consists in changing aceto-acetic ethyl ester into 
a compound known as acetone-dicarboxylic ethyl ester, condens- 
ing this with phenylhydrazine, methyjating, saponifying and finally 
heating the product to eliminate carbon dioxide, thus producing 

While antipyrin was selling at $\-Z^ an ounce, it was possible 
to profitably make it by the second process in spite of poor yield, 
but since the drop in price to about 18 cts. an ounce, it will not 
now pay to do so. 

Preparation of Antipyrin. 

The original method pursued in manufacturing antipyrin em- 
braces the following four steps : 

I. Formation of phenylhydrazine from benzol. 

II. Preparation of aceto-acetic ethyl ester from^alcohol and so- 
dium acetate. 

III. Condensation of phenylhydrazine and aceto-acetic ethyl 
ester to form phenyl methyl-pyrazolon. 

IV. Methylation of the pyrazolon compound and formation of 

Steps three and four were protected by the patent but one and 
two are of equal importance in its manufacture. 

In forming phenylhydrazine (QHjNHNHj), benzol (C^HJ from 
coal tar is changed into nitrobenzol (CgH^NO,) by the action of con- 
centrated nitric acid, then reduced to anilin (C^H^NHj) by nascent 
hydrogen released by the action of hydrochloric acid on scrap iron 
in accordance with the following reaction : 

C.H.NO, + 3H, = C,H, NH, + 2H,0. 

The next step consists in diazotising anilin with nitrous acid 
and reducing the resulting product with moist stannous chloride 
so as to form phenylhydrazine. On diazotising anilin hydro- 
chloride by aid of nitrous acid, diazobenzene chloride results, the 
equation being : 

C,H;-.NH2 . HCl + HNOg = C,H, • N : N • CI -f 2\{p. 

By adding a paste of stannous chloride to the clear mobile brown 
diazo-solution, it becomes creamy white and so stiff it can just 
be poured. A couple of hours standing insures complete reduc- 


tion ; neutralization with caustic soda follows and the phenylhy- 
drazine formed is then distilled under diminished pressure. 
The following are the equations : 

(1) CgH, . N : N . CI + 2H2= C,H,- NH • NH, + HCl. 

(2) HCl + NaOH =x NaCl + H,0. 

II. The formation of the aceto-acetic ethyl ester is accomplished 
by the action of metallic sodium upon ethyl acetate, followed by 
treatment with acetic acid. The ethyl acetate is obtained by act- 
ing upon sodium acetate with a mixture of alcohol and sulphuric 

Ethyl alcohol, added in small portions to sulphuric acid, pro- 
duces ethyl^sulphuric acid according to the reaction, 

C^H, . OH + H HSO, = CjH, • HSO, + H^O. 

This ethyl sulphuric acid is poured upon dried, fused and broken 
pieces of sodium acetate previously placed in a copper still. After 
standing twelve hours, the contents of the still are raised to the boil- 
ing point in order to form ethyl acetate which passes over as dis- 
tillate. Ebullition is kept up as long as any ethyl acetate is being 
formed. The collected distillate is then freed from water by redis- 
tillation over calcium chloride. The production of the ethyl acetate 
occurs as follows : 

C^H, • HSO, + NaOOC CH3 = C^H.OOC • CH, + NaHSO,. 

In order to form aceto-acetic ethyl ester, the ethyl acetate with 
small pieces of metallic sodium drawn into wire are placed in a still 
having a reversed condenser and left until the sodium has disap- 
peared after which a moderate excess of 50 per cent, acetic acid is 
added and the contents agitated. On standing, it separates into 
two layers, the upper one of which is subjected to fractional dis- 
tillation and the portion which passes over between 175-185° C. 

The reactions taking place during these operations, though 
quite complicated, have been worked out by Genther, Wislicenus 
and Claison. The sodium, replacing the hydroxyl hydrogen 
of some alcohol present in the distillate, forms sodium ethylate 
(CjHjONa). This ethylate then acts upon ethyl acetate producing 
an intermediate addition compound of the structure CH,C \ (ONa) 
(OCaHJ.^. The following equation shows this addition : 


CHj.CO OC,H,+ NaOC,H,=CH,C : (ONa) (OCjHj),. 

The intermediate ethoxy-sodio compound reacts at once upon 
more ethyl acetate, the resulting product being sodio-aceto-acetic 
ethyl ester (CH, • C (ONa) : CH . COOC.H,). 



CH3 C : (OC,H,), + (H,) CH • COOC.H, = 


CH, • C : CH • COOC.H, + 2C,H, OH. 

Addition of acetic acid replaces the sodium of sodio-aceto-acetic 
ester by hydrogen and generates sodium acetate. The formuhi for 
aceto-acetic ethyl ester is : 


The question of the tautomeric rearrangement of this ester from 
the enolic to the ketonic structure has been fully discussed in 
chemical literature. For most of its reactions, the ketonic struc- 
ture ofTers the best explanation, though in some cases the enolic 
form must be assumed. For convenience and because Knorr 
favored the ketonic form of this ester, it will be adopted in ex- 
plaining the formation of antipyrin. 

in. Condensation of phenylhydrazine and aceto-acetic ethyl 

When the ketonic group (C — CO — C) of aceto-acetic ethyl 
ester \y\lQ - CO - C]H/- COOC,H,) is treated with phenylhydra- 
zine, there occurs a condensation with rise of temperature and 
elimination of water. The condensed molecule is phenylhydra- 
zone-aceto-acetic ethyl ester. The change is as follows : 

CH3 • C CH, COOC,H, N - NH • C^H, 

II +11 

O H, 

Aceto-acelic ethyl ester. -f- Phenylhydrazine. 


+ H„0 



Phenylhydrazone ace'o- -f Water, 
acetic ethyl ester 

When the phenylhydrazine and the ester are mixed in the pro- 
portions of I to 1.3, the solution boils vigorously, globules of 

264 ^^^ QUARTERLY. 

water separate and float through the mass of the liquid which rap- 
idly changes in color from light yellowish brown to reddish brown. 
Upon warming this over a water bath for a couple of hours it 
darkens and thickens until it can just be poured. During the heat- 
ing, alcohol is eliminated, a second condensation takes place, and 
a closed ring results known as the pyrazolon ring. This con- 
densation is shown by the equation : 


H iOC.H, - 
N- N-C,H, 

Phenylhydrazone -aceto-acetic ethyl ester. 

CjH, • OH + CH3 • C ■ CHj CO 

N— N-C,H,. 

Alcohol. -|- Phenyl-methyl-pyrazolon. 

If we turn our formula upside down and number the groups, we 
observe that the phenyl occupies position i, the methyl position 3 
and the carbonyl position 5. while the whole ring constitutes pyra- 

(2) (I) 

N N - QH, 

CH3-C-CH2- CO 

(3) (4) (5) 

(I) Phenyl (3) methyl (5) pyrazolon. 

After the removal of the alcohol and water, the dark thick fluid 
is stirred and poured while warm into a little over its own volume 
of ether. In a few minutes reddish particles begin to separate and 
soon solidify into a dark reddish crystalline magma, insoluble in 
ether. This mass is allowed to stand until cold, when it is filtered 
and washed very thoroughly with ether until all coloring matter is 
removed and it is left as a white crystalline powder. 

IV. Methylation of the pyrazolon compound. 

This is accomplished by taking equal weights of freshly distilled 
methyl iodide, pyrazolon and methyl alcohol, mixing them together 
in sealed glass pressure tubes and heating to 100° C. in a specially 
constructed bomb furnace that allows of careful regulation. The 
temperature must not go much above 100^ C. or there will be great 
danger of explosion from excessive internal pressure due to sudden 


liberation of vapors produced by decomposition. At the end of 
three hours the containers are allowed to cool. 

According to Knorr, the chemical reactions during this mcthyl- 
ation involve first the direct addition of methyl iodide to the pyra- 
zolon molecule; second, the spontaneous elimination of hydriodic 
acid and, third, the rearrangement of the bonds connecting the 
atoms in the ring to form a stable compound. In the pyrazolon 
ring the nitrogen of the second position is trivalent, but the added 
methyl iodide makes it become pentavalent and forms an inter- 
mediate product with the structure 

I (2) (I) 

CH3.N N-C,H, 


(3) (4) (5) 

(I) Phenyl (3) methyl (5) pyrazolon (2) meth iodide. 

Almost immediately hydriodic acid (HI) is formed by the I 
atom leaving position (2) and uniting with one of the H atoms of 
position (4). This change necessitates the disappearance of the 
extra bond between positions (2) and (3) and the appearance of a 
dual bond between (3) and (4), thus restoring the N to tri valence 
as in the following : 

(2) (I) 

N N 

I I +HI. 


(3) (4) (5) 

(l) Phenyl (2:3) dimethyl (5) pyrazolon. 

When the tu^ s from the furnace have thoroughly cooled, the 
contained hydroiodide solution is removed into an open vessel. In 
order to avoid spattering, care must be taken in 0[>ening the tubes 
to allow the internal pressure to be gradually relieved. Water and 
sulphurous acid are next added and the solution boiled until most 
of the methyl alcohol has been driven off. The sulphurous acid 
acts upon the hydriodic acid forming free iodine, sulphur, water 
and basic (i) phenyl (2:3) dimethyl (5) pyrazolon. The latter com- 
pound is afttipytin. Washing this mixture with chloroform three 
or four times dissolves out the antipyrin. The chloroform solution 
is evaporated to a paste and just sufficient warm toluene added to 
dissolve it. Upon slowly cooling the antipyrin crystallizes out. 



According to Riedel, of Berlin, the production of antipyrin upon 
a commercial scale involves but one operation for condensation 
and methylation. His article states that phenylhydrazine lOO 
parts, aceto-acetic ethyl ester 1 25 parts, sodium methyl-sulphonate 
150 parts, sodium iodide 150 parts, hydriodic acid 5 parts, and 
methyl alcohol 100 parts, are' heated at 160-180° C. in an auto- 
clave under a pressure of 12-15 atmospheres for ten hours. A still 
later report from the same authority states that the sodium iodide 
is not necessary in this process, provided four to five times the 
above- quoted amount of hydriodic acid be used. 

The reactions taking place are, first, the formation of methyl 
iodide from the sodium methyl-sulphonate and sodium iodide, 

CH.SO^Na + Nal = CH,I + Na,SO,. 

Sod. methyl-sulphonate. Sod. iodide. Methyl iodide. Sod. sulphate. 

second, the methyl iodide reacts in its turn upon the phenyl- 
methyl-pyrazolon formed by the reaction of the hydrazine upon 
the ester, forming antipyrin hydroiodide. 

C,.H„N,0 + CH.I = C„H„N,O.HI 

Phenyl-mcthyl-pyrazolon. Methyl-iodide. Antipyrin hydroiodide. 

In order to isolate the antip} rin from the contents in the auto- 
clave, it is taken up with a mixture of alcohol and ether, filtered 
and washed, thus separating it from the by-product — sodium sul- 
phate. The alcohol and ether are distilled off, the residue dis 
solved in water and made alkaline, the solution extracted by chlo- 
roform, the chloroform distilled off and boiling toluene added, from 
which, on cooling, the antipyrin crystallizes. 

Knorr's Second Method. 

The second method pursued by Knorr in preparing antipyrin 
consists (I) in the formation of the symmetrical methyl-phenylhy- 
drazine and (II) condensing this with aceto-acetic ester to form 
antipyrin. This method differs from the other, in that here the 
phenylhydrazine, and not the phenyl-methyl-pyrazolon, is methyl- 

I. Phenylhydrazine possesses basic properties and, even in the 
form of its salts, is capable of reacting with organic acid chlorides, 
but not readily with metallic sodium. When benzoyl chloride 
(CjHjj* CO • CI) is added to phenylhydrazine sulphate (C^H^ NH- 
NH, • HjSOJ dibenzoyl-phenylhydrazine is produced with hydro- 


chloric and sulphuric acids as by-products. This body has the 


C.H. N(CO • C.H.) • NH(CO • C.H,). 

The replacement of the benzoyl for hydrogen in the hydrazine 
causes it to acquire acid properties, thus enabling metallic sodium 
to replace the remaining hydrogen. The sodium is gradually added 
to an alcoholic solution of the benzoyl derivative producing sodio- 
dibenzoyl-phenylhydrazine having the following formula : 

C.H, • N(CO • C.H,) N(Na)(CO • C.H.). 

The next step consists in the methylation of this and is accom- 
plished by heating it with methyl iodide for five hours in a sealed 
tube upon a water bath. The sodium of the sodio-dibenzoyl phenyl 
hydrazine is replaced by the methyl of the methyl iodide and forms 
methyl-dibenzoyl phenylhydrazine, the formula being* 

C,H.N(CO • C,H,) • N(CH,)(CO • C,H.). 

The white crystals obtained are then intimately mixed with an 
equal weight of solid caustic potash (KOH) and distilled. The 
distillate contains the symmetrical methyl-phenylhydrazine, the 
formula of which is : 


Potassium benzoate (KO COC^Hj) is the by-product. 

II. The condensation of the symmetrical methyl-phenylhydra- 
zine with aceto-acetic ethyl ester is stated by Knorr's patent as 
follows : " Equivalent quantities of ethylic aceto-acetate (aceto- 
acetic ethyl ester) and symmetrical methyl-phenylhydrazine are 
heated in an oil bath at 140° C. until water and alcohol are no 
longer evolved, the antipyrin formed is then extracted from the 
melt by hot water." This condensation reaction with the forma- 
tion of the pyrazolon ring is thus expressed graphically : 


H H + 

Symmetrical methyl-phenylhydrazine. 

O :H O CjH, 


Aceto-acetic ethyl ester. 


CH3-N N-C,H, 

CH, - C = CH - CO 

+ Hp + HO C,H, 

Antipyrin. Water. Alcohol. 

In this case where the substituted, in place of the unsubstituted 
I)henylhydrazine is used, the condensation works slightly different 
than in the first method, for here the ketonic oxygen of the aceto- 
iicetic ethyl ester, in order to form water, obtains one of its hy- 
drogens from an adjacent— CHj — group, necessarily changing 
this to a = CH — group in the antipyrin nucleus. 

Knorr's Third Method. 

The third method for the preparation of antipyrin consists in (1) 
the alteration of the aceto-acetic ethyl ester to acetone-dicar- 
boxylic ethyl ester, and (II) condensation of this with phenylhy- 
drazine. As a practical process this is useless, because it requires 
too many steps and, owing to secondary reactions, a large loss 
of product results. It is of interest from a chemical standpoint, 
showing as it does possibilities of organic group alterations. 

I. The alteration of the aceto-acetic ethyl ester (CHjCO'CHj. 
COOQHg) is accomplished by passing dry chlorine gas over it for 
half an hour. About 30 per cent, chlorine absorption occurs. 
The product is distilled; the fraction coming over at 188-189° C. 
is composed principally of gamma-chlor-aceto-acetic ethyl ester 
(ClCHaCO-CHa-COOCaHg). This chlor-ester, in ether or benzol, 
is then treated with potassium cyanide whereby a nitril is formed 
which, under greatly diminished presssure, distils at 145-160° C, 
The nitril has the formula CN-CHj-CO-CHaCOOQH^. Acetone- 
dicarboxylic ethyl ester (M.P. 134° C.) is obtained from the nitril 
in alcoholic acid solution by treatment with concentrated hydro- 
chloric acid whereby the nitril (CN) is changed to the carboxyl 
group (COOH), ammonia (NHg) and water (HgO) being split out. 
according to the usual saponification reaction. As this occurs in 
acidulated alcoholic solution the hydrogen in the carboxyl is re- 
placed at once by an ethyl group forming the acetone dicarboxylic 
ethyl ester (CaHgOOCCHj-COCH.COOQHJ. Its relation to 
acetone, from which it derives its name, is seen by a comparison 
of the following formulas : 

(I) (2) 

^^"-CHj '^^--CH, COOCjH, 

Acetone. Acetone-di-carboxylic ethyl ester. 


The two hydrogens of the acetone are theoretically replaced by 
— ^COOCjH^— groups. 

II. The condensation of acetone dicarboxylic ethyl ester is il- 
lustrated as follows: 

N-NC.H, o ,C,Hp 

Hj H CH,-C-CH,-CO 


Phenylhydrazine. -f Acetone di-carboxylic ethyl ester. 

N -N - C,H, 

CH, - C - CH, - CO + H.O + H • OC,H.. 

1 (3) (4) (S) 


/(.)PhenyI(S)Vyr-oIon (3)methyK _^^^^ + Alcohol. 

Vcarboxylic ethyl ester. / ' 

In order to produce*antipyrin from the pyrazolon condensation 
product, a methyl group has to be attached to the nitrogen num- 
bered (2) and the — CH, • COO • C Hj, group in position (3) must 
be changed to — CH,. 

The introduction of the methyl into position (2) is accomplished 
by adding methyl iodide and eliminating hydriodic acid (HI), the 
H coming from position (4) and the I from position (2), thus pro- 
ducing a compound with the following formula. 

(2) (I) 

CH3-N- N-C,H, 

! I 

CH,- C= CH - CO 

I (3) (4) (5) 


Saponification with acid changes the ethyl of — CH — COOC^H^ 
to H, thus producing a carboxyl in the group (— CH2 — COOH). 
On heating this, carbon dioxide (CO,) is evolved, leaving aCH, 
group for position (3) and giving us antipyrin. 

The Formula of Antipyrin. 

Knorr's elementary analysis gives the following figures for anti- 
pynn : 







70.2% 704% 

70.0% 69.9% 


6.4% 6.7% 

6.7% 6.6% 





The empirical formula which agrees most closely with these 
figures is C,jH^,NjO. Such a formula, however, throws little light 
upon the structure of the molecule, the groups present and their 
relation to each other. In order to understand why antipyrin is 
held to have the graphic structure assigned it by Knorr and here 

(2) (I) 

N N 

I I 

C = CH - C 

(3) (4) (5) 

CH3-N N-C,H, 

I I 

CH3 - C = CH - CO 

we must study its methods of formation, its reactions and its de- 
compositions with other bodies of known atomic arrangement. 
We shall show that the antipyrin molecule contains two nitrogens 
adjacent to each other ; that one of these nitrogens has a phenyl, 
the other a methyl group, attached ; that in the nucleus of anti- 
pyrin and attached to one of these nitrogens is a — CO — group 


and to the other a ■— C = group, carrying a methyl ; and that con 

necting the — C = and the — CO — is a = CH — and not a 

— CH^ — group. Then, knowing the relation of the atoms in the 
molecule, it may be readily understood how methyl iodide reacts 
with (i) phenyl (3) methyl (5) pyrazolon to produce antipyrin. 

Phenylhydrazine contains, in its molecule, two adjacent nitro- 
gens and there is nothing to lead us to suspect that they have 
changed v/hen entering into the structure of antipyrin. When 
antipyrin is distilled with a reducing agent, like zinc dust, and the 
products separated, anilin (C^Hj-NHj) having a phenyl, and methyl- 
amine (CHj.NHj) having a methyl group attached to nitrogens, 
are found. This shows that one of the nitrogens in antipyrin 
(Cj,Hj,N,0) has a phenyl and the other a methyl group attached. 
If the position of the phenyl in the antipyrin be arbitrarily called 
(i), then as the nitrogens have been shown to be adjacent, the 
position of the methyl is (2). This grouping is 



(•) (>) 

Cn,-N N-CH, 

The structure of symmetrical methyl-phenylhydrazine. 

CH,-N— - N-CH, 

s I I . . 

H H 

which, with aceto-acetic ester, condenses directly to antipyrin (see 
Knorr's second method), gives further evidence that grouping A 
exists in antipyrin. 

The presence of the CH, — C = and the CO groups in the anti- 
pyrin molecule is shown by the method of formation from aceta- 
acetic ethyl ester and phenylhydrazine. On the assumption that 
aceto-acetic ethyP ester has a ketonic structure, it contains a 
— CO — ds a ketonic group and a — CO — as a part of a car- 
boxy 1 ester — (CO • OCjH^) — group. As ketonic groups are 
known to condense with hydrazines and eliminate water, it is fair 
to assume that such a change occurs in this instance, for water and 
alcohol are by-products of the reaction. 

Graphically this condensation may be illustrated as follows : 

CH, C • CH^ • COOC.H, N-NH • C.H, 

O H, 

Aceto-acetic ethyl ester, -f- Phenylhydrazine. zizz 

N-NH C.H, 


Phenylhydrazone of ace oacetic ethyl ester. -f" W*tcr. 

In antipyrin, we have seen that a methyl group is attached to 
the second nitrogen. By this condensation it is manifest that the 

CHj-C = from aceto-acetic ethyl ester is also attached to the nitro- 
gen numbered (2) of grouping A, as in the following: 

(2) (I) 




That such a structure as given in this formula cannot exist, unless 
nitrogen can be tetravalent, is evident, and therefore this structure 


is exceedingly improbable. That the condensation occurs is cer- 
tain. How then can we explain this puzzling situation. We 
might assume that it has become pentavalent, but as nothing more 
than one methyl can be shown as attached to nitrogen (2) of anti- 
pyrin, this assumption is equally untenable. There remains but 
one further assumption. It must be trivalent. 

Since the product of condensation, the phenylhydrazone of aceto- 
acetic ester, has a double bond but no methyl attached to nitrogen 
(2) and antipyrin a single bond with a methyl, it is evident that 
this CH group must have entered the structure after the conden- 

■— C = to nil 

sation. Attaching then, the CH,— C = to nitrogen (2) gives anti- 
pyrin grouping : 


(2) (I) 

CH3-C = 


This formation of a trivalent nitrogen at position (2) and a 

CH3 — C = group at position (3) may also be shown by the 

assumption of an enolic instead of a ketonic structure for aceto- 


acetic ethyl ester; that is, CH, — C = CH — COOC^H^. Graph- 
ically, the equation of the condensation would appear 

HN - NHCgH, CH3 - C = CH - COOC^H, 

I I 

Hj OH' 

Phenylhydrazine -f- A ce to- acetic ethyl ester ^- 

HN - N - C,H, H,0 

CH,- C = CH - COOCjH, 

Phenylhydrazone-aceto-acetic ester -\- Water. 

Should the condensation occur in this way, one might expect a 


chlor-compound of the structure CH, — C = CH — COOC^H^ to 

act likewise with phenylhydrazine. Lederer found instead, how- 
ever, that the chlorine of such a compound condensed with the 


hydrogen attached to the phenyl nitrogen of phenylhydrazine. 
This reaction tends to show that in the case of antipyrin the ke- 
tonic and not the enoh'c aceto-acetic ethyl ester reacts with pheny- 
hydrazine, for, by analogy with the chlor-compound, the enolic 
form ought to condense and produce an isomeric reversed hydra- 
zone leading to a reversed antipyrin. Lederer made such a com- 
pound and called it iso- antipyrin. Chemically, it was (i) phenyl 
(2: 5) dimethyl (3) pyrazolon, while antipyrin is (i) phenyl (2:3) 
dimethyl (5) pyrazolon. This assumption of the enolic condensa- 
tion has one point in its favor, for by it a = CH — group follows 

the CH, — C = of phenyl hydrazone acelo-acetic ethyl ester, and 

by tracing this group through subsequent reactions to the antipy- 
rin molecule there is offered a simple explanation of why such a 
group is found present in antipyrin. 

It has already been stated that alcohol (HO-C^HJ, as well 
as water, is a product of the ketonic hydrazone condensation. 
The — O • C,Hj — of the alcohol could only have come from the 

— CO • OCjHj — of the ester, because it is the only — O • C^H^ — 
of the reacting bodies. It remains to discover the source of the 
hydrogen necessary to form H • OC,Hj. It could not^have come 
from the benzol ring of the phenylhydrazine (H,N ■ N • H • C^HJ, 
for a phenyl (C^H^) and not a phenylene(CgHJ exists in antipyrin. 
It is highly improbable that it came from the — CH, — group 
of aceto-acetic ethyl ester, for such a condensation of an ester to 
form alcohol is unknown. The only alternative is the hydrogen 
attached to the phenyl nitrogen of the phenylhydrazine. The 
reaction, forming alcohol, giving a ring structure and placing a 

— CO — group next to nitrogen (i), is : 

N -N-C.H, 

H = 

CH3 - C - CH, - CO • OC,H, 

Phenylhydrazone of aceto-acetic ethyl ester. 

N N-C,H, 

+ H.OC,H, 


(I) Phenyl (3) methyl (5) pyrazolon -|- Alcohol. 


Introducing the — CO — into the antipyrin grouping (see B), 
there is obtained 


(*) (O 
CH, - N N ■ CH. 

I I 

CH, - C = - CO 


We have yet to show how the C H, — C = and the — CO — of 
this grouping are connected. From the structure of aceto -acetic 
ethyl ester 



(CH3 - C - CH, - CO - OC,H,) 

we might infer that following the CH3 C = in antipyrin would 
come a — CH, — group in position (4) as is the case in (i ) phenyl 
(3) methyl (5) pyrazolon, A — CH, — group, however, would not 
satisfy the bonds of carbon (3) in grouping C. for it has two com- 
bining bonds and a—CH,— group can furnish but one, needing 
the other to attach to — CO — and form a ring structure. The fol- 
lowing three reactions of antipyrin are offered as proof that the re- 
maining group is = CH — . 

Reaction I. Antipyrin, with nitrous or nitric acid, forms only 
mono-nitroso- or mono-nitro substitution derivatives. The methyls 
certainly are not nitrated. That the phenyl is not nitrated is shown 
by distilling the nitro- derivative with zinc dust when anilin 
(C.H^NHj), just as in the case of antipyrin, is formed. Were a 
nitro group upon the phenyl in the nitro-derivative, phenylene 
diamine (CgH^(NH,)a) would be the product of the reaction. As 
all the elements as yet shown in the ring of antipyrin (see C) have 
their bonds satisfied by phenyl^ methyl or oxygen, the only pos- 
sible replacable hydrogen is at position (4). As dinitroso- or dinitro- 
derivatives appear not to be formed by antipyrin, the conclusion is 
justified that at position (4) there is a single hydrogen, that is 
a = CH — group. 

Reaction II. Antipyrin treated, in chloroform solution with bro- 
mine forms a monobrom-substitution derivative. Even in this 
case the bromine does not attack the phenyl for the brom-antipyrin 
resulting has different properties from a brom-antipyrin made with 


the bromine on the phenyl, using para-brom-phenylhydrazine. 
(BrC.H^NHNH) and following the regular method. But a 
single hydrogen, appearing in the by-product, hydrobromic acid 
(HBr), is eliminated by the action of bromine on antipyrin, so that 
the group attacked must have been a = CH — . 

Reaction III. Antipyrin. when treated with benzaldehyde and a 
condensing agent, hydrochloric acid, forms benzylidene di-antipy- 
rin. The by-product of the reaction is water and an elementary 
analysis of the condensed molecule gives figures which show that 
two molecules of antipyrin must have reacted. This furnishes the 
best proof of a = CH — group. The oxygen of the water is given 
by the aldehyde, the two hydrogens are furnished by two mole- 
cules antipyrin, therefore, there could have been but one reacting 
hydrogen in antipyrin, that is, a = CH — group. 

As a =CH — group is thus shown to exist in antipyrin, if this 
be placed in grouping C, it takes position (4) and the — CO — con- 
nected to nitrogen (i) takes position ($). The formula of antipyrin 

(2) (I) 

N N 

I I 

C = CH « C 

(3) (4) (5) 

The action of methyl iodide upon (i) phenyl (3) methyl (5) 
pyrazolon, may be interpreted in any of three ways. It has been 
shown that (i) phenyl (3) methyl (5) pyrazolon has no methyl at- 
tached to nitrogen (2) and has a — CH, — in position (4), while 
antipyrin has a methyl attached to nitrogen (2) and a = CH — in 
position (4). According to one interpretation, methyl iodide adds 
directly to nitrogen numbered (2) causing this to become penta- 
valent and giving the structure 

CH, - N N - C,H, 

CH, - C = CH - CO 


(*) (1) 

CH,I-N N-C,H, 

CH,- C - CH, - CO 


C - CHj - O 

(3) (4) (5) 

Hydriodic acid is eliminated, the iodine can obtain its hydrogen 
from no other source than the — CH, — . This compels the 
methylene (— CH,— ) to become methenyl (= CH — ) and requires 

VOL. XXII. — 19. 


the rearrangement of the double bond from between positions (2) 
(3) to (3) (4), thus producing antipyrin. According to the second 
interpretation methyl iodide adds to (i) phenyl (3) methyl (5) 
pyrazolon by the double bond between position (2) and (3) open- 
ing, the methyl adding to position (2), the iodine to position (3). 
This is shown in the following formula : 

(2) 0) 

rCH3i-.N . N-C,H, 

( — 

jl -C-CH^-CO 


CH3 (3) (4) (5) 

Loss of hydriodic acid (HI), the iodine coming from position 
(3), the hydrogen from position (4) would leave a double bond be- 
tween positions (3) and (4). forming here a methenyl (== CH — ) 
group and producing antipyrin. 

According to the third interpretation the formula for (i) phenyl 
(3) methyl pyrazolon is assumed to be : 

(2) (I) 


I • I • « 

CH3 - C = CH - CO 

(3) (4) (5) 

Such a formula results from the supposition of aceto-acetic ethyl 
ester having an enolic structure. Methyl iodide adds to nitrogen 
numbered (2) causing this to become pentavalent. The meth- 
iodide compound resulting has the formula : 

I (2) (I) 


CH>N ~ N-C,H, 


(3) (4) (5) 

Loss of hydriodic acid (HI) from nitrogen (2) would leave it tri- 
valent again, and produce antipyrin. 

Within the past year a new formula has been proposed by 
Michaelis and Schwabe for antipyrin. They give it the structure 

(2) (I) 
CH3 - N -— N -C,H, 

II "^ 

CH3 - C -CH = C 

(3) (4) (5) 

where nitrogen (2) is pentavalent and the oxygen is connected 


between this nitrogen and carbon (5). While working on chlor- 
pyrazol derivatives, these investigators produced a body to which 
they assigned the following structure : 

(2) (I) 

N— N -CH, Br 

CH3 - C - CH = C - CI. 

(3) (4) (5) 

Treating this with methyl iodide there resulted a methiodide com- 
pound with the formula 


1(2) (I) 

CH3-N ~N-CH,Br . 


CH3-C-CH = C-C1. 

(3) (4) (5) 

This body, by the action of alcoholic potassium hydroxide 
formed the same parabrom-antipyrin as that obtained from the 
action of methyl iodide upon (i) parabrom-phenyl (3) methyl (5) 
pyrazolon according to the regular methylation reaction. 

In the case of the pyrazol body, unless one assumes a total re- 
arrangement of the bonds in the molecule, the hydroxyls from the 
potassium hydroxide substituted for iodine at (2) and chlorine at 
(5) must condense with elimination of water and formation of an 
oxygen between (2) and (5), thus giving a new structure for the 
nucleus of antipyrin. 

The Derivatives of Antipyrin. 

Over one hundred derivatives of antipyrin have been reported 
in the literature of the past eighteen years, but few of them have 
been more than scientific curiosities. Many have been tried medic- 
inally in the hope of findmg a pyrazolon with all the advantages 
and with but few of the disadvantages of antipyrin itself. Fortune 
did not favor the^e investigations and to-day antipyrin, taken in all 
its properties, stands pre-eminently the antipyretic of the pyrazolon 
series. Among the compounds approaching it are tolypyrin, sali- 
pyrin, tussol and ferripyrin. Tolypyrin is slower and more lasting 
in its action as an antipyretic ; salipyrin combines the antiseptic 
properties of salicylic acid with the antipyretic action of antipyrin; 
tussol, the mandelic acid salt of antipyrin, has proven very useful 
in whooping cough; while in ferripyrin, the styptic action of anti- 
pyrin is strengthened by ferric chloride. 


A general view of the antipyrin derivatives will first be given in 
a systematic organic classification ; following this will appear 
detailed tables of all the antipyrin compounds reported up to the 
present and arranged in accordance with the systematic classifi- 
cation. Lastly, will come a chronological bibliography. 

Classification of Antjfyrin Compounds. 

1. Isomer. 


II. Homologues. 

Toly pyrin. 

III. Aldehyde condensation products. 

{d) Aliphatic. 



Hydrochloride salts. 
{b) Aromatic. 



Hydrochloride salts. 

IV. Brom- and lod-derivatives. 

(a) Brom. 




{b) lod. 


Pseudo-antipyrin iod-methylate. 
" ** lod-ethylate. 

V. Nitroso- and Nitro-derivatives. 

{a) Nitroso. 


*' " hydrochloride. 

{b) Nitro. 


« II 


VI. Amido-derivatives. 
[a) Unsubstituted amido. 


{b) Alkyl-amido. 

{c) Benzylidene-amido. 



Ortho-oxy- benzylidene-amido-antipyrin. 

[d) Acyl-amido. 

Formyl-amido antipyrin. 



Di benzoyl-amido-antipyrin. 
{e) Amido-ester. 

Beta-antipy ryl-imido butyric-ethyl ester. 

VII. Urea-, Imide-, and Urethane-derivatives. 

{a) Ureas. 


Symmetrical diantipyryl-thio-urea. 

{b) Imide. 

Antipyrin-tartronyl- imide. 
(^) Urethane. 

VIII. Di-azo-, Azo-, and Hydrazone derivatives. 
(a) Diazo. 

Diazo-antipyrin chloride. 

Diazo-antipyrin amido-benzol. 
{b) Azo. 

Antipyrin (4) azo-beta-naphthol. 
[c) Hydrazone. 

Antipyryl-hydrazone of (i) phenyl (3) methyl 
(4) keto (5) pyrazolon. 
IX. Hydroxl-derivatives. 

{a) Nitrogen ring substitutions, 
(i) Hydroxy. 


4 Oxy-antipyrin. 

4 Oxy-antipyrin dibromide. 

(2) Alkyl-oxy. 

4 Meth-oxy-antipyrin. 
4 Eth-oxy-antipyrin. 

(3) Acid-oxy. 

4 Benzoyl-oxy-antipyrin. 
{b) Benzol ring substitutions, 
(i) Alkyl-oxy. 



Para-eth-oxy-antipyrin salicylate. 
[c) Phenolic compounds, 
(i) Mono-phenolic. 


Ortho (also meta and para)cresopyrine. 



(2) Di-phenolic. 




(3) Tri phenolic. 

{d) Pheno-alcoholic compound. 
(^)][Chlor-aldehydic compounds. 
Mono-chloral antipyrin. 
(/) Hydroxy-acid compounds, 
(i) Mono-oxy. 
Metallic salipyrins. 
Antipyrin salicyl-acetate. 

Antipyrin meta-oxy-benzoate. 
Antipyrin para-oxy-benzoate. 


(2) Dioxy. 


(3) Tri-oxy. 

X. Acid chloride-, Acid-, Acid salt-derivatives. 
{cC) Acid chloride. 

Antipyrin chlor-benzoylid. 
^b) Acid. 

(i) Sulphonic. 

Antipyrin sulphonic acid. 
Diantipyrin*di sulphonic acid. 

(2) Acetic. 

Diantipyrin-acetic acid. 
Tetrabrom-diantipyrin-acetic acid. 

(3) Carbonic. 

Antipyrin-benzol-para-carbonic acid. 
[c) Acid salt. 

Antipyrin ferrocyanhydride. 

Antipyrin thiocyanate. 

Antipyrin valerianate. 

Antipyrin benzoate. 
XI. Metallic salt compounds. 
{a) With iron; 


Antipyrin ferri- thiocyanate. 
{b) With zinc. 

Antipyrin zinc chloride. 
{c) With mercury. 

Antipyrin mercuric bromide. 

Antipyrin mercuric cyanide. 

[d) With platinum. 

Antipyrin platinum chloride. 

(e) With cadmium. 

Antipyrin cadmium iodide. 
XII. Compounds with alkaloids, antiseptics, or mixtures with 
{a) Alkaloidal. 


Antipyrin quinine valerianate. 
[b) Antiseptic. 

Antipyrin salol. 
The following tables give a digest of the method of preparation 
properties and references of each compound mentioned in the 
above classification, and offer a general report of the chemical in. 
vestigations made upon the derivatives of antipyrin. 

List of Abbreviations Used in Tables and Bibliography. 

A. = Justus Liebig's Annalen der chemie. 
Am. = American Chemical Journal. 

B. = Berichte der Deutche Chem. Gesellschaft. 
Bl. = Bulletin de la Societe Chim. de Paris. 

C. = Chemisches Centralblatt. 
Chem. N. = Chemical News. 

Cob. = Coblentz Newer Remedies. 

C. r. = Comptes rendus des Seances de Tacad. des Sciences- 

J. = Jahresbericht iiber die Fortsch. der Chem. 
J. P. 5= Jour. Pharm et Chimie. 
J. Pr. = Jour, fur practische Chemie. 
Soc. = Jour. London Chemical Society. 
' Hoechst* = Meister, Lucius and Brunning. 


lillifyiil ill! 

It ij. 

? 'Nil 

11 < 1 

i S 
1 ^ 

^^l3 llf-^.-Elils'l^ll 








" ■ -1 



a F'» 


' 1. 

£ ii 


■ 'i'^. 4ii o" tt 



s- S si's -si 

11 = 

1 li' 


I ^1 
^ IS 

iSii |a 


illll P 


lysl U-i 

a i 

rli ill I 

si"- K' I 

ill yi-i! i.i 

llfel-. .I':;. it«H 


'5£S"aui-2i. ' uiH u-7~ s45 "■< _ 

T i 
a I 

III i"^ 

ill iiii 

i J 

if. ^ 

i|' i I 

3"j Is* 

iKIli fill 

i m 

- a ' 

■|||||||i'| ;|ii4 

fil:^ ti 
iiii; it 



I 'i 

•'ii'i 4 


1 lli 


yillJllfi isliffrJililii 


- I 




- ill 



■ s-eI 



■ lli 


Sg-p . 



l^^ ! 







u , 

J 1 


» Mil 



I- ■ -3 tfl 

iiisfi ^l| 


a ■ -I s «^ 

III tilhi^-l 
■ill ^Hpslil 





tin am-i 


I' o 

I M 

'I «J 

I' s 

N O 

ii M ^ O S 2 • •« 

^ 2*0 "^ * 3 O? ^ 

225^22. 2 gfi 
rt Sui g a-z Z * S 






— .2 & CpC 

N >« M M H O 

<0Q OQU 


° •* ft. 


3-i C WO 31? 

E^'S— 3 « > 
« :S ^ 3 .2 be *» 
i H c iJ-o g b« 
£ rt'C-3 C-O.S 




















= ^ o 
JU y »» 3 

3 *> 3 t. 




^ 0"C.2 3 




•3 3 CB-O «« 9 

2 o^- c C.2 

^ V O cS eS a' 



<<— ^ 





3 f a . 

•r a 

n * s e^ 

■ .<• Ml C »*»^^ 

=5^.2 .go 
• vr « 

^« - 

. . • "■ *8 

B B Q S 
o •"" " "" 


* '5 -a -c" *1 3 

u 3 

i o 

» C9 

O o 


K 3 


f > V iv* ••" "5 

o 2-0 
^ - o 


= o 
a M 


6 o-i 
g 2 - 

3 « B 



S v 


1 ^ 

' Q 



i S 

j» Q 

' l-H 
















• u 

















C CT3 












c a 



•5 . 

*S 3 

• *^ 2* 

J5 « E 

I 4, .. y 

pro 2 ee 

<3.S I 

3 U-s 

•s o a 5:2 
3= < J. u 

•^ 4 u act 

<^ M M 
• • • 

•3-0 o 

U 3 V 






•3 o 

2 o a 


















• • 

















u >» 

>, o. 


.J. ** 

rt 3 ^4 
























4: ■'•3 



' I mill 





■ S 















■ill- 3il 

;.it|r|| i|:i 
ililfll ill J 

1 i^-^ Bl^ 

ill ,1M 


'lil ii'iiii mma 

i f 

i« - 






- y -11 

S O J-5S 

I S. il-fl 



% H 



ii ji 



Si 1i 


Is 5e 






'1 ii:l 



■rf| - ■ 1- 




o I ■§ 

X 1°. 
2- >' 

S.1I Hi 



Ai I. 

li s«;i*; s"!.:; 

" 3|!l 3|5 


3«-:,J ari'J 

' U^ '' 


■ " 1 » ii^>. . °-: 



li h 



._. §1 _°= 


gi-3 E ^5 


■=-= 1 E. '^■^ 


il ill H 

.! !rti 

> sis £: a 
3 -^"5 " j^ 


'IIK If 


' strpi! 






^ i 


1 -^^ ^^ 

I i'i I'i I'ill i 

II IfHis ilSI 1 


C = G V 




I s 'Ml - ' s !| -i.--l - !* ■ 










2 ^g § o 

1 1 - 'a 

2 a ©3:2 
TJ j3 •« fl 13 

o «« 

^ o 






■a . t: 
























'Or! ^"^ 
♦ «-.** 



D, S 

<i I 

I o 

^ CO 










3-D — 
CO = o 

« rt gx.9 





o o 

^ o o 




— u 



— z 




— *» o 


M ~ 

O Wi ^ 




o A S i. 

*> C * 2 

jc g o a o 

^ 3 u 

^ o .- «-• n 







fl boa 
o a o 

-c 9* 

U S 3 
rt 3 

'^ C )S ^ 

S rt s 
•5 ^3rt 




•a y^^-S 
^ a .s u 

O O 
'J -tJ 














*^ r^ 















• • 











3 . 


e e 


h • w 


_ C-.'-- "^ rt . 


a u 

iJ c *• 


3 ** 

>.g rt 

•3 3 
2«- fl 
^ O n 










•a I 



"3 fl 
a c 


2 &«*« 




us:; — 















* ?3 












• « 

S rt 
9 w 






ilsV I H 


ill JlJ 

iiil flij 

E= - 



1= if 

B lis 111 













O <)< 

S o 


O ™ o 
O— © S 

= 11.1 

5 M 

.,^*> ^ 




















.N •> 


»o « 



M ei M 

i ••• 

, ** 






•— • 






































•— 1 





o * 





o S 

c •** •-* 

« c " 

f •» S 
c a o 

o o*c 

S|g . 

S C>.'C OS 

• N 

•- ? 


- o 


o " 






a a 

I 5^ 







6 ' J!i 



•n . 





a. • 










S"F c 








.9 4,- 




a (d 






•- >■ 









Chronological Bibliography. 

1884. Druggist's Circular. Notice on antipyrin, 185. 

Guttman. Tae Medicinal effect of antipyrin. C. 55, 493; 

Berliner Klein. Woch.^ 21 ^ 303. 
Knorr, L. Benzylidene di-antipyrin. B. 2040 b. A. 2j(?, 214. 

. The constitution of the chinazin derivatives. B. 77, 

548, 2032 b. 
Nitro- and nitroso-antipyrin. B. 77, 2038 b. 

The physiological effect of antipyrin. B. 77, 2037 b. 

Pharmaceutical Jour. London. Antipyrin-use in France, 
administration of. 75, III. Ser., 341, 435, 501, 737. 

1885. Dittmar. Chloriod-compounds of antipyrin. B. 7<?, 1617a. 
Druggists* Circular, Notice on antipyrin. 23. 
Liweh. Crystalline forms and angles of antipyrin. J. 1082. 
Pharmaceutical Jour. London. Antipyrin as hemostatic. 

445, use 288, manufacture 1059. 

1886. Baumgarten's Jahresbericht. Antipyrin-use in typhoid, 2, 

Casett. Use of antipyrin as hemostatic. J. P. 122, 271. 
Pharmaceutical Jour. London. Antipyrin as anodyne, 77, 

Umbach, C. Antipyrin report. B. 778 R. Arch, furexperim. 

Pathol. 21, 161-168. 

1887. Brouardel and Leroye. Physiological effect of antipyrin. 

B. 335 R; C. r. soc. biolog. 1885. 104, 106. 
Cesari and Carrara. Physiological effect of antipyrin. B. 

335 R. Ann. di chim. et di Pharm. 4 Ser. 3, 258. 
Dupuis. Administration of antipyrin in sea sickness. J. P. 

725. 521. 
Knorr, L. Antipyrin. A. 2j8^ 203. 

Bis- and methyl-antipyrin. A. 2j8t 204-210 

. Brom-compounds of antipyrin. B. 263 R ; A, 2j8, 215. 

•. Nitroso- and nitro-antipyrin. A. 2j8, 212-214. 

. Synthetical investigation upon aceto-acetic ester. B. 

259 R, A. 2j8, 137 to 219. (Univ. Wurzburg.) 

wilh Meistery Lucuis and B running, Antipyrin from 

phen}lhydrazine and aceto-acetic ester. B. 262 R, 609 P. 
Pharmaceutical Journal. London. Antipyrin and anti- 
febrine. 170-264. Antipyrin with phenol, 621; with 
nitrous acid, 1085 j as substitute for morphine, 1066 ; 
characteristics of, 1066; in headache and sickness, 170, 
485, 1005; in France, 664, 970. 


See. Subcutaneous injections of antipyrin in place of morphine. 

B. 578 R. ; C. r. 105, 103-105 ; J. P. 125^ 227. 

1888. Capitain and Gley. Poisonous nature of antipyrin. B. 365 C. 
Depuy and Bonne. Antipyrin as agent in sea sickness. C. 

53> I. 229- 
Gay and Fortune. Character of antipyrin. J. P. 126^ 594- 

Henocque. Antipyrin as a hemostatic. J. P. 126^ 218. 

Iwanoff. Physiological effect of antipyrin. B. 750 c. 

Lindo. Reaction of antipyrin. B. 858 r.; Chem. N. 5<?, 51. 

Neudorfer. About the new antiseptics — antipyrin and creolin. 

C. 5p, I. 290. 

Pharmaceutical Jour., London. Antipyrin and tests. 19, 

471, 607, 861, 920, 806, 240, 988. 
Vulpius. Antipyrin and phenol. C. 5p 1. 

1889. Baumgarten's Jahresbericht. Medical uses of antipyrin. 

Druggists' Circular. Upon the purity of antipyrin. 226. 
Jaffe and Hilbert. Effect of antipyrin on animals. B. 597 c. 

1889. Mallard and Stark. Incompatibles of antipyrin. J. P. 1890, 

131* '63, 210; C. 1889, do, 11. 200. 

Manseau. Effect of quinine on solutions of antipyrin. C. (5o, 
•I. 561. 

Pavia. New properties of antipyrin. C. do, I. 

Pellizari. Formation of antipyrin tartrc nyl urea and antipyrin 
tartronyl imide.. B. 734 c; C. 1889, do, II. 330; B. 22^ 
236 R. C. 60^ 1. 16; Beil. 1899, IV. 548. 

Petit. Analgesin and antipyrin identical. C. 60 ^ 1. 734. 

Pharmaceutical Journ. London. Antipyrin with caffeine, 
711, with carbolic acid, 977, with chloral, 602, 977, with 
iodine, 162, with sodium salicylate, 1058, antipyrin in 
asthma, 791, improper sale of, 417, poisoning by, 1059. 

Sawadowski. Effect of antipyrin upon animals. B. 410, c. 

Tuczek. Poisonous action of antipyrin. C. 60, II. 150. 

Vigier. Incomp. of antipyrin and sod. salicylate. C. 60^ I. 

1890. Griiner. Note on antipyrin. C. d/, II. 216. 

Manseau. Antipyrin and albuminous urine. C. d/, II. 174. 

Mercier. Antipyrin in urine. C. d/, I. 734. 

Millard and Stark. Antipyrin and its relation to other medi- 
cines. C. d/, II. 288, Phar. Jour, /p, IV. 20. 

Reuter. Effect of antipyrin upon chl >ral hydrate. C. d/, I. 
649, J. P. 7JO, 354, 539. 


Sochagzewski. Valerianates of antipyrin and quinine. J. P. 

131, C. 1 89 1, 62^ II. 485. 
Spica. Incomp. of antipyrin and sod. salicylate. C. d/, II. 


1 89 1. Andreocci. Action of phosphorus pentasulphide on antipyrin. 

B. 648 c; C. 62^ II. 1954. 

Patein. Naphthol antipyrins. J. P. 132, 585. 
Riedel. Proceedings in the preparation of antipyrin. J. P. 132. 
608, Monat. scientif. April, 1891. 

1892. Altschul. Para-alkyl-oxy-antipyrin. B. 23^ 1852. 
Baumgarten's Jahresbericht. Antipyrin in diphtheria and 

tetanus. <?, 190, 181. 
Briihl. About a supposed alcohol of antipyrin. B. 395 a; 

C. 63, I., 481 ; C. 62y II. 318. 

Biirwell. Isova'erianic acid antipyrin. C. 63 I. 477. 
Cressati. Antipyrin benzoate and picrate. C. (5j, 1. 326. 
Jandricr. Nitro-derivs. of antipyrin. B. 324 c; C. r. //^, 

303; C. 63, I. 478; Soc. 730; J. P. 134, 419. 
Knorr and Tauf kirch. Upon the alcohol of antipyrin being 

an anilid. C. dj, I. 745. 
Lederer. Isoantipyrin. J. Pr. ^5, 91 ; C. <5j, I., 205 ; J. P. 

134. 35:- 

Meister, Lucius and Brunning. Antipyrin from Beta- 
brom-crotonic acid and methyl phenylhydrazine, followed 
by sodium hydroxide. B. 883 c; Deut. Pat. 64,444, 

Mollenhoff. Sulphooated antipyrins. B. 23 I. 1950 ; Beil. 
1899 IV. 737. 

Nef. Formation of antipyrin from (i) phenyl (3) methyl (5) 
pyrazol. B. 22 c. 

. Consideration of the properties of aceto- acetic ester and 

its condensation. C. dj, I. 20. 

Stolz. Para-ethoxy-antipyrin. B. 25, 1664, 1852 ; Soc. 1080. 

Tappeiner. Diuretic effect of antipyrin. C. (5j, II. 657. 

Zimanyi. Optical properties of antipyrin. C. 63 ^ I. 375. 

1893. Freudenberg. The physiological effect of antipyrin. C. 64, 

I. 489. 
Riedal. Tolysal. C. 64, I. 709. 
Stock. About antipyrin and tolypyrin. C. 64, I. 988. 
Thorns. Increase in our knowledge of the relation between 

chemical constitution and therapeutic effect. C. 64^ I. 70H. 

1894. Meister, Lucius and Brunning. Production of antipyrin 


from (i) phenyl (3) methyl (5) ethoxy-pyrazolon by methyl 

iodide. B. 282 R. 
Petit and Fcvre. Antipyrin ^-resorcylate. J. P. /jp, 106. 
Schaak. Determination of antipyrin. C. 65, II. 1022. 

1895. Fritz. Migranin. C. d(5, I. 970. 

. Antipyrin 's medicinal action. C. dd, II. 1086. 

Haase. Antipyrin with iron. C. 66y I. 497. 

Knorr. Antipyrin preparation from (i) phenyl (3) methyl (5) 

ethoxy-pyrazol. BI. 707. 
. Consideration of the phenolic formula for ( i ) phenyl 

(3) methyl (5) pyrazolon. B. 28, 706-712 ; for antipyrin. 

B 28, I. 706; A, 2pj, 27. 
Meister, Lucius and Brunning. Preparation of antipyrin 

from (i) phenyl (3) methyl pyrazol (5) oxyacetic ester. 

B. 1080 R. 

Patein and Dufau. Antipyrin compounds with pyrocatechin, 
guaiacol, veratrol, resorcin and hydrochinon. B. 914 R. 
Soc, 1896 A.I. 188; Bl. (3) IS, 611. 

. Compounds ofantipyrin with the di phenols. B. 914R.; 

C. r. 727, 532; Bl. (3) 75, 172, 611 ; C. d/, I. 1282; C. 
66 y II. 963 ; J. P., 141 y 402. 

Rothenburg. An antipyrin synthesis. C. 66, II. 216. 
Schuyten. Analysis of antipyrin by iodine. Volumetric. B. 

1021 K; Chem. Zt. 7p, 1786; C. 66, II. 922. 
', Antipyrin and zinc chloride. B. 986 R.; Chem. Zt. /p, 

Schuftan. Effects of formic, acetic and salicylic aldehydes upon 

antipyrin. B. I. 1181 ; C. 66, II. 34. 
Spiegel. Antipyrin and zinc chloride. B. 986 R. 
Stolz. For a knowledge of the antipyrin synthesis. B. 28, 

632, C. 66, I. 950. 

. Pseudoantipyrin. B. 28, 629. 

Van Itallie. Antipyrin and mercury salts. Soc. A. I., 260 ; 

C. 66, II. 1052; Ned Tydschr. Pharm., 1895, 7, 295. 
Winkler. Crystalline forms of antipyrin. C. 66, II. 913. 

1896. Beuttner. Reactions of antipyrin. C. 67, II. 133. 
Carrez. Antipyrin reactions and action on quinine. J. P. 

142, 253; Soc, II. 584. 
Gawalowski. Solubility of antipyrin. C. 67, I. 1015. 
Himmelbauer. Antipyrin from (i) phenyl (3) methyl (5) 

pyrazolon (2) carbonic methyl ester. J. Pr. (2), 5^, 187. 
. Homoantipyrin. J. Pr. (2), 5^, 191. 

. Methyl antipyrin. J. Pr. (2), 5^, 210. 

Knorr. Action of methyl iodide upon antipyrin. A. ^pj, 
7-19; Soc., 1897, A. I., 108; B. 959 R. 

. Antipyrin and ferric chloride. B. 813 R. 

. Homoantipyrin-pseudo-iod-roethylate. A. 2QJ, 23. 

Knorr. The constitution of antipyrin. A. ^pj, 34. 

and Geuther. Upon the reduction of nitroso-anti- 

pyrin. A. 2pj, 55 ; Soc., 1897 ; A. I. 112. 

Geisse and Labe. Antipyrin iod-methylate. C. 6j^ 

n. 1025; C. 68, J. 1206. 
— and Rabe. Antipyrin benzoyl chloride additions. B. 
961 R.; C. ^7, II. 1027; C. 1897, 68, I. 1207; Soc., 
1897 A. I. no; A. 2QJ, 42. 

and Stolz. Upon (4) amido-antipyrin and its con- 

densations. A. 2gjy 58; Soc. 1897, A. I. 112. 
— . Antipyryl hydrazone. A. ^pj, 69. 

Antipyryl urea and thio-urea. A. 2gj, 65 ; 6. 962 R.; 

Soc. 1897, A. I. 112. 

— (4) Antipyryl-urethane. B. 962 R.; A. 2pj, 66. 

Diazo-antipyrin chloride and couples. A. 2gj, 67 ; B. 

961 R.; Soc. 1897 ; A. I. 112; Bell. 1899, IV. 1489. 

Kraft and Weiland. Action of antipyrin in cathode light and 

alteration of boiling point. B. 2p, 2241. 
Marcourt. Compound of antipyrin and formaldehyde. Soc, 

1897 ; A. I. 298 ; C. 67 ; I. 1208. 

Patein and Dufau. Action of antipyrin on bodies possessing 
3 phenolic hydroxyls. Bl. (3) //, 611, 1048; C 6j, II. 779* 

340; B. 1151, R. . 
Action of antipyrin upoft gallic acid. Bl. (3) 15, 1050; 

C. 67, II. 240. 


Action of antipyrin upon hydroquinone. Bl. (3) 75, 

611 ; B. 1 141 R. 

Artion of antipyrin upon sodium salicylate, and anis'c 

acid. Bl. (3) 7/, 847. 

Combinations of antipyrin with oxybenzoic acids and 

their derivatives. Bl. (3) 75, 846, 1049. Soc. A. I. 150. 
Combinations of antipyrin with cresols. Bl. (3) 7/, 

609 ; B. 1 1 40 R. 

Pschoor. (4) Oxyantipyrin. A. -2pj, 50. 

Schuyten. Antipyrin thiocyanate and antipyrin mercuric chlor- 
ide. Soc. A. I. 575. 
1897. Kippenberger. Antipyrin salicylate. C. 68, I. 484. 


. Estimation of antipyrin. C. 68, I. 484, Soc. 1897, A. 

II. 292. 
Meyer. Constitution of antipyrin. C. 68, I. 37. 
Patein. Antipyrin compounds with phenols. C. 68, I. 540, 

750. Soc. I. 297. 
Schuyten. Antipyrin reactions with nitrites. J. P. 144, 172. 
. Quicksilver halogen compounds with antipyrin. C. 68, 

II. 614; Soc. 1898, A. I. 452; C. 1900, May.; C. r. 130, 


1898. Bougault. Reaciionsof antipyrin with iodine. J. P. 146, 161, 

C. 6g, II. 858. 

Pieux. Antipyrin and lactation, j. P. 7^7, 419. 

Niccari. Osmotic pressure of antipyrin solutions. Soc. A. I. 

Oechsuer de Commick. Action of hypochlorites upon anti- 
pyrin. Soc. I. 566. 

Patein. Compounds of antipyrin with aldehydes. J. P. 146, 
79. Soc. I. 493. 

Zschimmer, Eppler and Schimpff. Antipyrin pseudo alky- 
late. C. 6g, I. 555. 

1899. Beilstein. Antipyrin compounds. 3d Ed. IV. 509-513; 737, 

1489, 1 109. 

1900. Michaelis and Schwabe. Preparation of Para-mono-brom- 

antipyrin. B. 33, 2607. 

and Sudendorf. Preparation of antipyrin-benzol-para 

carbonic acid. B. jj, 2622. 



By henry S. MUNROE. 

The classifier here described is used in the ore testing laboratory 
of the School of Mines to prepare small samples of crushed ore for 
jigging, vanning, or washing in a batea or on small laboratory 
tables. The water current, controlled by a dial-cock, enters at a and 
filling the flask e rises through the tubes d and c. The velocity is 
determined and adjusted by noting the time required to fill a vessel of 
known capacity from the spout at the top. The funnel b is allowed 
to fill with water, until the static head balances the velocity head in 
the tubes c and d. By loosening each of the corks for a moment the 
air may be expelled from the apparatus. Now closing the funnel 
stop-cock the sample of ore to be water sized is placed in the funnel 
and gradually fed into the apparatus. The heavy grains fall against 
the rising current into the flask e while the light grains ascend 
through c and are discharged at the top. By retreating the con- 
tents of the flask with water currents of increasing velocity any 
desired subdivision of the ore sample may be secured. Or a num- 
ber of these classifiers may be used together, the first with current of 
maximum velocity discharging its lighter product into the funnel of 
the next, and so on through the series. In this case the level of the 
top of each funnel should be adjusted so thatthe water just reaches 
the edge when equilibrium is reached at the velocity of current 
employed. The water from the previous classifier then overflows 
at the top of the funnel and does not pass through the next classi- 
fier, the funnel acting as a small spitzkasten to get rid of this extra 
water. It will be found necessary, however, to treat the sample, 
before running it through such a series, in a classifier with a gentle 
current, say 5 mm. per second, to remove at once the finest slimes, 
which would otherwise be lost. 

In the School of Mines laboratory the wate^ used is supplied 
from a tank with constant head, and under these conditions this 
classifier gives remarkably accurate results. At the end of a test 
there will remain in the apparatus a large number of grains just 
supported by the water current, which will continue to dance up and 
down in the tubes for many minutes. Some of these grains will 
settle in the curve at the top, from which they may be dislodged by 
tapping the tube with the finger. Care should be taken not to feed 



the ore too rapidly. The ore in the tubes at any given time should 
be less than one-tenth of the volume of the water current. If more 
ore be charged, heavy material will be carried up (by the in- 
creased lifting power of the interstitial currents) that should nor- 

Scale one-eighth. 

mally find its way into the flask below, and further the equilibrium 
between the feed tube and the upper sorting tube will be destroyed, 
giving rise to irregularities in the flow. 

These classifiers are made by Eimer & Amend from designs of 
the writer, and may be secured from them with a special ring stand 


support, at moderate cost. The Erlenmeyer flask and the stop-cock 
funnel are stock articles. The opening through the cock of the funnel 
should be as large as possible. The conical shape of the flask facili- 
tates the transfer of the sands when necessary. It will be found 
convenient to have the lip of the flask ground so that it can be 
closed by a ground glass plate. If then the side inlet be corked 
the flask may be inverted in the funnel or other receptacle, and 
the sand transferred without emptying out the water. The ring 
stands supplied have a special ring support by which the flask may 
be held over the funnel, in a vertical or in an inclined position, and 
the sands transferred a little at a time as required. If the glass 
plate be fastened to the top of the flask with a loop of string or a 
rubber band, it may be used as a sliding valve to control the flow 
of sand. In this case a common funnel without a stop-cock may 
be used. This arrangement will be found convenient when large 
quantities are to be treated and where the classifiers are used singly. 
For this purpose an extra flask will be required, which should have 
a mouth of the same size as the other, that the two may be inter- 

The following table of geometrical progressions will be found 
convenient, the ratios being the sixth, seventh, eighth, ninth and 
tenth root of ten respectively. 

Table of Velocities of Water Currents in Millimeters 

PER Second. 























74 99 











100.00 ' 






)1 62 




79.44 100.00 

By shifting the decimal points any series can be extended in 
either direction. By using alternate values in any series it will be 
possible later to subdivide any product if it is found desirable to 
do so. In practice decimals of a millimeter may be neglected. 



By henry S. MUNROE. 

The laboratory slime tables here described have been added to 
the equipment of the ore-testing laboratory of the School of Mines 
to supplement the work of the Vezin laboratory jigs by an appa- 
ratus to treat slimes. The table consists of a strong metal frame 
about five feet long in which flat plates of pine, maple, slate, glass 
or other material may be secured. These plates are held by 
springs, as shown, and the joint between the frame and the plate 
is made by inserting a strand of cotton wicking. The inclination 
of the table is adjusted by the levelling screws with the aid of 
a slope level reading by means of a vernier to five minutes. 
The amount of water is regulated by a dial-cock. These adjust- 
ments are controlled by testing with a small sample of the mate- 
rial to be treated, and giving to the table such an inclination, and 
such an amount of water as will effect the desired separation. 

Data for these adjustments will be found in the paper by Pro- 
fessor Richards, •* Sorting Before Sizing" {Trans, Amer, Inst 
Mining Engineers^ Vol. 27, page 76). It will be found convenient 
and more rapid to adopt somewhat more water and smaller inclina- 
tions of the table than recommended by the author of this paper, 
especially for the preliminary washing, reserving the steeper incli- 
nations and reduced water currents for the more careful retreat- 
ment of middlings. 

To operate the table the corrugated feed tray is filled with wet 
material and gradually pushed under the water jets, care being 
taken not to feed too fast. After two or three minutes the flow of 
slime is interrupted and the material left on the table is washed 
for two or three minutes until most of the tails have run off. 
Finally the middle products and heads are successively washed off 
with the aid of a wash bottle with a large jet. By making at first 
a large proportion of middlings, clean heads and poor tailings 
may be secured. The middlings may be saved and washed by 
themselves later, varying the amount of water and the inclination 
of the table to suit. 

In the first experiments with this laboratory table it was found 
that the water banked up at the edges by capillary attraction, in- 
creasing the velocity of the current at this point and increasing the 


LADOf\ATOf\Y Slime: Table: 

Scale one-seventh. 







.of eruptives has until recently met with little favor. The reason 

VOL. xxn.— ax. 


loss in the tailings. The same cause reduced the thickness of the 
water fi]m below the normal a short distance from the edge, tend- 
ing to cause the formation of fingers of sand, unless considerable 
water was used. Both of these difficulties were overcome by con- 
structing the washing surface in the form of a panel four inches 
wide and one-twentieth of an inch deep, or about the average 
depth of the water film. For rapid and accurate work it will be 
found best to make a large number of water-sized products, so as 
to obtain in each case material easily treated on the table. 





Introduction 307 

Distribution of the Serpentine 308 

The Staten Island Serpentine 309-316 

Geological Features 309 

General Characters 310 

Microscopic Characters ^ 311 

The Hoboken Serpentine 316 

Geological Relations ... 316 

General Characters 316 

Microscopic Characters 317 

The New Rochelle Serpentine ....... To be continued 

Geological Relations i< ♦< << 

General Characters <« <« << 

Microscopic Characters . 

The Serpentine from Tenth Ave., New York 

The Serpentine from Aqueduct Shaft, No. 26 

The Serpentine at Rye 

Mineralogy of the Serpentines *« «< i« 

Origin of the Serpentines «« «» »< 

Age of the Serpentines «« «« «< 

Introduction, — Since Mather first described the serpentines near 
New York and ventured to deal with their genetic relations by classi- 
fying them in his *' Trappean Division," they have been the sub- 
ject of frequent discussion by geologists working in this field. The 
view expressed by Mather that the serpentines were modified forms 
.of eruptives has until recently met with little favor. The reason 

VOL. xxn. — ai. 

(t << It 


for this divergence of opinion is to be found, no doubt, in the fact 
that the methods now used in the study of such questions had not 
been developed at that time, and too much importance was placed 
on general characters and appearance in the field. In 1890, G. F. 
Merrill published a description of the serpentine from Aqueduct 
Shaft 26, New York City,* which appears to be the first paper giv- 
ing any information as to the microscopic nature of the rocks and 
the first to definitely trace their origin. His researches, however, 
did not extend beyond this locality. 

A paper by F. J. H. Merrill f entitled '• The Geology of the 
Crystalline Rocks of Southeastern New York," is also of im- 
portance in this connection. Dr. Merrill briefly outlines the general 
progress that has been made in the study of serpentines during 
the past few decades, and describes the New Rochelle area, whose 
serpentine, he shows is derived from pyroxene and amphibole. 
The igneous theory of origin first advanced by Mather is once 
more brought to notice and used with reference to this locality. 
Dr. Merrill states that he found no direct evidence bearing on the 
genesis of the Hoboken, Staten Island and Rye serpentines. 

Distribution of the Serpentines. 

The serpentines are exposed in a series of roughly lenticular 
areas that reach in a broad curve through Hoboken, Manhattan 
Island and New Rochelle to a point near the village of Rye, which 
is some thirty miles northeast of Staten Island. All show more 
or less elevation relative to the surrounding formations, this feature 
being very marked in the two localities first mentioned. It is to 
be noted that these rocks are included in a narrow belt of related 
types, which extend with brief interruptions from Maine to northern 

Footnote. The field work and the greater part of the investigation upon which 
this paper is based were carried out during the fall of 1896, and the following winter, 
while the writer was Fellow in Geology at Columbia University. A summary of the 
principal results was given before a meeting of the Mew York Academy of Sciences in 
May, 1897. The writer desires to acknowledge his indebtedness to Dr. Arthur Hollick 
for information as to the Staten Island localities, to Mr. G. A. Goodell for chemical 
analyses of the rocks, and especially to Professor I . F. Kemp, who gave advice and 
assistance in the preparation of the paper and placed the facilities of tke Geological De- 
partment at his disposal. 

*•* Notes on the Serpentiies of Essex County, New York," etc.. Proceed. U. S. Nat 
Mus., Vol. XII., pp. 595-600. 

f New York State Museum Report, 1896. 


The Staten Island Serpentine. 

Geological Features. — The serpentine outcrops at numerous local- 
ities, but, so far as known, only within the bounds of the central 
elevated area that forms such a prominent topographic feature 
of the Island. This ridge starts at the northeastern end, and fol- 
lows the shore line for about two miles. It then extends in a south- 
westerly direction somewhat beyond Richmond. It is a lense- 
shaped mass having an extreme length of about seven miles and 
a width of two and a half. The total area is not far from thirteen 
square miles. 

The northern and eastern limits of the ridge are marked by 
steep and, in places, precipitous slopes, but to the south and west 
the contours widen out, and its extent in these directions can not 
be accurately defined. According to Bien's topographic map, the 
greatest elevation above sea level is a little over 400 feet, while a 
considerable area is above the 200 foot contour. The surrounding 
plain ranges from tidewater to 80 feet above. 

The surface of the ridge is broken by low hills and rather wide 
valleys that have been rounded out by weathering and denudation. 
In places a heavy covering of drift in the shape of clay and gravel 
with occasional boulders of diabase, sandstone and gneiss is pres- 
ent, but it often gives way to a thin coating of humus resting di- 
rectly upon the rock. It is noteworthy that plant growth is abun- 
dant and of good size, notwithstanding the usual unfertility of lands 
derived from magnesian rocks. 

Between the ridge and the eastern shore is a belt of low-lying 
country composed of Cretaceous clays, morainal material and 
beach sand. Toward the south these deposits widen out and ex- 
tend across the Island in a westerly direction. The Cretaceous 
strata sometimes exhibit evidences of violent disturbance, being 
crumpled and folded like rocks that have been subjected to moun- 
tain-making forces. This phenomenon has been studied by Dr. 
Hollick, to whose paper reference should be made for a full de- 
scription and explanation.* 

Triassic sandstone borders the serpentine on the west, but the 
line of contact is not revealed by outcrops or noticeable physio- 
graphic ftatures. The sandstone is broken through at some dis- 
tance from the serpentine by a parallel dike of diabase. 

• « Dislocations in Certain Portions of the Atlantic Coastal Plain Strata and their 
Probable Causes." Trans. N. Y. Acad. Sci., XIV., pp. 8-20. 1894. 


It is probable that the sandstone and clays rest upon schist, in a 
southern continuation of the rocks exposed on Manhattan Island 
which are regarded as metamorphosed strata of Hudson River age. 
Cozzens * has recorded an interesting occurrence of pegmatite 
granite at Quarantine near Tompkinsville, that is now inaccessible. 
The same locality is described by Britton.t as follows : ** Fine 
granite outcrops on the shore of the Upper New York Bay, about 
four hundred feet southwest of the Tompkinsville steam-boat land- 
ing, and directly in front of the old building known as Nautilus 
Hall. The surface of the rock exposed at low tide is about eighty 
feet wide by fifty feet long, at high water mark the rock disappears 
beneath a hill of drift some fifteen feet in thickness. A little riiore 
of the same rock is exposed at a point about two hundred feet 
south of the main outcrop." Pegmatite veins or dikes are of fre- 
quent occurrence in the mica schist formation of Manhattan Is- 
land and in places are of considerable extent. 

General Characters, — A macroscopic examination of typical 
specimens shows little else than serpentine. Usually a few crys- 
tals or bunches of chromite and scales of talc and chlorite are vis- 
ible, but these minerals are relatively unimportant. 

On the eastern slope of Pavilion Hill, the rock is given a special 
character by the inclusions of amphibole crystals, usually of light 
gray color and prismatic habit. They form a network of interlac- 
ing fibers which at times entirely replaces the serpentine. This 
rock is also developed further south near Grant City and it is prob- 
ably present along the entire eastern border. Its presence is 
marked by sharper contours in the topography, the rock being 
much more resistant to erosion and weathering than serpentine. 

Other variations from the normal type are represented by talc 
and chlorite schists, thin bands of which accompany the trem- 

The color of the serpentine varies with the amount of iron oxides 
present and the degree of weathering which it shows. Normally, 
it is dark-green or almost black upon fresh surfaces. Weathering is 
first manifested by oxidation of the iron constituents which stain 
the rock yellow or reddish. Following this, bleaching effects are 
noticeable with a decrease in hardness. The final product is a 
soft, white, powdery mass much resembling talc. Specimens from 

* ** A Geological History of New York Island," 1843. 

f N. L. Britton, Geology of Richmond Co. School of Minks Quarterly, II., 



Castleton Corners have a mottled appearance, owing to the bronze- 
colored inclusions in a groundmass of very dark serpentine. 

A more or less well defined lamination is apparent wherever the 
rock has been exposed to weathering influences for any length of 
time. Pronounced schistose types do not occur, however, except 
on the edges of the area. 

Jointing is everywhere present, but it shows little regularity even 
within the limits of a single outcrop. At Grant City, the most 
marked system has a nearly east and west strike with a steep dip 
of 70® to 80^. Usually the rock is divided into masses of irregu • 
lar size and shape by the numerous joints. The surfaces are 
slickensided and striated, showing that the rock has been subjected 
to pressure. This may have been due to some extent to the in- 
crease in bulk by alteration of the original minerals, though it can 
hardly be believed that the intense crushing effects observed at 
many of the localities coald have been produced by such an agency. 
It is noticeable that within the interior the serpentine is much less 
disturbed. At Castleton Corners, a faint parallel arrangement of 
the constituents is observable on weathered faces, but otherwise 
the appearance is similar to that for massive igneous rock. The 
strongest developed system of jointing here has a N.W.-S.E. di- 

Microscopic Cfiaracters, — An examination of thin sections shows 
that the processes of alteration have been carried nearly to com- 
pletion. In the majority of cases only secondary minerals are 
present, including iron ores, carbonates, chlorite and talc, besides 
varieties of serpentine which are the most important. Fortunately 
the characteristic textures resulting from the change of the pri- 
mary minerals to serpentine are often well preserved and give 
a clue as to the derivation of the latter, when studied and com- 
pared with the results recently achieved in this petrographic field. 

One of the most interesting types, microscopically, is that found 
near Castleton Corners, where in excavating for the foundations of 
a building, the rock has been exposed to a depth of several feet. 
The precise locality is known as " the old Germania Brewery site." 
Compared with specimens from other parts of the area, it has suf- 
fered very little from secondary alteration. Glistening bronze 
prisms embedded in a ground mass of dark serpentine lend a 
striking resemblance to some of the basic types of gabbros. The 
prisms have a good cleavage parallel to the longer axis. 


In thin section the serpentinous groundmass breaks up into an 
interwoven mass of fibers and plates sprinkled with dust and larger 
particles of iron ores. The fibers are slightly pleochroic, yellow 
to light green in transverse sections and give low polarization 
colors of the first order. The prismatic individuals are built 
up from serpentine plates in parallel arrangement, which give 
aggregate extinction. They are also slightly pleochroic and bi-re- 
fnngent. Their whole appearance is strikingly similar to bastite, 
and it was suspected at first that they were pseudomorphs after 
orthprhombic pyroxene. Examination of fresher material, how- 
ever, revealed included remnants of a colorless mineral whose 
optical properties correspond to monoclinic amphibole. The 
prisms are of stocky habit, measuring about 20 mm. in length by 
half that or more in width. Cleavage is pronounced in the longi- 
tudinal direction, while a cross fracture at right angles is often pres- 
ent. The boundaries in the prism zone are sharp, the ends, how- 
ever, irregular and feathery, losing themselves in the surrounding 
serpentine. By alteration the amphibole assumes a faint yellow 
color from inclusions of serpentine and gradually changes its optical 
properties. The process of serpentinization is simple and uniform 
in all cases. In the first stages, minute veinlets of serpentine are 
developed along the cleavage planes of the amphibole at the ends. 
From here they gradually extend toward the center, at the same 
time reaching out into the lateral prisms and widening out until 
the crystal is entirely altered. By this process serpentine plates 
are developed with uniform arrangment parallel to the c axis of the 
amphibole. In transverse sections the plates are divided into two 
.sets which cross each other at angles corresponding to the horn- 
blende cleavage. When examined in convergent light, cleavage 
pieces show the emergence of an optical bisectrix with large 
axial angle. The serpentinization is accompanied by separation of 
magnetite dust. 

A colorless monoclinic pyroxene was observed in one of the 
specimens from this locality. It occurs in granular aggregates 
representing partially serpentinized crystals whose original boun- 
daries are not distinguishable. The high bi-refringence, large angle 
of extinction (o°-37^) and prismatic cleavage renders its identifica- 
tion very easy. As with the amphibole, alteration begins by the 
formation of serpentine along the edges and cleavage planes* 
When completed, a mass of fibers and plates occupies the position 


of the former crystal. They show no uniform arrangement as a 
rule, though within certain limits they may be parallel or may cross 
each other at a definite angle. Very little magnetite is set free by 
the alteration. 

ChromiU in irregular grains and aggregates, rarely in octahedra 
is quite abundant in all the specimens from this locality. It is 
usually quite opaque and difficult to distinguish from magnetite. 

ChUmtt and taU are persistent but not important secondary prod- 
ucts. Both minerals occur in thin blades which are distributed 

Fig. I. ScTpenliDe from Slnteo lil>iid,ihowiiig paitiill; mlleKd eijttal of olivine 
with a.ulomorpbic bouQtUries. 

through the serpentine. Apparently they have been derived from 
both amphibole and diopside. 

At Martling's Pond, about one mile northeast from Castleton 
Corners, there is an outcrop of serpentine showing similar macro- 
scopic characters to that just described, but with different structure 
of the ground mass. This peculiarity is important in that it alTords 
evidence of the former presence of olivine in considerable quanti- 
ties. Altered fragments and, in a single case, a crystal of this min- 
eral were also found in some of the freshest material. The crystal 


(Fig. I) shows the prismatic development with pyramidal ter- 
minations characteristic of olivine, gives parallel extinction and 
rather high polarizing effects. By alteration, the- olivine yields 
serpentine fibers that are so arranged as to give a reticulate appear- 
ance to the field. Small veins or threads of dark serpentine that 
polarizes quite strongly and contains abundant magnetite, surround 
lighter areas of so closely matted texture that they have little effect 
on polarized light. Several of these areas may be grouped to- 
gether representing the space formerly occupied by an olivine 

Fig. z. Serpentine from Staten IiliLiid, completelj Altered. The Ugbtei areu 
were formeitjr occnpjed b; olivine crfttals ud the original bonnduies of the Uttci >re 
still ditcemlble. 

crystal. This peculiar structure is conditioned by the processes of 
alteration which begin on the outer edge of the olivine and along the 
lines of fracture; during the first stages, the serpentine substance has 
opportunity to assume a definite arrangement, but later on appar- 
ently the conditions for crystallization are not so favorable. When 
well defined, the structure affords indubitable proof of the former 
presence of olivine. By secondary alterAtion,that is, by the weather- 
ing of the serpentine, and by the removal of the iron ores, the char- 
acters are gradually erased, and a homogeneous mass results which 
gives little or no clue as to its genesis. 


Pavilion Hill, near Tompkinsville, afTords the most extensive 
outcrop of the rock and is interesting from the variety of its types. 
This locality is on the extreme northeastern edge and not far 
from the pegmatite granite described by Britton. The hill on 
its eastern face has a very steep slope for some distance, and is 
formed by bare rock. One variety is a hard, dense mass of tre- 
molite crystals, thinly laminated and apparently resting on edge. 
Under the microscope it shows little that is worthy of note, being 
composed almost entirely of thin needles of tremolite in a perfectly 
fresh state. In close proximity is a dark, massive, completely 
altered serpentine that is sprinkled with silvery flakes of talc and 
that contains chromite and magnetite. This rock seems to have 
been derived largely from amphibole and olivine (Fig. 2). The 
variety most commonly met is dark-green and nearly massive. It 
is composed of serpentine, talc, chlorite, iron ores and tremolite. 

Talc schist and chlorite schist are developed along with the 
tremolite rock and their association is probably a genetic one. 
The best exposure is at Grant City in a shallow ravine just west of 
King's Highway. The chlorite forms a thin band that is included 
in tremolite schist. It has become very soft and friable through 
weathering. In thin section it shows blades of low refracting 
chlorite, with almost no binding material. The blades have strong 
pleochroism, dark bluish-green to yellow, and small axial angle. 
The talc schist offers little of interest from a microscopic stand- 
point, though its exact relation to the serpentine and tremolite rock 
would be well worth deterfnining. The occurrence of these sec- 
ondary phases of alteration on the border of the area naturally 
suggests an igneous contact. 

Chemical Composition. — The general character of the serpentine 
is illustrated by the following analysis of a specimen from Castle- 
ton Corners, which was made by Mr. G. A. Goodell : 











H,0 below 110° C. 




H,0 above i io<' C. 







From the relatively large amount of ferrous iron present, it is 
seen that the material could have suffered but little secondary 
alteration. This is in agreement also with the percentages of water 


and carbon dioxide found. On the assumption that all of the 
water is united in the serpentine, the proportion of this mineral to 
the total mass is about 4 to 5. 

The Hoboken Serpentine. 

Geological Relations. — This area lies about seven miles from the 

northern extremity of the Staten Island serpentine and nearly in 

its line of strike. It forms a low, oval prominence known as Castle 

Point that reaches slightly out into the Hudson River. The east- 

FiG. 3. Serpenline rrom Hobokea, m ilnictuTeleu mass of lerpentine Gbtei. 
ern edge has been cut away by blasting and the rock is exposed 
for a distance of several hundred feet. The highest point of the 
prominence is some 60 or 70 feet above the river, but a sharp slope 
on all sides soon carries it to the level of the surrounding deposits 
which are only a few feet above the high-water mark. Cook 
mentions that the rock was found by drilling about one-fifth of a 
mile north of the exposed area at a depth of 175 feet. The geo- 
logical position of the mass is similar to that of Staten Island, 
being in contact with gneiss on the east and overlain by Triassic 
sandstone on the west. 

General Characters — In general appearance the serpentine re- 


sembles that found at some of the localities on Staten Island. It is 
usually light green in color, rather soft and of granular texture. 
Evidence of disturbance is shown by the frequent and irregular 
jointing, which gives it a brecciated appearance and in the slicken- 
sided surfaces. The joint-fissures are frequently filled with vein 
serpentine and the surfaces are covered with a film of carbonates. 
Octahedra and aggregates of chromite are abundant. 

The rock is thoroughly serpentinized throughout. No traces of 
the original silicates could be found in any part of the exposure. 

Microscopic Characters, — Microscopically, the serpentine is sim- 
ilar to the more altered types of Staten Island, particularly those 
occurring near Grant City and northward. It is composed of fibers 
and blades in irregular or approximately parallel arrangement, 
but without any of the well-defined structures described in the 
previous pages (Fig. 3). The fibers are sometimes strongly 
pleochroic and show a change of color that suggests chlorite, 
ranging from dark green with a bluish tint to yellow. A partial 
alteration to this mineral in such cases appears probable. The 
bladed variety gives a biaxial interference figure in convergent 
light with a variable axial angle. 

A peculiar microscopic feature is the presence of colorless, low- 
polarizing bands, which cut across the field in continuous lines in- 
dependently of the arrangement of the serpentine units. Their 
bleached appearance suggests that they may represent the course 
taken by percolating waters. 

Carbonates (comprising both dolomite and calcite) are always 
in evidence. They exhibit no crystal form, being usually in 
rounded granules enclosed in serpentine. They also fill small fis- 
sues. Chromite is rather more abundant than in the Staten Is- 
land specimens. It is perfectly fresh and exhibits the usual me- 
tallic lustre and crystal form. 

Talc and chlorite with small amounts of iron ores comprise the 
the remaining components and call for no special comment. 

Chemical Composition, — The similarity to the rock of Staten Is- 
land is well shown by an analysis by Mr. G. A. Goodell. 

SiO, 36.90 MgO 23.75 

AIjO, 1.29 CaO 15.53 

Fe,0 5.79 H,0 13.14 

Cr,0, 0.45 

FeO 1.46 Total 98.31 

[To be continued.) 


[Contribattons from the Geological Departmeat of Columbia University, Vol. IX., No. 
76: School of Mines Quarterly, Vol. XXII., No. a, pp. 318 to 329 ; 1901.] 



Part I. Geological Relations. 

By gilbert VAN INGEN. 

(Text Figures i-a). 

[Investigation prosecuted with the aid of a grant from the John Strong Ne<rberry 
Fund of the Council of the Scientific Alliance of New York. ] 


The work of the Geological Survey of Arkansas under J. C. 
Branner showed, among many other results of interest, the 
presence in the vicinity of Batesville, Independence county, of a 
series of highly fossiliferous limestones of Middle Paleozoic age. 
Specimens of the rocks and fossils from these limestones were, in 
1890, submitted for examination to Professor Henry S. William^, 
then at Cornell University, and were recognized as representing 
two distinct faunas ; one of Ordovicic age, the other of Siluric age. 
More material was obtained in 1 891 by Mr. Stuart Weller and sent 
to Professor Williams. The latter material was prepared and 
studied by the author of the present paper who was at that time 
engaged in special studies under the direction of Professor Williams. 
A list of the fauna contained in the limestone at St. Clair spring 
was then prepared, and it proved to contain upwards of seventy- 
five species, of which number a large proportion was recognized 
as having been described from the Clinton-Niagara faunas of New 
York, Ohio, Indiana and Tennessee. 

The results of that preliminary study of the collections sent to 
Professor Williams have been in part published in the Annual Re- 
ports of the Geological Survey of Arkansas (Refs. Branner, 1889; 
Penrose. 1891 ; Hopkins, 1893), and in a paper entitled, — ^' On the 
Age of the Manganese Beds of the Batesville Region of Ar- 
kansas," by Professor Williams (1894). In these earlier publica- 


tions the authors confine their remarks to discussions of the 
physical characters and geological ages of the different formations, 
giving only a few short lists of the contained fossils. The earlier 
statements regarding the Ordovicic and Siluric faunas (Penrose, 
i89i,and Hopkins, 1893) were somewhat indefinite and confusing. 
This uncertainty was cleared up by Professor Williams' analysis of 
the faunas, in which the distinctness of the two faunas, Ordovicic 
and Siluric, was fully demonstrated. 

In a volume of the Arkansas Geological Survey soon to appear. 
Professor Williams has a chapter on " The Paleozoic Faunas of 
Northern Arkansas," the page proofs of which we have very kindly 
been permitted to consult. In this paper Professor Williams re- 
views the work done on the Paleozoic faunas of northern Arkansas, 
gives lists of all the collections sent in by the different field parties, 
enumerates the species constituting the several faunas represented 
in those collections, and from the study of these faunas straightens 
out the nomenclature and discusses the stratigraphic positions and 
relationships of the various formations. On pp. 286-288 of this 
report is given a numbered list of the St. Clair limestone fauna from 
St. Clair spring (427-Al of our notes), in which list 69 species are 
named. On page 289 are brief notes on a few of the species ; but 
beyond these there are no observations on the particular expres- 
sions presented by the individuals of this fauna. The names em- 
ployed in this list will in a few cases need correction in the light of 
the larger collections and more perfect material now in our hands. 
When this may be necessary reference will be made to the above- 
mentioned list, using the numbers there applied to the particular 
species under discussion. 

Further collections from these formations were made in 1896 by 
the author of the present paper who, in company with Mr. Stuart 
Weller of the University of Chicago, spent two weeks at Batesville 
collecting from the many prolific fossil localities in that vicinity. 
These later collections are especially rich in the fossils of the Siluric 
limestone at St. Clair Spring, and as the elaboration of this mate- 
rial has brought to light several points of interest and also a number 
of species new to the Niagaran fauna of the North American prov- 
ince, it has been considered advisable to describe and illustrate the 
entire fauna from that and a neighboring locality. The present 
section on the geological relations will be followed by a second part 
dealing with the Crustacea, and that by others on the remaining 


hard Polk Bayou StC/air Carbonic 

Fig. I. Sketcli nup ihowing ihe diitributioa of the Ordorlcic and Silnric forma- 
tions, and the loc«1iIici, indicated by number), from wbidi colkctioni were secured in 
the vicinjly of Batenille, Arkaoia*. 


classes of organisms, and the correlation of this fauna with other 
Siluric faunas of America and Europe. 

The term Siluric is employed in this paper in the same sense in 
which it is used by Clarke and Schuchert (1899), embracing all the 
formations between the Richmond beds of the Ordovicic and the 
Coeymans limestone (Lower Pentamerus) of the Helderbergian 
group of the New York series. 

The material upon which this paper is based is in the collection 
of the Department of Geology of ColumbiaJUniversity ; the ex- 
penses of the trip on which it was secured having been largely de- 
frayed by the appropriation for the Summer School of Geology of 
this university. 

The author takes pleasure in acknowledging his indebtedness to 
the following gentlemen : To his esteemed former instructor Pro- 
fessor Henry S. Williams, of Yale University, who in 1 890-1 892 
kindly permitted the author to prepare and study the collections 
submitted by the Arkansas geologists; to Mr. Stuart Weller, his 
companion in the field, for assistance in collecting and for the loan 
of several specimens from the University of Chicago collections ; to 
Professor R. P. Whitfield, of the American Museum of Natural 
History, for the critical examination of several of the supposedly 
new species ; and to Professor J. F. Kemp of Columbia Univer- 
sity, who made possible the collection and publication of the ma- 


The material described in the following pages was obtained from 
two localities north of Batesville. Arkansas. Both of these locali- 
ties are indicated on the geological map, Batesville sheet, issued 
with the report on manganese by Penrose (Penrose, i 891), and on 
the revised edition of the same map issued with the report on the 
marbles by Hopkins (Hopkins, 1893). The sketch map. Fig. i, 
page 320, accompanying this article is based upon that issued with 
the Penrose report on manganese. The locality that yielded the 
larger amount of the material is St. Clair Spring, on the Hickory 
Valley road, about eight miles north of Batesville. This locality 
is No. 427-Ai of the Columbia University Geological Survey sta- 
tions, and 1240-A of the United States Geological Survey stations 
listed in the recent report on the Paleozoic formations of northern 
Arkansas by Williams (1901, pp. 285-286). The second locality 
where the Siluric fauna was studied is the abandoned Cason Man- 


GANESE Mine, No. 426-C, which is in the southwest quarter of 
section 34, township 14 north, range 6 west; about three miles 
northwest of Batesville, and about four miles southwest of the St. 
Clair spring outcrop. At both places che Siluric fauna is found in 
a very crystalline limestone, of massive bedding, and of light 
gray, white, or pink color. 

The Section at the Cason Mine, 426--C. 

This locality was the only one visited by us where the Silurfc 
limestone, with its characteristic fauna, is to be seen lying above the 
Ordovicic limestone, with the manganese bearing Cason shale in- 
tervening. The section at this point has been described, as respects 
its general features, by Penrose (1891, p. 219-220); by Hopkins 
(1893, p. 225); and by Williams (1894, p. 327; and 1901, pp. 

The mine is on a hillside in a gully that opens into the valley of 
Miller creek, and is situated in the southwest quarter of section 34, 
township 14 north, range 6 west. The opening is an " old pros- 
pect pit" referred to by Penrose (1891, p. 220), which is said by 
him " to have been sunk into the deposit for 20 feet without reach- 
ing the bottom." The pit was found by us to be filled nearly to 
the top by wash from the hillside. This thickness, 20 feet or more, 
given to the Cason shale, the " deposit " referred to, is more than 
twice the actual thickness as determined by us. The mouth of the 
pit is in the Siluric limestone just above its base, and about six 
feet of the Cason shale show in the pit. Just below and on the 
south side of the opening, and ten feet below the base of the 
Siluric limestone, is an outcrop of limestone of Ordovicic age. 
This restricts the maximum possible thickness of the Cason shale 
at this point to ten feet. 

The outcrop of coarsely crystalline limestone, 426-Ci mentioned 
above, showing on the hillside below the opening of the pit, is 
about twenty feet in thickness. Its age is considered to be Ordo- 
vicic and of a somewhat higher horizon than that containing the 
Rhynchotrema capax fauna (426-Bi) at the ford of Polk Bayou, 
about two miles west of the Cason mine, which launa has recently 
been listed by Williams (1901, p. 279). Unfortunately for care- 
ful correlative purposes, no fossils were found in this limestone at 
the Cason mine, and the determination of its age as Ordovicic was 
made entirely on its lithological characteristics, which are the same 


as those of the Polk Bayou limestone (426-Bi), and quite distinct 
from those of the Siluric limestone ; the former being a ** sugar 
rock " and less compact than the latter. 

In the prospect pit and extending downward to a level that is 
about four feet above that of the Ordovicic limestone exposed on 
the hillside is a brown shale, 426-02. This is the Cason shale of 
Williams (1894. p. 328; and 1901, pp. 296-299), the characters 
of which have been described by Penrose (1891, pp. 219-220), 
and the microscopic structure of which, determined by Wolff, is 
given in the same report (1891, p. 170). Some interesting details 
concerning the manner of distribution of the manganese concre- 
tions contained in the Cason shale, not incorporated in the descrip- 
tions of that deposit furnished by Penrose and Williams, were 
observed by the present writer. 

This Cason shale is, as stated above, approximately ten feet 
thick. Its base was not seen, because of the filling of the pit by 
surface wash. The lowest part seen was a rather soft, incoherent, 
chocolate-brown shale, of irregular stratification, with some frag- 
mentary fossils, and full of flattened manganese concretions. These 
concretions vary in size up to three inches in length, though the 
greater number are between one and two inches. They are most 
abundant in the lowest visible layers of the shale, and become less 
abundant in the superjacent layers, as the shale loses its arenaceous 
facies and assumes the calcareous facies of the overlying Siluric 
limestone. The color and arenaceous character of the shale change 
gradually from the lower to the upper layers. The color varies 
from a dark chocolate-brown, through dirty yellow, to cream, pink 
and gray, as the rock changes from a soft shale to a compact lime- 
stone that merges into the overlying limestone, 426-C3. The 
change is brought about by the intercalation of thin seams of 
limestone which gradually become thicker and finally entirely re- 
place the diminishing thicknesses of the shale. This change from 
the shaly facies to the calcareous facies is so gradual that the line of 
contact between the two beds can with difficulty be decided upon. 
I have taken the boundary line at the top of the highest bed con- 
taining well developed nodules of manganese, whatever their size. 

A lull series of specimens of the rock and manganese concretions 
was collected from every few inches of the Cason shale, which 
specimens showed well the mode of occurrence of the ore. 

VOL. XXII.— 22. 


The manganese seems to be here in the original form in which 
it was accumulated, enclosed in a mud or clay that was probably 
the residual soil resulting from the subaerial erosion of the Ordo- 
vicic limestone. Such erosion took place during a portion of late 
Ordovictc and early Siluric time, when the Ordovicic limestones 
already deposited had been elevated above sea level. For some 
notes on the elevation of the continents during this time, see 
Weller's paper on the epi -continental nature of the Siluric sea 
(Weller, 1898); also Weller (1900, pp. 12-22); and Wiluams 
(1901, pp. 295-301). The manganese ore found in place in the 
Cason shale at the Cason mine proved to be non-marketable be- 
cause of its high percentage of contained phosphorus {Penrose, 
1891, p. 221), which fact caused the abandonment of the workings 
there. The marketable ore, on the other hand, seems to be in all 

Fig. 3. Diagram illastraliog the relalioiu of the varioui geological formations to 
each other, and to the deposits of good and non.marketable manganese ore. Not 
drawn to scale. 

cases restricted to the recrystallized ore which, after erosion, has 
been accumulated in the lower portions of the residual clay that 
fills depressions in the eroded Ordovicic limestone. Residual clays 
filling depressions in the Siluric limestonealone, were, on the other 
hand, found to be lacking in manganese ore. It is only when the 
erosion has been so long continued as to have formed hollows 
below the level of the Cason shale, and the drainage has been of 
such nature as to permit of little transportation of the eroded 
material, that the marketable non-phosphorus-bearing ore is found, 
recrystallized, and forming large irregular masses in the residual 

A suite of carefully selected specimens of the Cason shale rocks 
should be submitted to chemical and microscopical analysis, with 


a view to ascertaining the exact conditions under which this man- 
ganese bearing shale was deposited. In the mean time it is the 
author s opinion that the shale was formed during a period of ter- 
restrial conditions, but whether the place of deposition was upon 
land, in fresh water, or in sea-coast swamps or deltas, cannot at 
present be determined. For some observations on the various 
modes of deposition of manganese ores the reader should consult 
ihe admirable statement by Penrose (1891, pp. 550-569); also 
Williams (1894, p. 329). 

Certain of the facts stated above indicate that the Cason shale 
should be considered as of Siluric age, and as representing the 
very earliest portion of Siluric time, during which land conditions 
existed over this part of the Arkansas region. 

The fossils found in the Cason shale are all in a very fragmen- 
tary state, and occur only in the upper portions, after the thin seams 
of limestone begin to be intercalated with the shale ; in other 
words after the commencement of marine deposition. Only one 
specimen at all recognizable was found by us, that a species of 
Straparollus of uncertain specific identity, from the upper part 
of the shale bed. 

Professor Williams (1901 , pp. 297-299) presents some evidence 
in favor of the Siluric age of the Cason shale, derived from the 
examination of specimens furnished by the Arkansas geologists. 
As suggested by Professor Williams, the specimen No. 31 is prob- 
ably a loose fragment derived from the limestone that lies above 
the shale bed. The two species cited are very abundant in the 
Siluric limestone. The evidence from the next specimen, No. 32, 
is of value, since the specimen is without doubt derived from the 
shale itself, as is indicated by ** the concretionary-like buttons or 
pellets." The presence in this specimen of Plectambonites seri- 
CEUS intermedius, Ringueberg and Atrypina disparilis Hall is 
sufficient to stamp part of the deposit as of Siluric age, and more 
particularly, as stated by Professor Williams, of early Siluric age. 
The exact portion of the Cason shale form which this specimen 
was obtained is, however, unknown though it in all probability 
came from the upper layers. 

The Siluric limestone, 426-C39 that overlies the Cason shale, is 
seen oil the hillside above the pit. About twenty feet in thickness 
was exposed, and other outcrops further up the hillside, but not 


examined by us, indicate a total thickness of over lOO feet for this 
formation. The rock is a compact crystalline limestone, of gray 
or pink color. The lower part, just above the Cason shale, is thinly 
bedded, and stained with the chocolate-brown and yellow of the 
manganese, but contains no manganese concretions. Higher up 
this lormation is a heavily bedded, pure limestone, with occasional 
bands containing crinoidal fragments and other organic remains. 
The collection of fossils secured from this bed, 426--C39 was from 
such a fossiliferous band about twelve feet above the base, and was 
not as large as might be desired. The contained fauna is about 
the same as that from the St. Clair spring locality, 427-A I, though 
not so large in point of number of species. Certain topographic 
facts indicate that the horizon of this 426-C3 fauna is somewhat 
lower than that of 427-Al at St. Clair spring, though more care- 
ful work will be necessary to determine this point. 

St. Clair Spring, 427-Ai. 

The particular point from which our Siluric fossils were obtained 
in this neighborhood is a small gully in the hillside just back 
(east) of the St. Clair spring on Miller creek, at the east side of 
the road from Batesville to Hickory Valley, and about eight miles 
north of Batesville. This locality is described by Penrose (1891, 
p. 134}, Hopkins (1893, P« 234) ; and the fauna from which is briefly 
noticed by Williams (1894, pp. 330-331), and again in greater 
detail by the latter author (1901, pp. 286-288). 

The Siluric limestone is here exposed in an escarpment that ex- 
tends along the westerly face of the ridge that forms the eastern 
side of the valley of Miller creek. Just back of the St. Clair 
spring is a small gully, and in this gully was collected our material 
from the large blocks that had become loosened from the fifteen 
foot escarpment of the limestone. The only rock seen in place in 
this gully above the point where our material was obtained was 
the single bed,427-Al, of Siluric limestone ; and as our collections 
were carefully made from loose masses very little removed from 
the parent ledge, there is no possibility of our material containing 
representatives of any other fauna than that of the Siluric lime- 
stone. The lower portions of both gully and hillside are covered 
by talus and residual soil so that the underlying formations can 
not be seen. 

The bed in question is a massive, crystalline, pure limestone ; 


abundantly fossiliferous. In color the rock varies from gray to 
white and pink. Solution in hydrochloric acid failed to show the 
presence of any grains of quartz or flakes of mica. The only 
particles found, as residue after solution, were small ocherous 
masses that resembled oxidized iron pyrites. These facts may be 
taken as indication that the rock had its origin in a marine basin, 
the waters of which were quite free from terriginous sediments. 
In many places the rock is an aggregate of crinoidal fragments, 
and then presents on its freshly broken surface the sparkling ap- 
pearance of a metamorphic marble, due to the cleavage faces of 
the calcite in the crinoid fragments. In other restricted portions 
the matrix is almost pure white calcite. In the latter event the 
contained fossils are found to be in the most perfect state of pres- 
ervation ; the brachiopods having usually both valves in articula- 
tion. A species of Conocardium (sub gen. Rhipidocardium, 
Fischer, 1887) with a broadly winged carina on each valve has 
usually the two valves united. There are few shells that have the 
appearance of being water-worn or crushed; such imperfections 
as have been observed being due to the recrystallization of the 
calcite. These facts seem to indicate that the waters in which this 
fauna lived and flourished were rather quiet and free from any cur- 
rents of a strength sufficient to transport along the bottom the 
hard parts of the dead organisms. The trilobites of this fauna are 
in all cases in a fragmentary condition, as the specimens are the 
cast-off" and dismembered moults that have fallen to the bottom. 

A still unexplained peculiarity of the distribution of trilobite 
fragments is well exhibited in this limestone. Some portions hold 
large numbers of the cranidia of iLLiENUS, while the pygidia are 
found more abundantly in other portions. Of Ceraurus the 
cephala are abundant ; the pygidia rare. Of Dalmanites both the 
cephala and the pygidia are found together in about equal num- 
bers, while fragments of the thorax are rare. Lichas is represented 
by abundant pygidia and by few cephala: Arges, an allied form, 
presenting the opposite proportion. So little is known of the 
moulting habits of modern Crustacea that almost no assistance can 
be had from that source in the attempt to explain this irregular 
distribution of the separate elements of the trilobite tests. 

Another feature of considerable interest is the small size ot many 
of the species constituting the fauna; reminding one of the depau- 
perate St. Louis fauna of Spergen Hill, and the sub-Carboniferous 


fauna of Windsor, Nova Scotia. Among our Siluric fauna the 
brachiopods are especially noticeable on account of the small size 
of all their species ; which are only a third or half the size attained 
by the same species at Waldron, Indiana; and the Wisconsin, and 
New York localities of the Niagaran shales and limestones. Some 
genera of the trilobites are represented by specimens of normal 
size, comparing favorably with those from the other localities ; and 
with these are associated great numbers of smaller specimens of 
young individuals. iLLiENus, Bronteuj«, Dalmanites, Ceraurus 
are of this category. The PROEXiDiE, AciDASPiDiE, LiCHADiDiE, 
Ampyx, are all of small size. The Gastropoda are all small, so also 
are the Lamellibranchiata and Cephalopoda. No explanation for 
this decrease in size of the species of a fauna will be here attempted, 
although it is hoped that the detailed study of the species may af- 
ford some clue to the probable cause. 
Columbia University, March 14, 1901. 

Branner, John C. 

1889. Annual Report of the Geological Survey of Arkansas for 1889, 

vol. I. 

1896. Thickness of the Paleozoic sediments in Arkansas. 

{Amer, Jour, Set., Ser. 4, vol. II., pp. 229-236 ; map. 

1897. The former extension of the Appalachians across Mississippi, 

Louisiana and Texas. 

(Amer. Jour. Sci,, Ser. 4, vol. IV., pp. 357-371 ; a maps. 

Nov., 1897.) 

Clarke, John M. (and Charles Schuchert). 

1899. The nomenclature of the New York series of geological forma- 

{Science, N. S., vol. X., No. 259, pp. 874-878.) 
Reprinted {Amer. Geologist, vol. XXV., pp. 11 4- 119). 

Hopkins, T. C. 

1893. Marbles and other limestones. 

Annual report of the Geological Survey of Arkansas for 
1890, vol. IV., pp. xxi. and 443; numerous text figures and 
6 maps. 8vo. Little Rock, Ark. 1893. 

Penrose, R. A. F., Jr. 

1 89 1. Manganese: its uses, ores and deposits. 

Annual report of the Geological Survey of Arkansas for 
1890, vol. I., pp. xxi. and 642; 40 figures, 1 map (col.). 
8vo. Little Rock, Ark. 1891. 


Williams, Henry Shalkr. 

1894. On the age of the manganese beds of the Bates ville region 
of Arkansas. 

{Anur. Jour. Sct\, Ser. 3, vol. XLVIIL, pp. 325-331. Oc- 
tober, 1894.) 

1 901. The Paleozoic faunas of northern Arkansas. 

(Arkansas Geological Survey, chapter VIL, pp.268-362.) 
This work has not yet been issued, page proof only having 
been seen. 




Cycle II. 

Let Fig. 9 be the P. V. and Fig. 10 the 0<p diagram of this 



FW. 9. 

FIG. 10. 


In the compression cycles the volume ratio " will enter into 


many of our formulae and we will find it convenient to write 




= r. 

* Continued from page 252, Vol. 22. 


The compression is adiabatic, hence 

,;) = Tj-^-'- (2) 

During addition of heat v^ = v^ and therefore 

/*c = A 7- 


y 7/ 

If we write 7^^ = i + ^ y. = ^ as in the previous cycle we get 

Pc--P,X^PaVX (3) 

t: - rjv-ix (4) 

Adiabatic expansion gives 

^d = ^.(r)'^ or if /^ = A 


""" "' (~/r)' " '''''^' " 'P^^ ^ ''"^'* ^^^ 


7;=7;-f=r„xv. (6) 


Applying the perfect gas law 

P 7^ « 
^- = a:, 

A^'» _ A.r''^a_ _ r, 

T, Tjy-iXr - '^' 


Pi^A P V Xy „ 
' TXy 

The heat discharged 

^. = Qt; - T,) = c; 7:x^> - I). (7) 


The work done is 

W=. H,-H,^H,- CT{X\ - I) (8) 

H H ^^' 






Cycle II. 

Sjnmbol. Formula as First Derived. Formula Reduced. 

A a(^:)' pr 






J 5) 

f . . . . t'. 

7^» "; (r arbitrary) 



•A 7-- PJ'X 

■^'c ^6 



^« ^4' + #rj ^*^ 



V (^^\> vx'y 


T, T ^ TXy 



H, CIT^-T^) CT{X-^ - I) 

W H,-H, H,- CTiX •< - X-) 

H, H, 

R, v^-v^ vlX^ - I) 

M.E.P. jZ y ^-^y--" 



^, CJog.5 C,\og,X 


R, A- A pJcr^- 

R T; - T; TJ,R<-'X- I) 


Cycle II. A. 

Let Fig. II be the P. V. and Fig. 1 2 the Of diagram for the 


TIO, 1 1. 

PIO. 12. 

Then we have, since the compression is as in Cycle II., 
Also for C the heat addition being as before 

^. = «'». 

P, = PiX=.pj^X, 
T,~ T,X ^ Tjy-^X. 





The point D lies arbitrarily between C and the atmospheric line 
on the adiabatic. 

From this point we will consider two cases: i°, the general 
where v^ is greater than v^, and 2°, a particular case where 



^4 > "^a a"d Pd > A 
Then we have 

^d = ^« 

/>.=A^(y (8) 

r.=. Tr-*-'X 

V \^~' 


^1 (9) 

V, == t'^ 




•~ "A" "A 

A' = A^ (8') 

r/ = Tx 

1 a 

2/ / = V. 

p: = A 




■ T => T 



a a 


Apply the perfect gas law 

fa a 


= R, 

A^'»_ Ar^'a _ r> 

T, - Tjy-^r 

pv p ffXv „ 



r, =/?. 


Pl^l _Pa^a_ r, 

p 1' p V. „ 

^, ~ 7. ^i ~ 

a 2t 


PX _ ^'aP, 

T' ~ T 


Heat is abstracted as follows : 

"t=C,{T,-T,) Hi^C,{T,-T:^ (13') 

The work is given by 
IV = H,-H, W^H,- CJIX- I) (14') 


Volume range is 

-^."'^.-^'.^^.-y (15) ^; = ^'a-^6=^«(l-^) (15') 

w w w 

M.E.P.=/^ =/ (16) M.E.P.=/-- (16') 

i> 'i 


Entropy range is the same for both cases, 

Rip = C; log^ ^ = C; log, .V. (17) 

Mean of mean temperatures of heat addition and abstraction, 




_ I H, -^H, 
- 2 C, log. X 




\ H, + CJSX- I) 
C, log. ^ 






C/log. X ' 
Pressure range is same for both cases 

Mean effective volume 


Temperature range is also the same for both 


Cycle II. B. 

Let Fig. 13 and Fig. 14 be the P.V. and Oip diagram respec- 
tively of the cycle. 


A. E 



FIG. 13. 

FIG. 14. 


Assume same results as before up to the point c. Take p^ 
something less than atmosphere, i. e.. 



A>A>0- (') 


Through D and a point E whose volume is greater than the orig- 
inal we have an isothermal 


"A "\ />. / A 


'A " V A / A 




^.= ^',.'^^(^^)^' (5) 

A = A " (6) 

Apply the perfect gas law to the points 

AJ'» _ A^ar^ _ r> 

fee __^ t ai n ^___ r> 

7; - Tjy-^Xr - "^ 

\ 1 

^<. ^v.1//^^^-' ' 

A^. PJ'..^-'{P,?i-' _j. 

• ^«^'(^')ma)^ 


During the isothermal compression heat must be abstracted, the 
amount can best be calculated by Of coordinates. Call this 
amount m, then, 


f d - f . = (y. - f ») - (f . - fa) 

f . - f » = C. log. -f = C. log. ^' 



Besides this amount m we must abstract a quantity C^{T^^ T^) 
isopiestically, whence 

H,= cjj- 7;)+ t: j(rjog.^+ <^,iog.[^^(^')'^]) 


W^H^- H„ (8) 

£=i-5- (9) 

+ ^P 

The volume range is 

^,= ^'<.-^6 = ^'„'^M-. 



M.E.P. = /-„- = f-T— i , (II) 

"■ -'4-'(^:)'-;] 




.E.T. = J ( 

C, log. X 


K=Pc-Pi^ PXryx - A. 

M.E.V. =/ 


^r= ^{{ry'^X- I) as for II. 



Cycle II. C. 

Let Fig. 15 be the P.V. and Fig. 16 the Oip diagrams of the 



FIG. 15. 

FIG. 16. 

All values for the compression and heat addition found in Cycle 
II. may here be assumed. The point D lies at the intersection of 
two curves, one an adiabatic through C, the other an isothermal 
through A and the relations can be written. From the adiabatic 

"' - ". {';)"■ 

From the isothermal relation 


Equating we get 


° \PJ "Pa 

VOL. XXII. — 23. 


r -)-'="• 




This is the pressure at which the intersection will take place. 
By substitution we get 

^i^-^J^"-' (2) 

T, = 7; (3) 

Applying the perfect gas law to D 


rr* *"^ I "^" ^T^ ^^ d.\.» 

^ Xy-^ T 


All the heat is abstracted at constant temperature during the 
compression D to A. The entropy range is evidently the same as 
for heat addition and this is 




Rf = C, log, X (4) 

^,= Ufa-f.)=KC,iog,X (5) 

W=I/,-H,= H,-T^CJog,X (6) 
//, TC log. X 


= ^. - ^'* = ^ a [^^-' - !. ] (9) 

M.E.P. = -^j^- =/ ■ - "r ^ -.- (10) 


v^ I XV-. 




^, = A - A = A [rX- X'^-y^ 

Rj. = 7; (r*-'^ — i) as before 






Cycle II. 

Formula as Brst derived. 


Formula reduced. 

• • • • pj" 
















'^A^'^cf) ^"'-^-^ 




Tlfi-f^ ^„^. log.^' 

.■H,-H, H,-T^C^og,X 

I — 


I — 

r c" log X 


C-.log.^' C.log.^ 



( Clog. A' ^'•) 





M.E.P / 




Hy - T^C, log. X 


P,-Pi PifX- X^-y) 

M.E.V / 




T~ T 

c a 

f^,-KC,log,X ] 

TXry-'X- I) 

Cycle III. 
Let Fig. 17 be its P.V. and Fig. 18 its Of diagram. 


FIG. 17. 


The compression results of Cycle II. may be assumed, hence 




A = PJ", (2) 

T,= Kr-'- (3) 

Heat is added isopiestically, hence calling C^ the specific heat at 
constant pressure we have 

P p 6 


1 V 

.•.7;-r,F=7>'-r. (4) 

p,=p,=pr> (5) 

^'.-^»5 = ^.>'=7>'- (6) 

Adiabatic expansion gives for final pressure of one atmosphere 

)v = z,^K=n'.. (7) 


^d = 



A = 




1 . 


.•.7; = 




the perfect gas law 

X a a 

7; - 


7"* ^ 


as in II. 


ArX J' _ ; 


An AnJ^_ ^p 
„ = r F ~ 

a a 

Hence the formulae are verified. 
Heat is abstracted isopiestically 


» C,UY- I) (10) 

W^H,-H,^H,-C,T^{Y-i) (,i) 

Z7 f^ ^2 CJIY- i) , , 


Volume range is 

R, = v^-v^ = v^\Y--'^ (13) 
Whence for mean effective pressure 

M.E.P. -y '^-/fi-^'iO:=-L> (.4) 


^f = fc - f » = <^p log. j' = ^p log. F (15) 

^p = A-A = A(r-i) (17) 

M E V = 7 ^ - /^a.TlGl^«(?J^) f ,8) 





Rr^T^-T^^Tir-'Y-i) (19) 

Cycle III. 

Symbol. Formula as derived. Formula reduced. 

A pXtS ^■■■•■P:r 

r r 

A Ar' 


z'e i-ty -^Y 

■^ b 

P h 

A A- 

^*(' + crj ^-'^-•^ 

'...•- </ I — 1^ 1/ 

Pa " 


V, vA-rh vY 



cjj,-t:) c;riY-i) 

H,-H, H,-CTlY-i) 

I — 

I — 




CAog,^ Clog.F 

[ H,-C,T J, Y-i) 


M.E.P / 




■P^-Pa A(r-I) 

M.E.V / 






Cycle III. A. 
Fig. 19 is its P.V. and Fig. 20 its Otp diagram. 


no. i». 




'-"'^ \ 



^— — --''''''"^ 


FIG. 20 


Assume the results of III. up to point C. The point D is situ- 
ated anywhere on the adiatoatic through C between C and atmos 



P,>P,>P. (1) 


^d > ^«- (2) 

This latter (2) will not necessarily follow from (i) but where it 
does not hold the cycle is decidedly'imperfect and this case is here 
neglected, i, e,, the case where the isometric DE cuts the adiabatic 

We have then 

A = A (5) 

^»=^.=^<,>'(^;)^ (6) 

Apply the perfect gas law to D and £. 

P \' 

r a I Y 

p,v YV'\ 


pitj^L^yp-d ._ p 

" \pj 
This verifies the formulae. 

Heat is abstracted in two parts and the amount is 

h,= cit,-t:)+cit-t:^ (8) 




The work done is 
and efficiency 

fF= //, - H^ 




M.E.P.=/p =/ 


"■ -["(^f-] 


Rf s C" log^ F as before for III. 
^, = A - A = A(?^ -0 as in III. 

■^■^-^\ R^ /-^lc;iog.Kj 




As before III. the temperature range 










Cycle III. B. 
Figs. 21 and 22 are its diagrams 








_— -"^-^ 


FIG. 21 

FIG. 22. 


All results of III. -^ up to period D may be assumed except 

that p^ which was there arbitrary was assumed greater than p^ is 

here less than /^, /. c, 

P.>Pa>0 (I) 

We had 




irough E and D there must pass an isothermal and 





j,^,, A=^ a^^j,./a\;a 

• "A "A " v./ A 



T.= T,= Tj[^^y 


Applying the perfect gas law to E 


py. ^" " \pJ 


Heat abstracted i ° isothermally a quantity w. 

2° isopiestically " //. 

»-£;(?•. -r.) 



f rf - f . = (f 6 - ^.) - (f , - fa). 

= C, log, 5 = C log. !•. 




= C^\og,Y-C^\og,Y[^^p\ 

= C,o,.Y[.-{^P\ 
K(^-)V_. + Klog.F[i-(^;)T]|. (9) 

P o 

£= I- 


K-n-^.-^.V^jS- -\\ 











R^^ T^^T^^ T^{yy'iV^ i) as before III. (17) 

Cycle III. C. 
Let Figs. 23 and 24 be its diagrams. 







FIO. 23. 

FIG 94. 


We will assume all results to C already derived. The point D 
is determined by the intersection of the adiabatic through C with 
the isothermal through A, From the adiabatic relation 

P V 


From the isothermal relation 


'' "A 

'\pj ''Pa r\paf 

• p =.A 


By substitution 



T ^T 


=a-a=a(?"'- \ ) 





^. = Uf. - f *) = T^C^ log. F (4) 

fF= //.-//, = /^. - 7;C; log, F (5) 

' //, 

^. = n,-^'» = ^„[^'>-~^--5.] (6) 

I' I Fr-i 

^,=A-A = A ?"'- V • (8) 

M.E.v../-^-==y^'r/^.^.'og,^ ^^^ 




" ' (10) 

"Mciog r"*" ^°) 

c, log, r 

Rt^T,-T^= TXf-'Y- I] as in III. (i i) 

Cycle III. C. 

Symbol. Fonnula as derived. Fonnula reduced. 

A pXvS f-^ 

a a 

r r 

^-Ct)" ^-^^ 


A Pr 

T V 

f, V' -^F 

/ Fi-> 

^^ ^aY'-' 

T, r. 

^, Uf.-f,) TS,\og,Y 

^' H,-H, H,-TC log, Y 

^ ^-H, '~ //, ' 

^9' <^p log, ^' Cp log. F 


*\ /?f ^ ^AQog^F^ ''^ 




^.-n ^'a(l^^-'-J) 

M.E.P / 








r H, - TC log. Y 


PSf - y-'') 

T-T,^ Tly-^-^Y-x) 

Cycle IV. 
Figs. 25 and 26 are its diagrams. 




FIG. 26.. 


FIG. 26. 

We may assume the results already obtained for the compres- 
sion but beyond that new conditions arise. By isothermal heating 
the curve approaches the atmospheric line and there will be a cer- 
tain quantity of heat that will bring the isothermal down to the 
atmospheric line leaving a subsequent adiabatic expansion an im- 
possibility. This quantity of course depends on the location of B^ 
/. €,, the amount of previous compression. The higher the pre- 


vious compression the more heat may we add isothermally before 
reaching atmospheric pressure. 

The quantity of heat which will make adiabatic expansion im- 
possible and stop the isothermal on the atmospheric line can best 
be determined from 0(p relations. Denote this quantity by Q, 


Fig. 45. 

On the Oif diagram Fig. 45 the point 3 lies at the intersection 
of the isothermal 23 drawn at temperature O2 the compression 
temperature and the isopiestic 1 3 drawn from atmospheric tem- 
perature 0\ to the intersection 3. In each case the entropy range 




Apply now to the Cycle IV. 


= Tyr-^ log. r-' 


This is the amount of heat that will bring C down to atmosphere 
with no adiabatic expansion. In order that the cycle may exist 
according to the hypothetical definition we must add less heat 
than this quantity Q. Hence we have the equation of condition 
for the existance of the cycle 


//, < T,r-' log, r^-' 
^. < 7; log. ^a . 



A similar method can be used to find the amount of expansion 
or resulting pressure and volume after addition of //j, BTC/ of 

Draw on both diagrams the isopiestic through the termination 
C of the isothermal and cutting th^ adiabatic AB at point c'. 




T T 

<Pc - f,' = ^i, 'og. 7- = <^p log. Y 

= CJog, "^ 



- (1:)- 





A A 



And the amount of heat necessary for this isothermal expansion 
from B to C 

^.= 7;c;iog.(|^)';. 


"- I 


r (-'p c^ ' 


'°^' A = nc^ - Q' 


"^ ,,, = z 

Then will 


That is to say if we start at state B add a quantity of heat //j, 
isothermally the resulting pressure is 



A-^-^>^-. (3) 

A ^ !:., 
A "^ ^»' 

-• = ^ 
^. ' 

J/. = v.e' = y- «^, (4) 

^. = ^» = 2:r>-'. (5) 

A = A. (6) 

" ''a' r V^A' 



We had III., 



^'d=^'a<'''"'- (7) 

VOL. XXII. — 34. 




^- = ^« (a )'^ ^ ^' ^PJ^^ " ^"''^'^^ " ^•'''"'- ^^^ 

Apply the perfect gas law 

p V 

* a a 


= i? 

a a 

Verifying the formulae. 


W^ H,-H,^H,- C^TSe^-' - i) 








r r-, I 1 

" ' "L rj 





.-. M.E.T. = ^T; + -i^-"-^^— -? J 

= ^r,.-. ^ ^iw-C^-'-^i)] (,3) 




^,=A-A=A(r^-i). (16) 

^r=7;-7;=7;(r>-'-i). (is) 

Cycle IV. 

Symbol. Fonnula as derived. Formula reduced. 

A pXvS ^-'" 


r r 

;■) 7;r- 

Equation of condition. . . H^<iT^ log^ y,* . . . //j < 7^„r^~^ log^ ;'^~i 



A Ar^ 
//, e^ 


v^ vf' ""^^ 

A r 

t; 7; 7;r>-' 

A A A 


7'^ z'. I y I '' vj'^-^ 





T, T^\^f)'' V-' 

^. Q^.- ^„) c^U^'-' - 

w" ^,-^. //,-c;7;(.''--i) 


* ^. r-7; 




i/^, + ^. 





/ ■ • 

rv-'7:r/f,+ c,7;(.'--*-i)-j 


»rf - ^'6 


M.E.P J 






A- A A(r-i) 




R • ■ • 




Pair-' - 


I) J 

T,-T^ TJjy-'-i) 

Cycle IV. A. 

Figs. 27 and 28 are its diagrams. 


FIG. 27. 




FIG. 28. 

We may assume the results of IV. up to point C. The point D 
lies somewhere on the adiabatic between C and atmosphere and is 
subject to the conditions 

A>A>A (0 


^U > ^a- (2) 




■•■^',=^.(W^^~'- ^3) 


•••:7;=:r„(^-)'vV- (4) 

^.= ^'. = ^'„(J"Ji^^-^ (5) 

■'A "Va' A "Va/ 

Apply the perfect gas law to Z> and £. 

A^rf_A^«A^^''"'A'' _ ;p 
r — I v-i ~ ^ 
' Pa-' Tj, y e^-^ 

A^:.^^*'"VA/_ _^ 


The heat abstracted is 



■■ -w, = cj,,^ 



(;:)'B:-]-^-'-.[e:)''"-] <^> 

W=H,-H, (7) 



^ ' H^ 

•^, = ^rf — ^» ' „ 

r a a a 


M.E.P. =/ ^ -,- 




M.E.V. =/-,— r 

M.E.T. = J 











Cycle IV. B. 

Let Figs. 29 and 30 be its diagrams. 

P iB 





FIG. 29. 

FIG. 80. 

The operations up to C are as in IV. and we may assume those 

The point D is subject to the condition 

and the point E to the condition 

V. > v^, ' (2) 



^'.= ^a(^" )''•''-' (3) 

T. = T. (^^ )'-".-- (4) 

d — *a 

T=T^= T 

' ' °''A 

; (f l"^''- (5) 

A = A 

t/. = v/'"- - r l-^ \*> V-» 

« d y, a 

(9. } ' '"■ («) 


Following the methods already adopted we can write 

f . - f a = ^, log. -f- 




IT- Jf, - H, (8) 


E^i-^- (9) 



^.=-.— .=^'„[(^;)>..--;]. 

.-. M.E.P. =/ 



^ Kfi"-' 


A / ; J 


.: M.E.V. =/ 


R -^•- ^' 

*~T,- Tjy- 



Cycle IV. C. 
Let Figs. 31 and 32 be its diagrams. 









FIG. 31. 

Assume results up to C as in IV. 

FIG. 32. 


The adiabatic through C must meet the isothermal through A 
to locate the point D, 

From the adiabatic relations 

From the isothermal relation 

(2/ \Y V 

By substitution 

d ' ^d 

X e c in ad 


.■.v^ = vy=rv^. (l) 

' * \P.v.) \ e'r-^pv. ) 

Pi^^=l (2) 

T, = 7;. (3) 

By inspection it is easily seen the perfect gas law is satisfied. 

^h -^ at 

•••^.= ^^1 (4) 

fF=/^,-//, = /^,(i --'_,) (5) 


. = ^'„-^»=^'„(^-p) (7) 

M.E.P.-y 7-^-77 (8) 


K=A-p,=A(r-^) (9) 

M.E.V. =/ 

M.E.T. = K^« + 7;) - 5"(i + r-'). (10) 

Cycle IV. — C. 

Symbol. Formula as derived. Formula reduced. 

r r 

^* ^<.(.:) ^j-^- 

Equation of condition //", > o ff^> o 

A Ar 


V v/^ ^"^ 

'A r 

t: 7; 7>-« 



(J ~a 

T, 7; 

^. 7:(^.-fj 


W^ H^-H^ H. 








^. + ^, 




M.E.P / 



I — 





M.E.V / 




7;-7; TJjy-^^i) 

I — 


Cycle V. 
Let Figs. 33 and 34 be the diagrams of the cycles. 




FIG. 33. 

WO. i4. 


If we add heat at increasing/, v, and 7" the curves of states will 
lie somewhere between the isometric and isopiestic on both dia- 
grams and the cycle is somewhere between III. and II. If the 


heat addition took place at decreasing /. increasing v and _;' the 
curve of states would lie between the isopiestic and the isothermal 
and the cycle lie between III. and IV. We cannot, however, cal- 
culate the appropriate set of formulae without knowing the law of 
variation of states. The number of ways of variation is infinite, 
and while any one might be assumed, nothing could be gained by 
the calculation unless the law of variation chosen were preem- 
inently simple or maintains in practice. Whatever it may be, how- 
ever, the previous discussion will enable us to class it pretty well 
without entering much into details. 

[G>iitributioii from the Physical Laboratory, Columbia University.] 


By henry ST. JOHN HYDE, Ph.B., A.M. 


The standard Clark cell can only be used under special condi- 
tions because it is only constant with 10,000 ohms. 

It is important to have a cell constant through low resistance 
and the " Mesco dry Battery" and other "dry cells" are con- 
stant with S-io ohms. 

The following experiments were made in order to ascertain the 
constancy of the E.M.P'. under varying conditions. 

Methods of Testing. 

The first method employed was the measurement of potential 
difference (P.D.) by the High Resistance Method. 

1 //. /P. f 


Fig. I. 




This method depends upon the principle that if the resistance in 
a circuit be constant, the current (and therefore the galvanometer 
deflection), is directly proportional to the voltage. 

The only change of resistance made in the circuit is that due to 
the internal resistance of the cell or battery E^ (Fig. i). 

The terminals were joined, from the galvanometer (Fig. i), and 
high resistance !H.R. (Fig. i) to a standard Clark cell (£",) and 

the deflection {d^ was noted. 

Then the deflection (rf,) was observed, given by the unknown 
cell (£,). 

The voltage was calculated from the proportion : 

It was considered better to find the value of one scale division 
in volts, and to multiply this value by the deflection of the un- 
known cells. 


i^ = ,* = ''constant." 

^2 = " constant " x rf^. 

The voltage of the standard Clark cell 

= 1.43 volts (approximately). 
= 1.438 legal volts at 59° F. 

From this value, must be subtracted i, in the last decimal place, 
for each 2^ above 59° F. 

Clark Cell = E^, 
I*" = -7-' = '' constant " = '^-^ = .0065 (legal volts at 70^ F.). 

•* GoNDA " Leclanche Cell = E^. 

^2 ='* constant " )S: ^^ = -^^^5 ^ 225= 1.465, 

.-. E.M.F. = 1.465 volts. 

E.M.F. OF A Daniell Cell. 
£2= "constant" X d^^ .0065 x 165 = 1. 074 volts. 

.-. E.M.F. = 1.074 volts. 



The next method employed was the potenttometer method. 

The arrangement for the comparison of standard cells by the 


t--' 1 

-e— iJ 

I 12} 

Fig. 2. 

potentiometer method, is shown by Fig. 2 (Poggendorfs Com- 
pensation Method). 

When the potential difference (P.D.) between the points, e^g^ 
along the resistance R^ is equal to E^ there will be no flow of cur- 
rent from the branch circuit containing E^, and the galvanometer 
will show no deflection. 

The resistance R,S^ may be a long GS wire wiih a movable 
contact gy but this resistance, if a bridge be employed, should be 
much greater than the internal resistance of the constant battery, ^^ 

A resistance r, of not less than io,ooo ohms is placed in series, 
with the standard cell, to prevent polarization during adjustment. 

When a balance has been obtained, the value R^ is read off, and 
another cell, E^, is substituted for -£,, and the value of ^, is deter- 
• mined for equilibrium. 


1 • 2 • • 1 * 2* 

In practice it is better to use two rheostats (Fig. 3) in place of 
a bridge wire. 


J ^000 OAj/ts. 



h-ta — © — iJ 

Fig. 3. 

In adjusting R^ to equilibrium, the sum of the resistance in both 
rheostats should be kept equal to their original values, preferably 
not less than 10,000 ohms. 

The most convenient arrangement is to employ a " potenti- 



ometer " or slide coil bridge. The connections with the " five 
arc potentiometer " are shown in Fig 4. 




Fig. 4. 

This potentiometer is equivalent to a bridge wire of 100,000 
ohms resistance, that may be read to one place in 100,000. 

In the following experiments, a resistance of 20,000 ohms was 
employed in the potentiometer, and the resistance was read to one 
place in 10,000. 


= 1.433 volts, 


Sliding Potentiometer Experiments. 

^j = 6614 ohms (found by experiment), 
Ey^ = 1.426 volts at 67* F., 
R^ =r 6648 ohms (found by experiment), 
-f 2 = -^ volts. 

6614 ohms : 6648 ohms : : 1.426 volts : x volts, 

6648 X 1.426 

.-. £j= 1-433 volts. 

^j = 7146 ohms (found by experiment), 

£j = 1.426 volts (given), 

i?2 = 7200 ohms (found by experiment), 

-£j = ;r volts. 

R R "E E 

7146 : 7200 :: 1.426 : x, 

7200 X 1.426 

.*. £j= 1.436 volts. 

= 1.436 volts. 


(3) Clark Cell. R^ = 7200 ohms. 

R ' R E ' E 

7200 : 72CX) : : 1.436 : x, 

7200 X 1.436 

= 1.436 volts. 

7200 ^^ 

.-. -£3= 1.436 volts. 

Comparison of a new Clark cell with a standard Clark cell. 
Standard cell, R^ = 7560 ohms (found by experiment). 

Ey^ = 1.426 legal volts at 67"^ F. 
New Clark cell, ^2~ 7-9H ohms (found by experiment). 

7560 : 7914 : : 1.426 : x, 

7914 X 1.426 
2 = 1.492 volts. 

New Clark cell, -£2= ^-492 volts. 

May 6, 1897. 

Testing Clark cells by the high resistance method. 

Standard Clark cell = 203 = dy 

E, 1.4^^ 

1° = -7 = ** constants " = — ^^ volts at 70° F. = .00705, 
d^ 203 ' ' ^ 

New Clark cell = 204 = d^, 

E^ = '* constant " x rfj "= -00705 X 204 = 1.440 volts. 

March ii, 1898. 

Testing Clark cells with the sliding potentiometer. 
Standard Clark cell of the laboratory. 

-^1 = S309 ohms (found by experiment). 

El = 1.433 volts (given). 

Clark cell No. i made by the writer. 

R^ = 5339 ohms (found by experiment). 

R ' R " E ' E 

5309: 5339:: 1-433 :•*■, 

13_39.x_ 1-433 ^ J y^j^g ^ E.M.F. Clark cell No. i. 



March 18, 1898. 
E.M.F. of the ''Mesco Dry BatUry*' cells, No. i, No, 2, No. j. 
Standard Clark cell of the Laboratory. 

R^ = 5269 ohms (found by experiment). 
A= ^-433 volts (given). 

Mesco Cell No. i. 
-^2 = 5510 ohms (found by experiment). 

5269:5510:: 1.433 '^ 

^^^^"V""^^^ = 1.498 volts = E.M.F. Mesco Cell No. i. 

Mesco Cell No. 2. 
/?2 = 5149 ohms (found by experiment). 

5269: 5149 •• 1-533 --^i 
5149 + 1.433 
— ~6 " ~ I 4003 volts = E.M.F. of Mesco cell No. 2. 

Mesco Cell No. 3. 
R2 = 5129 ohms (found by experiment). 

R ' R "E ' E 

5269:5129:: 1.433 ••*'» 

5120 X 1.433 

- — ^ — > = 1-394 volts = E.M.F. of Mesco cell No. 3. 

March 25, 1898. 

E.M.F. of the ''Mesco Dry Battery " cells No. /, 
No. 2, No, J {continued). 

Standard Clark cell of the laboratory. 

^j = 5229 ohms (found by experiment), 
^1= 1-433 volts (given). 

Mesco Cell No. i. 
R^ = 5450 ohms (found by experiment). 

R 'R "E -E 

5229: 5450:: 1.433:^, 

— ' — = 1.493 volts = E.M.F. of Mesco cell No. i. 

5229 ^^^ 

VOL. XXII. — 25. 


Mesco Cell No. 2. 
R^ = 5080 ohms (found by experiment). 

R ' R " E ' E 

1 * 2 * * 1 • 2» 
5229:5080:: 1.433 -^f 

—- = 1.392 volts = E.M.F. of Mesco cell No. 2. 

5229 ^^ 

Mesco Cell No. 3. 
R^ = 5060 ohms (found by experiment). 

R 'R "E 'E 

5229:5060:: 1.433 '^f 

50^_^_L43_3 ^ J gg ^^j^g ^ E.M.F. of Mesco cell No. 3. 
5229 ^ ^ 

April i, 1898. 
EM.E. 0/'* Mesco Dry Battay " cells No. /, No, 2, No, j. 
Standard Clark cell of laboratory. 

R^ = 5248 ohms (found by experiment — "short circuited"), 
E^ == 1.433 volts (given). 

Mesco Cell No. i. 
^2= 5455 ohms (found by experiment — "short circuited"). 

5248: 5455:: 1.433:^, 

5455 X 1-433 ^ ^^g^ ^^j^^ ^ E.M.F. of Mesco cell No. i. 

Mesco Cell No. 2. 
R^— 5086 ohms (found by experiment — "short circuited"). 

R ' R " E • E 

I * 2 • • I • 2» 
5248:5086:: 1.433 ••^• 

5^_ -^-l'"^^^ = 1.388 volts = E.M.F. of Mesco cell No. 2. 

Mesco Cell No. 3. 
^2= 5073 ohms (found by experiment — "short circuited"). 

R 'R " E ' E 

I 2 • • 1 * 2> 

5248:5073:: 1.433 :-f. 
-5°7_3_x^-433 ^ j^gj y^jjg ^ E.M.F. of Mesco cell No. 3. 


April 15, 1898. 

Mesco Cell No. 2. 

Effect of*' short circuiting." 

R^ s 5076 ohms (found by experiment — ^before " short circuiting "). 

It was then " short circuited " through 5 ohms, for ten minutes. 

^2 = 4858 ohms (found by experiment — after *' short circuiting '*). 

R ' R "E 'E 

XV. . «Va • . X-| • ■'"'At 
5248:4858:: 1.433:^, 

4858 X 1.433 

— 5 — ^ =c 1.324 volts = E.M.F. of Mesco cell No. 2. 

5248 ^ ^ 

after " short circuiting." 

Mesco Cell No. i. 
R^ = 5452 ohms (found by experiment). 
Standard Clark cell. 

^j = 5275 ohms (found by experiment). 

5275:5452:: 1.433:^, 

5452 2< 1.433 ^ ^ ^^j^^ ^ E.M.F. of Mesco cell No. i. 

April 29, 1898. 
Standard Clark cell. 

/?j = 5288 ohms (found by experiment), 
A= 1-433 volts (given). 

Mesco Cell No. i. 
R^ = 5456 ohms (found by experiment). 

R 'R "E ' E 

5288:5456:: 1.433 ^-n 

5456_x^ i.433_ ^ j^^g ^^j^^ ^ E.M.F. of Mesco cell No. i. 

Mesco Cell No. 2. 
R^ = 5075 ohms (found by experiment). 

R * R - E ' E 

5288:5075:: 1.433 '-^l 

^"^^^00''"^^= 1.37s volts = E.M.F. of Mesco cell No. 2. 


Mesco Cell No. 3. 
^j= 5091 ohms (found by experiment). 

R 'R " E 'E 

12** 1 * V 

5288:5091 :: 1.433-^, 
5001 X 1.433 
7^88^ ' ~ ^-379 v^l^s = E.M.F. of Mesco cell No. 3. 

May 6, 1898. 

Mesco cell No. i was completely ••short circuited" for 5 

R^ =s 5039 ohms (found by experiment). 

Standard Clark cell. 

R^ = 5369 ohms (found by experiment), 

E^ = 1.433 volts (given). 

R • R " E 'E 

12** I * 2» 

5369: 5039 •• 1-433 ••^, 

-— — >-- — = 1.344 volts = E.M.F. of Mesco cell No. i. 
5369 ^ 

(After a complete " short circuit " of 5 minutes.) 

Mesco Cell No. 2. 
After ** short circuiting" for 15 minutes through 5 ohms. 
/?2 = 4870 ohms (found by experiment). 

R • F " E ' E 

5369:4870:: 1.433:-^, 

4870 X 1.433 

> = 1. 299 volts = E.M.F. of Mesco cell No. 2. 

5369 ^^ 

(After ** short circuiting " for 15 minutes through 5 ohms.) 

May 13, 1898. 
Standard Clark cell of laboratory. 

R^ = 53^9 ohms (found by experiment). 
i?j= 1.433 volts (given). 

Mesco Cell No. i. 
^2 = 5319 ohms (found by experiment). 

5319: 5319:: 1.433 :;r, 

5319 X 1.433 ^ J y^ug ^ E.M.F. of Mesco cell No. i. 



Mesco Cell No. 2. 
R^ = 5070 ohms (found by experiment). 

R^:R^\:E^: E^, 

5319: 5070:: 1.433 '-^^ 

5070 X 1 .433 ^ J g ^^f^ ^ E.M.F. of Mesco cell No. 2. 

Mesco Cell No. 3. 
^3 = 5 1 1 8 ohms (found by experiment). 


S 1 18 X 1.433 ^ J g ^^j^g ^ E.M.F. of Mesco cell No. 3. 

March 10, 1899. 
Standard Clark cell of laboratory, 
R^ = 5043 ohms (found by experiment), 
E^ = 1.433 volts (given). 

Mesco Cell No. i. 
^2 = 5115 ohms (found by experiment). 

R ' R " E ' E 
5043 :511s:: 1.433 :;r, 

5 1 15 X 1.433 ^ J ^^j^ ^ E.M.F. of Mesco cell No. i. 

S043 ^^ 

Mesco Cell No. 2. 

R^ = 489 s ohms (found by experiment). 

R ' R " E 'E 
1 • 2 • • 1 • 2 

5043 :4895 :: 1.433:-^ 

4895 X 1.433 ^ ^ ^^j^g _ E.M.F. of Mesco cell No. 2. 

5043 ^^ 

Mesco Cell No. 3. 

/?2 = 4916 ohms (found by experiment). 

R 'R "E E 

1 • 2 * * 1 • 2» 

5043:4916:: 1.433 :;r, 

4916 X 1-433 ^ J g y^jjg ^ E.M.F. of Mesco cell No. 3. 



March 17, 1899. 

Efficiency of a Cell, 

Polarization and Recovery, 





Fjg. 5. 

Read the voltmeter with key d open ; this gives the E.M.F. of 
the cell E. 

Then close d and read the voltmeter ; this gives the potential 
difference (P.D.) across R, EJ 

Then E-- E '\s the potential difference (P.D.) across the cell E", 

The internal resistance, ^', may then be found by the propor- 
tion E'\E"\\R\R'. 

For polarization keep b closed for two minutes, then open b for 
an instant, and read the voltmeter; this gives E, 

A reading should also be taken with b closed for potential 
difference (P.D.) across R^ (£'), and the internal resistance again 

calculated from the above formula. 

The current may be obtained from the formula, C= „. 

Continue these readings every 2 minutes for 10 minutes. 
For recovery of E.M.F. keep b open, and read the voltmeter 
every 2 minutes, for 10 minutes. 

Voltmeter Reading in -i^ Volt. 


R=i2 ohms. 



E E' 


' 41. 29 

2 35- 
4 34. 
6 1 34. 
« , 34. 1 
10 ' 33-5 

Mesco Cell, No. I. 










' = B\ 

41 — 29= 12. 

E \E'\\R\R', 

29 : 1 2 : : 2 : ;r, 


- = .827 ohms = /?', internal resistance. 
29 ' 

April 7, 1899. 

Mesco Cell No. i. 

Voltmeter reading in ^ volt. 

DErrERMiNATiONS. R = o ohms. 


Time. , ^ 

1 E. 1 E'. 


2 35. 
4 34.5 
6 34. 
8 33.5 
10 ' 33- 




15.5 X o 


E^B = E\ 

42.5-27= 15.5. 


27 : 15.5 ::o :;r, 

= .574 ohms = internal resistance of Mesco cell No. I 

April 28, 1899. 

After two weeks " short circuit *' through 20 ohms. 
Voltmeter reading in -^^ volt = |^ volt. 

R = 2 ohms. 


24-9= 15. 

E :E'::R:R\ 

9 : 15:: 2.0 : X, 

15 X 2.0 

- = 3.33 ohms. 


Mesco Cell No. 2. 

Voltmeter readine: == — volts = E. 


^ = 5 ohms, drops to ^^ volts = E , 

E--E ^B\ 

40—32 = 8. 


32 : 8 :: 5 :jr. 

=B I.2S ohms = internal resistance of Mesco cell No. 2. 

32 ^ 

This cell was then left on " short circuit " for one week, through 
272 ohms. 

At the end of this time the following readings were made on 
this cell: 

Voltmeter reading: = - volts, 

^ 30 

^ = 5 ohms. 

36.3 27 

Voltmeter reading drops from - volt = i? to volt = £'. 

^ ^ 30 30 

^- £'=£'', 

36 — 27 = 9. 


27:9:: Six, 


= 1.7 ohms = internal resistance of Mesco cell No. 2. 

27 ^ 

Mesco Cell No. 3. 

Voltmeter reading = volts = E, 

^ 30 

/? = 5 ohms, drops to volts = E\ 

E^E ^E\ 

40.5-34 = 6.5. 


34 :6.S : : 5 ohms \x, 
6.5 X 5 


= .955 ohms = internal resistance of Mesco cell No. 3. 



Tabulated results of tests made on the Me.sco dry battery cells. 

Mesco Cell, No. i. 


E. M. F. 

1.498 volts. 

Condition of Experiment. 

March i8, 1898. , 

Sliding potentiometer method. 

** 25, ** 

1-493 ** 

April I, ** 

1.489 (short circuited). 

short circuit. 

*• 15. " i 

1. 481 volts 

** 29, •* 

1.478 *• 

May 6, ** 

1.344 *' 

complete short circuit for 5 minutes. 

13. ** 

X.433 ** 

March 10, 1899. 

1.433 " 

Mesco cell, No. I, was set aside from May 13, 1898, until March 10, 1899. 

Mesco Cell No. i. — Continued, 

March 17, 1899. 

April 7, 1899. 

April 28, 1899. 

May 5, 1899. 

Internal Resistance. 

.827 ohms - R^, 

Condition op Experiment. 

.574 ohms = i?''. 

3.33 ohms ^ i?'. 

3.84 ohmsz= iV''. 


2 ohms. 


= volts. 



^9 volt. 



— volt. 





— ohms. 


^'•S volts. 





•5-S volt. 


E' '.Ef'','.R'.R'. 

After short circuiting for two weeks, 
through 20 ohms. 

Voltmeter reading in volt showed 

^ 30 

-- = .77 volt. 
After one week's recovery. 

Voltmeter reading =1 ^'*' ^ — 1. 14 volts. 


i? =r 5 ohms. 




March 1 8, 1898. 

** 25, " 
April /, *♦ 

'5, •* 



May 6, 


Mesco Cell, No. 2. 

E. M. F. 

Condition op Experimbnt. 

1.400 volts 

Sliding potentiometer method. 

1.392 ** 

i.3«8 *' 

Short circuited. 

1.324 " 

1 Short circuited for lo minutes through 

5 ohms. 

1-375 ** 

' After two weeks' recovery. 

1.299 *« 

1 After short circuiting for 15 minutes 

through 5 ohms. 

r.365 ** 

After one weeks' recovery. 

1.390 »• 

Sliding potentiometer metho<l. 


March lo, 1899. 

Mesco Cell, No. 2, was set aside from May 13, 1898, imtil March 10, 1899. 

April 28, 1899. 

May 5, 1899. 

Mesco Cell, No. 2. — Continued, 

Internal Resistance. 
1.25 ohms = R' . 

1.70 ohms -- R' . 

Condition of Experiment. 



Voltmeter reading 

E _-:40. 

£':£'':: R:R'. 
32 : 8 :: 5 : jc. 

After ** short circuiting" for one 
week, through 272 ohms. 


March 18, 1898. 

** 25, *♦ 
April I, " 

*' 29, " 
May I 3, ♦* 
March 10, 1^99. 
May 5, 


Mesco Cell No. 3. 

E. M. F. 

1.394 volts. 



< ( 



Condition of Experiment. 
Sliding potentiometer method. 




Short circuited. 

Sliding potentiometer method. 

t 4 



Short circuited through 5 ohms. 
Internal resistance = .955 ohms. 

Mesco cell No. 3 was set aside from May 13, 1898, until March 10, 1899. 


That the E.M.F. is constant to 1% or better, under all usual 
conditions, and this type of cell need only be standardized at long 

That time has but little effect on E.M.F. 


Even when polarized, the cells recover, almost completely, their 
original E.M.F. 

The author desires to express his sincere thanks to Mr. H. C. 
Parker, for his faithful supervision of the experiments described in 
this article. 


By H. M. HOWE. 

The removal of arsenic during the second hour of Mr. Scherr's 
roast, while probably due in large part to the volatilization of sul- 
phides of arsenic, and to the reduction of the non-volatile arsen- 
iates to the volatile arsenious acid by the excess of unoxidized 
copper, iron, and sulphur still present, as Mr. Scherr points out, 
may be due in part also to the decomposition of arseniates by the 
sulphuric anhydride set free by the decomposition of iron sul- 
phates ; and we see on looking back that the investigation might 
have been made more instructive by determining the ferrous and 
ferric sulphate in the several samples. 

Mr. Scherr explains the fact that arsenic, in spite of its rapid 
removal during the second hour, is but very slowly removed there- 
after, by holding that this removal, once the arsenic has formed 
arseniate, can be brought about only by the joint action of sul- 
phuric anhydride to decompose the arseniate and of a reducing 
agent, which, while like sulphur, iron, and copper, it is strong 
enough to reduce the fixed arseniates to the volatile arsenious 
acid, must not, like carbon, be strong enough to reduce the sul- 
phuric anhydride itself, and thereby prevent that anhydride from 
decomposing the arseniates. Thus, in his view, during the second 
hour the combined action of the nascent sulphuric anhydride as a 
decomposer, with the unoxidized sulphur, iron, and copper as re- 
ducing agents, causes arsenic to be expelled; whereas the reason 
why the joint action of sulphuric anhydride and charcoal does not 
cause arsenic to be expelled in the sixth hour is that the charcoal 
is so powerful a reducing agent as to destroy the sulphuric anhy- 


There is another possible explanation, viz., that at the end of the 
second hour nearly the whole of the superficial op accessible ar- 
senic, /. e., the arsenic in the exterior of the individual particles, 
has already been removed, and that what is left is .stubbornly held 
in the interior of those particles, where neither sulphuric anhydride 
to decompose nor charcoal to reduce could get access to it ; in 
short, that the reason why sulphuric anhydride and charcoal jointly 
are unable to expel arsenic during the later hours is simply that 
they are mechanically prevented from reaching it. We might test 
these two suppositions by adding, say at the fifth hour, in one pan 
charcoal and in the other pan pyrite. In this latter pan we should 
have the conditions which, according to Mr. Scherr, led to the ck- 
pulsion of arsenic in the second hour ; and if his explanation be 
correc^we should find arsenic expelled in this pan, but not in the 
one to which charcoal had been added : whereas if the reason why 
arsenic was not expelled in the latter part of the present roast was 
simply that the residual arsenic was so deep-seated as to be me- 
chanically protected from the sulphuric anhydride and charcoal, 
this same protection should prevent its removal by the joint action 
of sulphuric anhydride and the sulphur and iron of the pyrite, and 
arsenic should not be expelled in either pan. 

Another test would be to grind the matte before adding the 
charcoal, and thus destroy this supposed mechanical protection. 

Following Plattner's assertion that ferric arseniate is readily de- 
composed, it might be well to study the temperature and other 
conditions most favorable to its decomposition, and then repeat 
Mr. Scherr's experiments under these favorable conditions. 



Died: i March, 1901, 

Charles Hill Bergen was born at Red Bank, New Jersey, July 
30, 1 870, only son of John Bowne Bergen and Amanda Hill Bergen. 

He attended the Red Bank graded school, graduating class of 
'85, and the Lawrenceville School, class of '88. 


He entered Columbia University and was graduated a civil en- 
gineer from the School of Mines, class of '92. In his junior year 
he was historian of his class. 

He was a member of the Phi Gamma Delta fraternity, Omega 

In 1892 he entered the employ of the Brooklyn Heights R. R. 
Co.; becoming Building Superintendent and Assistant Engineer 
he had charge of the operation of the Montague Street Cable Rail- 
way, and the designing and construction of that company's build- 
ings. In 1899 he became Chief Engineer of the Wilson & Bailey 
Co., Brooklyn, and in 1900 Secretary and Treasurer of the Man- 
hattan Paint Co., New York. In I9CX> he joined the A. D. Granger 
Co., New York, with whom he was connected at the time of his 

In May, 1895, he married Henrietta M. French, who died in 
January, 1901, leaving 'one child, Portia F. Bergen. 

'92 Reunions Committee, No. 32 Nassau Street, New York 

At a Meeting of the Alumni of Columbia University, Class of 
'92, Arts and Mines, held on Monday, March fourth, 1901, the 
following action was taken : 

'* Whereas, Our beloved friend and classmate, Charles Hill 
Bergen, was taken from us on Friday, March first, 1901 ; there- 
fore be it 

** Resolved^ That the Alumni of Columbia University class of 
'92, Arts and Mines, extend to the family their sympathy for the 
loss they have sustained ; and furthermore be it 

" Resolved, That Messrs. Chrystie, Granger and Merz be ap- 
pointed a Committee to express and send on behalf of the class 
an appropriate message of condolence to his family, and to cause 
to be inserted in the University publications a suitable sketch of 
his life." 

A True Copy : 

Arthur T. Hewlett, 






President McKinley has just appointed Professor Chandler as a member 
of the Board of Visitors to the U. S. Naval Observatory. The Board is 
organized under the provisions of a recent act of Congress calling for "a 
board of six visitors 10 the Naval Observatory, four to be astronomers of 
high professional standing and two to be eminent citizens of the United 
Slates.** The astronomical members of the Board are Professors Picker- 
ing (Harvard), Young (Princeton), Hall (University of Michigan) and 
Stone (University of Virginia) ; while President Harper of Chicago 
shares with Professor Chandler the honor of being an * eminent citizen ' 

An optional one-hour course, Chem. 10, has been announced for next 
year, to be given by Mr Whitaker, upon Theoretical Inorganic Chem- 
istry, including the rare elements This course is open to all students of 
the University who have attended the courses, Chem. i, and Chem. 7. 

Mr. Whitaker has also been appointed to represent the University 
Committee on Entrance Examinations, in the subject of Chemistry, at 
this year's entrance examinations. 

At the last meeting of the Trustees of Columbia University, the very 
unusual, indeed the unique, step was taken, of appointing three officers 
of the Chemical Department to the grade of adjunct professor. Two of 
these appointments. Dr. E. H. Miller and Mr. M. T. Bogert, are old 
School of Mines men. Dr. Miller graduated in the chemical course, 
as Ph.B., in 1891, taking the degree of A.M. in 1892, and that of Ph.D. 
two years later. Ever since his last year in College he has been serving 
the University, first as assistant in assaying, later as tutor, and from 1897 
until now, as instructor in quantitative analysis an'l assaying. For the 
last two years he has been in charge of the quantitative and assay labora- 
tories, taking the place of Prof. P. deP. Ricketts, and has also been in 
charge of the apparatus room. 

Mr. Bogert graduated from Columbia College in 1890, and from the 
chemical course as Ph.B. in 1894, and the same year entered the 
organic laboratory as assistant to Professor Colby. In 1897 he was 
made first tutor, and then instructor in organic chemistry, and for the 
last fis^ years he has had full cha-ge of the laboratory and lecture in- 
struction in that subject, succeeding the late Professor Colby. He has 
recently been appointed to take charge of the apparatus room for the 
next two years. 

Mr. Bogert has been elected vice-chairman of the New York Section 
of the American Chemical Society, and a member of the executive com- 
mittee of the New York Section of the Society of Chemical Industry. 
He has been, for some years, the secretary of the Chemists* Club of this 

As will be noticed, both Professors Miller and Bogert succeed to 
chairs which existed in former years, and have been allowed to lapse 
temporarily. Their promotion is a source of great satisfaction to all their 


colleagues because, during their long service in the University, they have 
worked not only with industry and ability, but also with very marked 
success, to build up their respective laboratories. The very striking ad- 
vances in the numbers of both regular and postgraduate students, in the 
courses of instruction, and, above all, in the amount and quality of the 
work done, in the analytical and organic laboratories, have been due in 
very great measure to their able and untiring efforts. 

Dr. J. L. R. Morgan is a more recent addition to the Department, 
having been appointed to the position of tutor in chemical physics and 
chemical philosophy in 1897. He graduated at Rutgers as B.Sc., in 
189 a and studied for the next three years in Leipzig, where he received 
his Ph.D. degree. Dr. Morgan has been very successful in building up 
the new courses in modern theoretical chemistry, and the establishment 
of this new chair places Columbia in line with Cornell, Johns Hopkins 
and Wisconsin, where special attention is devoted to this branch of the 

Professor Morgan has just been made a member of the Board of Re- 
viewers of the Journal of Physical Chemistry, and M. Rossetof Paris has 
asked to be allowed to translate into French one of his books, ''An out- 
' line of the Theory of Solution and its Results " (N. Y., John Wiley & 
Sons, 1897). 


The extraordinary increase of students in all lines of analytical chem- 
istry has filled both the main laboratories as well as the special rooms to 
their full capacity during the present year. The incorporation of gas an- 
alysis in the various courses in quantitative analysis, has extended this 
line of work to the students in electrical and mining engineering, as well 
as improved the facilities for the chemists by transferring the apparatus 
to the clectrol>tic room where the temperature is practically uniform. 
As about seventy-five students do gas analysis in the course of the year, 
there is greatly needed a special room which can be devoted to this purpose. 

Drs. Sherman and Joliet have been made tutors in analytical chem- 
istry instead of lecturers; this places the work in quantitative analysis 
on a more permanent basis. 

There has recently been added to the quantitative laboratory an At- 
water-Blakeslee calorimeter which will prove useful in research work, both 
on foods and fuels, and also in continuing the work on •* Heat of Com- 
bustion as a Factor in the Analytical Examination of Oils," on which 
some results have been recently published by Dr. Sherman in the Journal 
of the American Chemical Society. 

Dr. Tingle's work on the synthesis of amines by use of alkyl salicy- 
lates, has turned out successfully, and the results appeared in the Ameri- 
can Chemical Journal for February. He is now at work on the reactions 
between platinum salts and ferrocyanides. 

Besides the thesis work already mentioned, the following subjects are 
under investigation : 

"Comparison of Methods for the Determination of Mercury/* G. N. 

"The Use of Test Lead as a Reducing Agent for Iron, in Zinciferous 
Ores," E. I. Shepard. 


''The Estimation of Cadmium/* R. W. Page. 
** The Estimation of Sulphur in Pig Iron," W. L. Lindsay. 
**The Estimation of Manganese in Spiegel," T. F. Hildreth. 
**The Conversion of Lead Sulphate to Barium Sulphate in Lead 
Slag," G. C. Hubbard. 

"The Composition of the Ferrocyanides of Cadmium," E. H. Miller. 


Omx\% to a change of scheme the regular students in the School of 
Chemistry did not attend in this laboratory during the first term, but in 
spite of this, the four courses. Chemistry 26, 27, 28 and 29 were all taken 
by |X)stgraduates and special students, nine in all. This term the lab- 
oratories have been crowded with both regular and special students to the 
number of thirty, and it has been difficult to find room for them to work 

Speciil attention has been given to the course by Mr. Tucker on elec- 
tro-chemistry, a subject which has proved of great interest to the students, 
and which is of very great importance, nowadays, both in applied chem- 
istry an<l in metallurgy. 

The course in dyeing and calico printing has also been attended by 
largely increased numbers of students, both regular and special, many of 
the latter being personally interested in textile industries in various parts 
of the country. 

Mr. Tucker, ia conjunction with Mr. H. R. Moody, candidate for the 
degree of Ph.D., has conducted a research throughout the year with the 
electric furnace, studying some new reactions which are brought about at 
the temperature of the electric arc. It was first necessary to devise a suit- 
able form of electric furnace, the one finally adopted being somewhat 
similar to that of Moissan, but made entirely of carbon. The work has 
included the following : 

The production of mixed carbides with a view to the production of 
ethylene on their decomposition with water. 

The reduction of metallic aluminium from its oxide by means of cal- 
cium carbide. 

The analysis of gaseous mixtures containing acetylene, ethylene, meth- 
ane and hydrogen. 

The formation of nitrides. 

The formation of phosphides. 

The formation of borides. 

Mr. Meer, E.E., 1901, is engaged on'the study of the formation of 
sodium nitrite from nitrate by electrolysis. 

Mr. C. W. Stoddart, candidate for the degree of M.A., has undertaken 
a research into the formation of permanganic acid by electrolysis. 

Mr. O. W. Palmenberg, Chemistry i9ot, has taken for his thesis sub- 
ject the transformation of acetylene into ethylene, and, at the same time, 
Mr. A. W. LePrince, of the same class, has taken for the subject of his 
thesis, the formation of ethyl sulphuric acid, with a view to obtaining al- 
cohol. It is hoped that these experiments will throw some new light on 
the possible production of alcohol from calcium carbide. 

Mr. A. B. John, special student, is engaged in the fractional distilla- 
tion of Mexican petroleum. 



The number of students, undergraduate and graduate, taking the three 
courses offered in the laboratory of Physical Chemistry has been so great 
that the lack of working space has been severely felt. The laboratory 
has already been enlarged once this year by the absorption of another 
large room* but extra accommodation will soon be necessary for research 

Of the investigations now in progress the most important is that which 
is being carried on by Mr. F. M. Becket (M.A. 1898) on the electrical 
conductivity of fused electrolytes. For this work a delicate pyrometer 
is necessary, and the one being used is so satisfactory, that it merits a 
short description. It is of the portable type, as made by the Siemens- 
Halske Company, and consists of a suspension voltmeter reading to one 
hundred millivolts, each division corresponding to one-tenth of a milli- 
volt (0.000 1 volts) and a sixty.inch platinum-rhodium junction. Both 
of these portions are guaranteed accurate by the Reichsanstalt of Berlin, 
the temperatures being read directly up to i,ooo°C. from a special scale 
on the voltmeter, one scale division corresponding to 5° C, ard being 
so large that it may be subdivided, with practice, into fifths, giving esti- 
mated temperatures to a single degree centigrade. This type of instru- 
ment is so much more easily read and adjusted than the ordinary Le 
Chatelier form with reflecting galvanometer, which necessitates empiri- 
cal calibration and temperatute reading by interpolation, that even if it 
were not quite as delicate, it would be ideal for the present purpose. 


Several changes have been made in the undergraduate courses in organic 
chemistry. Hereafter the subject of the lecturefs to third-year chemists 
will be the chemistry of methane and its derivatives (Chemistry 31), the 
fourth-year course being devoted to the carbocyclic and heterocyclic 
compounds (Chemistry 32). The general laboratory course (Chemistry 
^^3) has been set off by itself, to permit of its being taken by candidates 
for higher degrees, who have already a sufficient grounding in the theo- 
retical side of the subject. The elementary lecture course (Chemistry 20) 
has been remodelled in order to adapt it to the needs of non-professional 
students in the College, the School of Medicine, and so forth. In the 
future this course will be more popular and less technical in character, the 
lectures being liberally illustrated wiih material from the chemical mu- 
seum, and by experimental demonstration. The recitation* hour has been 
dropped, making it a two-hour course. 

At the November meeting of the New York Section of the American 
Chemical Society, Mr. D. C. Eccles, a graduate student in organic chem- 
istry, read a paper on antipyrine and its derivatives, illustrating his dis- 
cussion with numerous charts and diagrams intended to show the im- 
portance of this substance, both chemically and medicinally, and the 
great number of compounds which have been prepared from it. The 
paper is given in full in this number of the Quarterly. 

Dr. Gotthelf is continuing his investigations on the ketodihydroqui- 
nazolines, and read a paper upon this subject before the April mee 
the New York Sect on of the American Chemical Society. 

VOL. XXII.— 26. 


Dr. Caspari has contributed to the February number of the American 
Chemical Journal articles upon the new radio-active substances, and upon 
the higher superoxides of hydrogen. 

Through the kindness of a friend ihe organic laboratory has received 
twenty-five pounds of crude limonene, which will prove most interesting 
material for researches upon the chemistry of the terpenes. 


The pei-sonnel of the Department has been changed since last Autumn 
only by the lamented death of Mr. L. Le Count, who at that time was 
assistant in civil engineering. That vacancy was filled by Mr. Myron 
S. Falk, C.E., a graduate of the class of 1899. 

The general work of the Department has been progressing along its 
usual linei with such advances as another year's experiences and oppor- 
tunities make practicable. At the present time marked advances are 
being made particularly in the field of railroad work. The fourth year 
students have just completed a very full and thorough treatment of the 
economic theory of railroad location and are beginning lectures under 
Professor Lovell relating to signaling and interlocking. Both of these por- 
tions of railroad work, and others connected with the same general subject, 
constitute lines along which the specialty of railroad engineering is being 
developed as rapidly as possible. The same general observation can be 
made in reference to the work in roads and pavements, masonry struc- 
tures and the entire field of iron and steel structural work and founda- 
tions. The hydraulic laboratory of the Department of Civil Engineering 
is still in an unfinished state, owing to certain complications which fol- 
lowed the transference of the H. R. Worthington Company's business to 
its new owners. Arrange ments are in progress now, however, which it 
is hoped will lead to an early completion of the laboratory. 

The members of the fourth-year class are all actively engaged in the 
preparation of thesis work. A c(»nsiderable amount of original investiga- 
tion is now in progress in various fields, which will result not only in 
most useful experiences for the students who are performing the work, 
but will lead to valuable practical data in two or three lines of pro- 
fessional practice. In these investigations the laboratories of the Depart- 
ment of Civil Engineering and of the Department of Physics are exten- 
sively used. Work of this class is being actively encouraged in the 
Department and forms a most effective supplement to the large amount of 
engineering design done by the students in civil engineering throughout 
their course of study. 

llie departmental library has been enlarged by the books formerly be- 
longing to Mr. Le Count, whose father has very kindly and thoughtfully 
made a gift of them to the Department in memory of his son. 


The students of the th rd and fourth years have completed the six 
weeks of afternoors devoted to historical research, and the results of 
their labors, which will be shown in the annual exhibition in June, are 
interesting for their quaMty quite as much as for their quantity. In all 
over five hundred sheets of drawings have been handed in, each week's 
work of each student being devoted to a specified topic of the history of 


mediaeval architecture, and accompanied by a written essay on the re- 
sults of the week's studies. The Avery Library and the Department 
Library, with its great collection of prints and photographs, have fur- 
nished the documents for these studies. Not a few of the drawings pro- 
duced are of permanent and practical value either as illustrative material 
for future lectures, or as drawings for publication in future treatises. This 
work has been under the general direction of Mr. Kress and Mr. Partridge. 

The sixth competition for the McKim Fellowships has recently closed. 
The preliminary sketches by competitors in and near New York were 
made at the Department on Saturday, March 9th, and the final drawings 
were handed in April 27th. The subject is *' A Presbyterian Church 
with a Dome,*' and the design is required to be in the style of the Italian 
Renaissance. Nineteen competitors were present at the school for the 
preliminary sketch, and as there are eleven Columbia graduates in Paris 
who have also registered as competitors, the number of designs handed 
in promises to be the largest thus far recorded in any fellowship competi- 

The subjects of the senior theses show the usual variety. They com- 
prise a Country Club (two theses), U. S. Post Office, Museum of Arts 
and Sciences, Cemetery with Crematory and Chapel, School of Architec- 
ture, Gothic Mausoleum, Hippodrome, University Hall, City Hall for 
Greater New York, County Court House and City Hotel. 

Mr. Hombostel, lecturer on architecture in this school, has been giving 
to a class in the Art Students' League a course of lectures on architec- 
ture and practical perspective. Profe sor Hamlin delivered before the 
Architectural League on Thursday, April nth, the address on '< Italian 
Gardens " which he gave before the Institute of Ar. hitects at their con 
vention in Washington last December. 

Three of the lectures in the February course at the American Museum 
of Natural History were given by officers of this department : the first on 
"Trees in the Gardens of the Renaissance," by Professor Hamlin; the 
second and third on the '' Life and Characteristics of Trees and Forests," 
by Mr. C. P. Warren. 


A Text Book of Important Minerais and Rocks ^ with tables for the de- 
termination of minerals. By S. R. Tillman, Professor of Chemistry, 
Mineralogy and Geology, U. S. Military Academy, West Point N. Y. 
New York, John Wiley & Sons. 8vo. Pp. 186; 35 figures. Cloth, 
I2.00, net. 

This book is planned to meet the conditions prevailing at the Military 
Academy at West Point where the students are " well fitted for the work 
when they begin it, ' and have *' excellent opportunity for the examina- 
tion and comparison of specimens and for laljoratory work in determin- 
ing them," and where the purpose is to impart a fair knowledge of the 
important minerals and r^cks. 


Seven pages are devoted to chemical and physical properties, and 
about twenty to crystallography. Some seventy-five species are de cribed 
briefly, in general in order of the basic constituent. The descriptions 
are followed by convenient tables for determination and these by a short 
chapter on the common rocks. 

As a guide for mature students desiring a brief course the book will 
be useful. For the general student too little space is given to descrip- 
tions of methods, tests and crystals ; and the list of species, which ex- 
cludes entire economic groups like the nickel minerals and the borates, 
and such con mon species as barite and epidote is too restricted. 

A. J. M. 

Kurzes Lehrbuch der Analytischen Chemie, Von Dr. F. P. Tread- 
well. Franz Deuticke, Lepsig und Wien, 1899. 426 Seite. 

The first volume of Professor Tread well's book on Analytical Chemistry 
is devoted entirely to qualitative analysis and is of such great merit as to 
warrant a special review although the book is not yet completed by the 
appeirance of the volume on quantitative analysis 

Qualitative analysis is treated not from the descriptive standpoint but 
from the point of view of modern inorganic chemistry. A brief but dis- 
tinct statement of the ionic theory, hydrolysis and solubility product 
precedes the main portion of the work devoted to the reactions of the 
bases and acids 

These reactions are most accurately expressed, those which are rever- 
sible are so designated and composition is indicated by a liberal use of 
graphic formulae. 

A most valuable feature of the book is the introduction of the reac- 
tions of the rare elements with schemes for their separation and ident- 

The book is no adap ed to the wants of a beginner in chemistry for 
use as a laboratory guide, but as a text and reference book for students 
who are learning at the same time inorganic chemistry and qualitative 
analysis it cannot be too highly commended. 

E. H. M. 

OJena Normal if%. 
biass. \» . ^v 

C. A. F. Kahlbaum's Chas- Schleicher & 

Chemlcalty Pure Organic 
Inorflanic ChBmlcals. 




ISlh St and 3rd Ava.. Ngw York. 

Chemicals S Chemical Apparatus 


SCHOTT & GENOSSEN, Manufacturers of the famous Jena Olisa 

C. A. F. KAHLBAUM, Chemicals. 


CARL ZEISS, Microscopes. 

FRANZ SCHMIDT & HAENSCH, Polariscopes and Spectroscopes. 

DF5M0UTIS & CO., Purest Hammered Platinum, 


GREINER & FRIEDRICH'S, Fine Stopcock Ware. 

MAX KOHL, Physical Apparatus. 

TENDER, Reagent Bottles with Black Indelible Letters. 

KAVALIER'S Famous Bohemian Glassware. 

The above finns are known to furnish the best goods in (heir retpeclive linei 
We are tfaeir representatives for the Unitetl States snd Canada and carry a consider 
able stock of these goods on hand. Kohl's Apparatus and Stender's Hottleaaie 
mpoiled to order only. 

Jena Glass the Glass of the Fntnre 

Neumann Bros. 

Book Binders 

6 and T EAST 16TH ST., 



Books bound in any style. 
Single volumes or in quantities 

Woodbridge School, 


School of Mines Preparatory School. 

417 Madison Avenue. 

Between 48th and 49th Streets, NEW YORK CITY 

Twentieth Y^ear Begins October ist, igoi. 

THE school is well equipped with physical and chemical laboratories, in which the 
students are required to perform a complete set of experiments illustrative of their 

recitations in physics and general chemistry, A special laboratory is devoted to 
qualitative chemical analysis for advanced students. 

Five hundred Students of Columbia School of Mines have been instructed in the 
Woodbridge School. Also a large number have been prepared for Massachusetts In- 
stitute of Technology, Stevens Institute, Sheffield Scientific School, Lawrence Sden- 
ific School, Troy Polytechnic Institute, Cornell University, and the Classical, Medical, 
and Law Departments of Harvard, Yale, Columbia and Princeton. 

A summer sch4X)l for students who have failed in June begins August 13th and 
coaches men for the Fall examinations. All classes are limited to five. CoUege men 
are coached in Freshman and Sophomore Mathematics and Quantitative Analysis. 

An advanced course for older students prepares them to enter the second -jrear class. 
Last June one of our students received his degree in Electrical Engineering in 
three years. 


J A^ i^~ 20 '11 

Vol. XXII. No. 4. - JULY, 1901. 


, , > I 

THE -, y 




» ♦ t 


A. J. MOSES, Prof, of Minermlogy. JOS. STRUTHERS, Ph.D., 

J. F. KEMP, Prof, of Geology. Lecturer in Metallurgy. 

R. P££LE, Adj. Prof. Mining. I. H. WOOLSON, Instructor in 

A. D. P. HAMLIN, Mechanical Engineering. 

Adj. Prof, of Architecture. W. H. FREEDMAN, E.E., 

R. E. MAYER, Instructor in Drawing. S. O. MILLER, Aeeiatant in 

B. WALLER, Analytical Chemist. Mechanical Engineering. 

Manaarlnff Editor, R. E. MAYER, 
Business Manager, S. O. MILLER, 


The Quantitative Determination of Cadmium. By Edmund H. Miller 

and Robert W. Page , 391 

The Serpentines of Manhattan Island and Vicinity and their Accom- 
panying Minerals. By D. H. Newland 399 

A Method of Cyclic Analysis of Heat Engines. By Charles £. Lucke, 

M.S 4x1 

A List of the Crystal Forms of Calcite with their Interfacial Angles. 

By Austin F. Rogers 429 

A Laboratory Classifier — Supplementary Note. By H. S. Munroe 449 

A Correction — Thomas Egleston Memorial 451 

Abstracts in Analytical Chemistry. By Elwyn Waller 45a 

Abstracts — Current Mining Decisions .^. 456 

Book Reviews 475 



Registered at the New York Post Office as Second Class Matter. ^ 

All Remittancea ehoald be made payable to Order of **The School of Minee Quarterly 



Tool and Die Steel of th« Beat Quality, Sheet Steel, 
Spring and Macbineiy Steel. Steel for Edge-Toola and 
Hardware of Every Deacription. Polished Drill Rods 
and Fine Wlr«. Steel Forgings, Coiled SpiinKSi 'tc. 

WeMcra OBce and Warcbonie. ' Baaliin Of&ce aod Warebosac 

64 and 00 So. OUaton StrMt, 480 Feacl SUvet. 


16,000,000 BARRELS 








Makes perrect joint immediaiely ; doei not have to 

be followed up. 

Mali esjranl that will tost for years on all pressures 

or steam 

[loes not rot, bum, blow or squeeze out. 

Weighs 30 per cent less than many other packings, 

therefore the cheapest and best. 

JENKINS BROS., New York, Boston, Philadelphia, Chicago 




Vol. XXII. JULY, 1901. No. 4. 

[Contribution from the Havemeyer Laboratories of Columbia University, No. 45.] 




The work described in this paper was suggested by the difficulties 
encountered in some experiments on the determination of cadmium, 
which were undertaken before analyzing a number of precipitates 
of cadmium ferrocyanide. The methods compared are the electro- 
lytic, the carbonate and the phosphate. The work was done on a 
solution of cadmium chloride, found to be free from impurities, 
which after thorough mixing was sealed up in a number of bottles 
to avoid evaporation. 

Electrolytic Determination. 

The cyanide * method was used under the following conditions : 
Thirty cubic centimeters of the cadmium chloride solution were 
diluted to one hundred and fifty cubic centimeters and one gram 
of potassium cyanide added, this amount being a slight excess 
over that required to redissolve the cadmium cyanide first formed. 
The solution was then electrolyzed by a current of 0.1-0.15 am- 
pere for sixteen hours. The kathode was a platinum cylinder having 
an area of 90 square centimeters, the anode a spiral of heavy 
platinum wire. The deposition was found to be complete in every 

* Riban, Analysis Chimique Quantitative par Electrolyse, Paris, 1899. Rimbach, 
Zeit f. Anal. Chem., Vol 37, p. 284, 1898. 

VOL. xxn.— 27. 391 



case, the electrolyte was colorless and the deposited metal bright 
and adherent. The results are given in the following table : 

Table I. 


taken cc. 


in p^ms. 


in houn. 

Bulk in cc. 
































in grams. 


Average weight of cadmium, coinciding with three determina- 
tions 0.2088 gram. 

It was found that a large excess of potassium cyanide was to be 
avoided and also the presence of other salts. When cadmium is 
separated as sulphide and then dissolved in hydrochloric acid the 
excess should be removed by boiling and not be neutralized by 
sodium carbonate, as the resulting sodium chloride interferes with 
the electrolytic precipitation. Three grams of potassium cyanide 
under the preceding conditions gave results which were lower and 
less uniform. 

The Carbonate Method. 

This method was used merely as a check on the electrolytic re- 
sults before proceeding to the phosphate determination. The 
cadmium chloride solution was precipitated by a slight excess of 
sodium carbonate boiled, and the precipitate washed till it gave no 
evidence of alkali when tested with phenol-phthalein ; dried, sepa- 
rated from the paper etc., ignited and weighed as cadmium oxide. 
The results expressed in grams of cadmium were 0.2098, 0.2106, 
0.2087. These are fairly satisfactory considering the difficulties of 
the method and serve to confirm the accuracy of the results ob- 
tained by electrolysis. So the strength of the cadmium solution 
is taken to be 0.2088 gram of cadmium in 30 cubic centimeters, 
the amount used throughout. 

The Phosphate Method. 

Drewsen * is apparently the only authority for giving to the 
cadmium ammonium phosphate one and one-quarter molecules of 

* Gmelia-Krant, Handbuch, 6th Auflage, III., 74, 


water of crystallization. So before starting work upon this method 
it seemed advisable to make a series of experiments to verify this 
statement in order that determinations could be made by weighing 
the cadmium ammonium phosphate as well as the pyrophosphate. 
Eight determinations were accordingly made and the results show 
that there is an even molecule of water present just as is the case 
with the double ammonium salts of manganese and cobalt.'*' These 
determinations were made as follows: The precipitates, thrown 
down by either microcosmic salt or by the diammonium phos- 
phate were dried to constant weight at 100-103° C. It was found 
that the weight remained practically constant up to 1 10° C. and then 
diminished rapidly, so that it was not safe to use a temperature 
above 105° C. in drying. After obtaining the weights at 100- 
103*^ C. the precipitates were ignited and weighed as pyrophos- 
phate, and from these two weights the molecular weight of the 
CdNH^PO^xHjO was calculated. The results were all between 
the limits of 242.64 and 244.4 compared with 243.85, the molecular 
weight of CdNH^PO^HjO. The average result was 243.79, corre- 
sponding to 1. 016 molecules of water. 

The only information on the quantitative estimation of cadmium 
as phosphate is in a recent paper by Austin ; t it concludes thus : 

" Cadmium may be estimated with accuracy as the pyrophos- 
phate if the precipitate by microcosmic salt in the nearly neutral 
solution containing ammonium chloride in the proportion of ten 
grams to one hundred cubic centimeters is allowed to stand several 
hours before filtering. In this way all the cadmium separates out 
from the solution as a beautiful crystalline mass of cadmium am- 
monium phosphate of ideal constitution. The conditions must, 
however, be preserved with care ; there must be no excess of am- 
monia, no free acid and no excess of ammonium salt beyond the 
quantity indicated, while that amount is necessary." 

A series of determinations was made following Austin's direc- 
tions as closely as possible ; the results were invariably low and on 
testing the filtrate with hydrogen sulphide a heavy precipitate of 
cadmium sulphide was always obtained. (See table, page 396.) 

Instead of experimenting further with Austin's conditions, a 
more promising field seemed to be the adaptation of the conditions 

* Dakin, Zeit. fUr AnaL Chem., Dec., 1900. 

t Am. J. Science, 158, 214, 1899. Alto Zeit fUr Anorg. Chemie, 22, 207, 1899* 


given by Dakin ^ for the precipitation of zinc ammonium phos- 
phate to the determination of cadmium. The essential differences 
are the use of ammonium phosphate instead of microcosmic salt 
and the absence of the ammonium chloride held to be so neces- 
sary by Austin for the precipitation of either zinc or cadmium. 

The next series of determinations (Nos. 9-12 in the following 
table) were made following exactly Dakin's conditions. The cad- 
mium chloride solutions were diluted to a bulk of 150CC. and 0.5 
cc. of hydrochloric acid added and then nearly neutralized with 
ammonia; then heated on a water bath and 35 cc. (2.6 grams) of 
ammonium phosphate solution added and the heating continued 
for fifteen minutes. The phosphate solution was made exactly ac- 
cording to Dakin's directions and contained 80% of (NHJ^HPO^ 
to 20% of NH^HjPO^. After warming, the precipitates were 
allowed to stand over night before filtering ; they were flocculent 
at first, became partly crystalline on warming and after standing 
over night consisted entirely of large scale-like crystals of a pearly 
lustre, which settled immediately and were readily washed. They 
were filtered on asbestos in a Gooch crucible and washed first with 
a I % solution of the precipitant, then with 60% alcohol, next dried 
and finally ignited, using a low fiame at first, to pyrophosphate 
and weighed. 

As the asbestos used by Dakin was slightly soluble in the am- 
monium phosphate, he determined the weight of the Gooch crucible 
after weighing the precipitates — dissolving out the pyrophosphate 
with dilute nitric acid. The results in the table give both methods 
of procedure, direct weight and by difference. The determina- 
tions made by dissolving out the precipitate are invariably high 
while those made by direct weight are slightly low. These figures 
lead to an investigation of the solubility of the asbestos in nitric 
acid as well as in the phosphate solution which showed constant 
loss with either solution, even where the asbestos had been treated 
first with boiling nitric acid and then been allowed to stand one 
portion with nitric acid, the other with a strong ammonium phos- 
phate solution for two weeks, and explained satisfactorily the error 
in using an asbestos filter. 

To avoid this source of error a series of determinations was next 
made in which the precipitates were filtered on balanced papers. 

*Zcit. filr Anal. Chem., 39, 273, 1900. 


This precipitate is particularly well adopted to this method as the 
crystals are large, easily washed and show no tendency to creep or 
run through. The reagent used was 100% (NHj^HPO^ which 
we believe better than the mixture used by Dakin, as the dihydrogen 
ammonium salt gives no precipitate with cadmium. It is easily 
obtained by adding ammonia to the purchased reagent till it gives 
a slight pink with phenolphtalein. 

The conditions for precipitation were as follows : 

To the cadmium chloride solution diluted to 150 cc. and barely 
acid with hydrochloric acid add 35 cc. (2.9 grams) of (NHJ,HPO^, 
about fifteen times the weight of the cadmium present, and allow 
to stand over night. In Nos. 13-16 (see Table II.) the precipitant 
was added to the warm solution and then heated on a water bath 
for fifteen minutes. This practice, while causing the conversion to 
the crystalline condition to take place more rapidly, is not to be 
recommended as there is danger of the composition of the precipi- 
tate being changed by loss of ammonia. The precipitation is best 
made in the cold (Nos. 17-20), then allowed to stand over night. 

Precipitates 1 3-20 were all filtered on balance papers (after drying 
at 105° C. for several hours) washed first with a hot 1% solution of 
(NHJjHPO^ then with 60% alcohol and weighed after drying to 
constant weight at 100-103° C. 

In Nos. 15, 16, 18 and 20 the precipitates were dissolved in 
dilute nitric acid, evaporated and ignited in platinum to Cd^PjO,. 
Weighing as CdNH^PO^H,0 is to be preferred as it is shorter and 
the danger of reduction and volatilization is avoided. 

These results are shown by the table on the following page. 

The next series was made to determine the effect of ammonium 
chloride as this salt is likely to be present in varying quantities. 
The conditions were identical with Nos. 17-20 in the preceding 
table except that each solution contained fifteen grams of ammon- 
ium chloride. The results expressed in grams of cadmium were 
0.2080,0.2082,0.2081,0.2083. The filtrates gave in each case a 
slight but distinct precipitate of cadmium sulphide when tested 
with hydrogen sulphide. These results show that ammonium 
chloride, in any such quantity, exerts a slight solvent action and is 
to be avoided when great accuracy is required. 





















4, '4 







w ■ 1 ' ^ 




jo ttuiu^ ui JOX13 

00 t>. t>. OsO 00 

M M M M ■■ M 



N ■* 

I I I i I I +I + I + I + I 11+4 

•uinjmpB^ JO 
souaj ill *Ord*PD 

m punoj *o*d*PD 

Q M M M 00 O 
O « M O ^ fj 

00 On On Ox On On 

M 00 *4 00 M QO gvOO 

Ox O" On t^ O N 
M ro ro fO -^ ^ 
ro fO fO fO **0 f^ 

* -♦— 

vO C\»nOx^ONN On 
t^O t^O l^vO l^vO 
f> fO fO fO '^ fO fO ^o 

w 8 

• • 



c acid. 


ith nitri 

JO suivjis uj Joxr^ 

•luniuipr^ JO «mJ3i 
"! O^H*Od*HNPD 

, 'suieoS UI panoj 
; 0'H*Od*HXPD 


• • 

I I 


fn fO 

+ 4- + 



00 On 
On On 



On N 

O ITiTfi^ ITiN TfO 

00 00 00 00 00 X) 00 30 00 00 30 
CO QO 00 00 00 X 00 CO 00 00 00 

ooopooooo o o 

00 CO 00 00 00 00 CO 00 00 

M N rO '^ *OVO t^QO On O M N f^ ^ irivO t^oo On Q 






i S3- 


The effect of heating on the precipitate was next investigated, 
and the results given in the following table : 





, -g 

• '^ 




t i& 

' as" 

3 w 





1 u 

5 3 




: .4427 


— .C044 

Heated I hour and filtered at once. 



1 .4382 


— .0063 

iC (( (< <t M 


i .2088 




' Heated on water bath I hour. 






(( ( ( (( a it 




■ 1583 


Boiled gently i hour. 






(i (( 






Boiled moderately. 



, .3346 



1 (( (< 

The danger, abready mentioned, of hastening the conversion to 
the crystalline condition by heating is fully shown by this table. 
In Nos. I and 2 the CdNH^PO^H^O precipitates were heated one 
hour in a water bath, until they were entirely crystalline and 
then filtered at once ; the other conditions were the same as Nos. 
1 3-20 in Table II., where accurate results were obtained. The 
physical properties of the precipitates were changed by the heat- 
ing, the crystals were finer and a portion of the precipitate was 
decidedly pulverulent. Nos. 3 and 4 were heated at a higher 
temperature on the water-bath until the precipitates became crys- 
talline and then stood over night before being filtered. In these, 
a larger proportion of the pulverulent precipitate was noticed. 
Nos. 5 and 6 were boiled gently ; here the precipitates consisted 
chiefly of the dense pulverulent particles with a small quantity of 
coarse scale-like crystals resembling those formed in the cold. In 
Nos. 7 and 8 the boiling was more vigorous and here the precipi- 
' tates were entirely pulverulent and in no way resembled those ob- 
tained in the cold. 

This change in appearance, together with the low results, indi- 
cate certainly a change in the composition of the precipitate- 
Ammonia is given off in noticeable quantities during the boiling. 

The coarse crystalline precipitate formed in the cold and the fine 
pulverulent precipitate, made by boiling small quantities in sepa- 



rate beakers, on account of great trouble due to bumping, were 
analyzed with the following results : 

Theory for 

Crystalline Precipitate. 

Pulverulent Precipitate. 


^ 10.83 







99.81 9& 


The analyses show the loss of both water and ammonia by boil- 
ing and suggest that the normal cadmium orthophosphate would be 
produced were the boiling sufficiently prolonged and at the same 
time confirm the determination of the water of crystallization in 
the crystalline precipitate. 

The points which we wish to emphasize are : 

1. That the electrolytic determination of cadmium is most ac- 
curate and satisfactory provided a large excess of potassium cyanide 
and the presence of other salts are avoided. 

2. That the carbonate method is the most troublesome and the 
least accurate. 

3. That cadmium ammonium phosphate contains one molecule 
of water of crystallization and can be dried without decomposition 
at ioo-i03°C. 

4. That Austin's method for cadmium is unsatisfactory. 

5. That asbestos filters should be avoided on account of the 
solvent action of either ammonium phosphate or nitric acid. 

6. That cadmium can be determined with great accuracy by 
precipitating in the cold, in a neutral solution by a large excess of 
(NHJj HPO^ and allowing to stand over night; it can be weighed 
on balanced filters or ignited to pyrophosphate. If precipitated in 
a boiling solution or on prolonged boiling it is liable to change in 
composition and give low results. 

We recommend this method as preferable to the carbonate 
method in all respects and while no more accurate than the elec. 
trolytic, it has the practical advantage of requiring no extraordinary 
or expensive apparatus. 

Quantitative Laboratory, June, 1901. 




(Continued from page 317, Vol. XXII.) 

The New Rochelle Serpentine. 

Geological Relations. — About one mile from the New Rochelle 
railway station, on the northern shore of Davenport's Neck, there 
is a small but interesting body of serpentine, which is mentioned in 
Mather's report and which has been described briefly by J. D. Dana* 
and F. J. H. Merrill.t According to Dana the serpentine lies 
along the axis of an anticlinal that widens out in a northerly direc- 
tion. He believes it to be closely related to limestone and to owe 
its origin to the alteration of included silicates, especially pyroxene. 
Merrill states that he found crystals of bronzite and actinolite in the 
process of serpentinization. The question of genesis is discussed at 
considerable length by him, with important results, of which the 
following is a brief extract: "In all probability the magnesian 
silicate rocks, which by their alteration have yielded these serpen- 
tines, were similar in their origin to the amphibolites and pyrox- 
enites which abound in Westchester county. * * * In composition 
the amphibolite and pyroxenite beds of the Manhattan group bear 
the same relation to the strata which enclose them as the in- 
trusive mass of the Palisades bears to the beds of sandstone and 
arkose between which it is now included. There is nothing but 
their somewhat foliated condition to suggest that they are of sedi- 
mentary origin and this characteristic has been shown to result 
from dynamo-metamorphism.** 

The prevailing rock of this section is a hornblendic mica-schist 
of very variable composition but usually containing both quartz 
and feldspar in considerable amounts, besides garnet, epidote and 
iron oxides. The basic phases approach amphibolite. When ex- 
posed the contact with the serpentine is usually sharp. In places, 

* Geological Relations of the Limestone Belts of Westchester Co., N. Y. Am, 
Jour. Set., III., Vol. 20, pp. 30, €/sf^, 

f Reprint from Nev York State Museum Report, 1896, p. 30. 


small veins of the latter extend into the adjacent rock but they 
are not continuous for any distance. They may well have been 
formed by extrusion of the serpentine under the pressure that was 
developed during alteration. In other respects also there is little 
geological evidence upon which to base conclusions as to the 
genesis or relative age of the rocks. 

General Characters. — In composition and appearance the serpen- 
tine shows much variety. Carbonates are usually in evidence and 
sometimes form a large part of the mass. Chromite, magnetite, 
tremolite and discolorations of limonite may be recognized in most 
of the specimens. In color the serpentine varies from green in 
different shades to reddish or white. A peculiar mottled appear- 
ance results when bunches of iron ores and calcite alternate with 

Microscopic Characters. — A study of thin sections shows that the 
rock has been derived largely from a colorless amphtbole, similar 
to that described in the serpentine of Staten Island. Included 
remnants of this mineral were frequently found and in various 
stages of serpentinization. The individuals are long and thinly 

Pyroxem occurs in partially altered but recognizable crystals. 
Merrill refers to bronzite, but the writer cannot confirm this obser- 
vation or the presence of any other variety of orthorhombic 
pyroxene. Wherever observed this mineral appeared nearly color- 
less and showed oblique extinction in prismatic sections. It is 
optically identical with the Staten Island pyroxene which has 
been described as diopside. 

Spuiel is one of the most distinctive characteristics of the rock 
and differentiates it from the other outcrops. This mineral occurs 
abundantly in well-defined octahedra that measure 2-3 mm. in 
diameter. It has a bright green color and belongs to the chrome 
spinels. A peculiar feature noticed in most of the crystals is the 
inclusion of irregular grains of chromite which usually occupy the 
centers and which are of varying size. Whether they represent orig- 
inal inclusions cannot be definitely determined, but it appears more 
likely that they have not been formed by alteration. It is rather 
to be inferred from the nature of the minerals that the chromite 
has contributed a portion of the material that has gone to build the 
spinel. Other individuals when examined with low powers show 
a cloudiness in the central portions, which under greater magnifi- 


cation is resolved into minute blades of reddish color possessing a 
sub-metallic lustre and medium translucency. These have rounded 
or rude hexagonal boundaries and appear to be arranged in lines par- 
allel to the octahedral faces of the spinel. For the purpose of deter- 
mining the nature of the inclusions a separation of the constituents 
was attempted. This was carried out successfully by digesting 
in hydrochloric acid several grams of rock which had been pul- 
verized to a coarse size. After several treatments with acid a residue 

Fic. 4. S«rpenllne from New Rochetle. Spinet crystal! will) inclunou of chiomite 
embedded in a gronndouss of KrpeDtine, 

was obtained which consisted of spinel, chromite, magnetite and 
silicates. The particles of magnetite were removed by a magnet 
and the silicates by passing the material through Thoulet's solution. 
The remainder, a fairly pure concentrate of chromite and spinel was 
placed under the microscope and the minerals separated by hand. 
Qualitative chemical tests on the spine! showed the presence of 
magnesium, iron, chromium, and small amounts of aluminum and 
calcium but no titanium. These results would indicate that the 
inclusions are some form of iron, possibly hematite. The borders 


of the spinel crystals show alteration to a colorless, low polarizing 
mineral resembling talc. 

Chromite is quite abundant, rather more so than in the serpen- 
tines from the other localities. It is usually irregular in outline, 
rarely showing quadrilateral boundaries. 

Magnetite calls for no special comment 

Utnonite is a secondary constituent resulting from the alteration 
of the other iron ores. It is distributed in fine particles, oftentimes 
so thickly as to stain the rock a bright yellow. 

Zaisite was not positively identified, but it maybe represented by 
small, brownish, biaxial crystals with high relief and low birefring- 

Calcite is always present It occurs in the form of grains em- 
bedded in the serpentine ; in larger areas made up of several crystal 
aggregates showing rhombohedral cleavage ; and as veins of variable 
width and extent Although specimens of rock may be found 
which approach ophiolite, in general the carbonates are of minor 
importance relative to the serpentine. The small embedded grains 
are to be regarded as by-products of the serpentinization, while 
the larger areas and veins have been formed by leaching of the rock 
by surface waters and subsequent deposition. Chemical tests show 
that the carbonates always contain magnesia ; sometimes in suffi- 
cient quantity to be classed as dolomite. 

The Serpentine from Tenth Avenue. 

This locality is now inaccessible, but its general features have been 
described and placed on record. According to Gale,* the rock 
was formerly exposed along Tenth Avenue from 59th Street north 
to 63d Street in a nearly north and south direction. He men- 
tions that the serpentine occupied several conical hills and was of 
variable character. Chamberlain has given the following descrip- 
tion : ** An extensive bed of serpentine lies between the Hudson 
River and Tenth Avenue from 63d to 58th Streets. It is usually 
of an olivine green color, at times nearly black. Intermixed with 
calcite so as to present a mottled appearance, it is termed ophiolite." 

A thin section of the serpentine is preserved in the petrographical 
collection of Columbia University, which shows similarity to the 
New Rochelle occurrence. It is made up of fibrous tremolite, ser- 

• Geology of the First Dist N. Y. Geol. Survey, p. 582, 1842. 


pentine and calcite. The tremolite is partially serpentinized, the 
edges being altered, while the central portions are still fresh and 
transparent. Included grains of calcite, a by-product of the altera- 
tions, and small veins of the same mineral comprise a large part of 
the mass. 

An analysis by Dr. Thomson of Edinburgh gave the following 
results : 

SiOj 44-43 FeO 9-37 

MgO 28.80 MnO 1.38 

CaO 5.06 Uf> 8.58 

Total 97.62 

The similarity in composition between this and the Hoboken 
and Staten Island rocks is noticeable. A portion of the iron no 
doubt exists in the ferric condition although it is all reported 
above as FeO. 

Serpentine from Aqueduct Shaft 26. 

Dr. G. P. Merrill * has given a petrographic description of this 
rock, found on Manhattan Island, from which the following is ex- 
tracted : " This serpentine occurs in a coarsely crystalline, white, 
granular dolomite. It plainly originates through the hydration of 
a white monoclinic pyroxene showing under the microscope 
nearly rectangular prismatic cleavages and giving extinction angles 
as high as 44^. The alteration is accompanied with the formation 
of abundant secondary calcite. * * * Under the microscope the 
serpentine shows a platy, almost fibrous structure, the plates in each 
case lying approximately parallel with the vertical axes of the 
crystals from which they are derived." An analysis by Mr. 
Catlett showed the results below : 

















Total 1 


♦ *« Notes on Serpentinoui Rocks of Essex County, New York,'* etc., Procad, of 
the U, S, Nat. Mus,^ Vol. XII., pp. 595-600. 


From the statement by Dr. Merrill that the serpentine occurs 
in dolomite, it appears that the relations differ from the other 
localities. The evidence is hardly sufficient, however, to base any 
conclusions as to the nature of the original rock, and unfortunately 
no further information is available at present. 

The Serpentine at Rye. 

An area of several acres in the eastern part of Westchester 
county consists of serpentine. The exact locality is a little more 
than a mile north of the village of Rye and about an equal dis- 
tance west from Port Chester. The outcrops, of which there are 
several, show the rock to be completely serpentinized and some- 
what disintegrated. In thin section no clue could be found as to 
the origin of the mass and nothing of interest was noted. The 
area, like those of Staten Island and Hoboken, is lenticular. 

Mineralogy of the Serpentines. 

The characteristic constituents of the serpentines have been de- 
scribed in the preceding pages. There is also a series of accom- 
panying minerals, which are closely related genetically and which 
may be regarded as by-products, some having been derived from 
the alteration of the parent mass, while others owe their origin to 
the leaching and disintegration of the latter by atmospheric 

The serpentinization of complex rocks, as these must have been, 
cannot take place without a dissolution of the chemical molecules 
and a readjustment of materials. This is essentially a process of 
oxidation and hydration and is accompanied by the formation of 
various oxidized compounds whose bases do not enter into the 
composition of serpentine or are in excess of the molecular pro- 
portions. In the case of olivine, an ortho-silicate of magnesium 
and ferrous iron, which alters readily to serpentine under natural 
conditions, the change may be expressed, according to Tschermak, 
as follows : 

2(Mg,SiO,) + CO, + 2H,0 = H,Mg3Si,0, + MgC03 
SCFe^SiOJ + 4O + 3H,0 = Fe,03(0H), + 2Fe30, + sSiO,. 

The equations indicate the formation of four by-products, but it 
seems probable that the excess of magnesia in the first will combine 
with the free silica to form serpentine, and it is also certain that a 


portion of the iron may replace the magnesia. The processes be- 
come much more comiplicated when the starting point is pyroxene 
or amphibole, which contains lime and alumina as additional bases. 

Weathering and leaching of the serpentine by ground waters 
contribute to the development of a third generation of minerals 
represented by iron ores, carbonates and vein serpentine. Iron 
oxides are especially susceptible to these influences. When the 
rock has been exposed for a long period, it has a thoroughly 
bleached appearance, is porous and granular, the grains of mag- 
netite being almost completely removed. Chromite does not 
break up so readily as magnetite, but nevertheless, yields to the 
same influences. 

The by-products are found in the groundmass of the serpentine, 
where they occur usually in granular form, in surface depressions, 
joints and cavities. 

The following minerals are known to occur with the serpentines : 

Aragonite, according to Dana,* occurs in fibrous crusts and 
other forms at Hoboken, no doubt accompanying the serpentine. 

Asbestos. ^'Qoih varieties may be found in close proximity at 
Pavilion Hill, The fibrous amphibole or tremolite is distinguish- 
able from chrysotile by its lighter color and superior hardness. 
The latter variety is common at Hoboken and New Rochelle. A 
fine specimen of asbestos with fibers nearly a foot in length was 
picked up at St. George, Staten Island. 

Brucite occurs at Staten Island, Hoboken and New Rochelle. 
The first two localities are mentioned in Dana's Mineralogy, b^t 
could not be confirmed by the writer. Thinly foliated speci- 
mens of small size were collected at New Rochelle. 

Calcite in granular masses is common at nearly all the serpen- 
tine localities, while small veins may be seen at New Rochelle. 
A specimen of serpentine from Hoboken, collected by Prof. J. F. 
Kemp, shows a large number of calcite crystals implanted in a 
matrix of finely divided carbonates. It was probably taken from 
the wall of a fissure. The crystals are of slender habit, being 
bounded by a steep rhombohedron and are much distorted. 

CJialcedony is recorded by Mather as occurring at New Rochellcf 
It is abundant and of a red color near Stevens Institute, Hoboken. 

* «• System of Mineralogy,'' 1900, p. 283. 
f Report on the First District of N. Y., 1843, p. 426. 


Chlorite has already been mentioned and described under the 
Staten Island area. 

Chromite is described under the different localities. This mineral 
has been regarded as an alteration product of olivine, but such 
derivation seems very doubtful. 

Dolomite is of common occurrence, but seldom in crystals. Beck* 
describes a peculiar variety from Staten Island : " Dolomite closely 
resembling gurhofite is found at the Quarantine. It is white, com- 
pact, has a flat conchoidal fracture and is so hard as sometimes to 
strike fire with steel. Exceedingly tough and difficult of solution, 
except when in a very fine powder. Sp. gr. 2.712." He gives the 
following analysis: 

CaC03 5275 

MgC03 42.25 

Insol SiOjj 5.00 

Iron oxide trace 

Total icx).cx) 

Hydroviagnesite, — The occurrence of this mineral at Hoboken 
is mentioned by Beck.f WachminsterJ gives for its chemical 
composition : 

SiO^, 0.57 

FCjOg, 0.27 

MgO, 42.41 

CO2, 36.82 

HA 18.53 

Loss 1.40 

Total 100.00 

Eliminating the small amounts of SiO^, FCjO, and loss by ignition 
the molecular ratio of the remaining constituents is : MgO ; 
1.084; CO,, 0.837; HjO, 1.023. This is a close approximation of 
the formula, 4MgCO,, Mg(OHX + 4Hp. 

Hematite in pulverulent form is quite widely distributed, though 
not found anywhere in large amounts. A jasperoid hematite out- 
crops at a short distance from the Hoboken serpentine but within 
the limits of the Triassic sandstone. 

♦"Natural History of New York," Part III., p. 255. 

t Ibid. 

X Sillvnan's Journal, Series XVIII., p. 167. 


Limonite forms extensive deposits on Staten Island and in years 
past has been of economic importance. The ore occupies superfic- 
ial depressions in the serpentine and is never of great thickness. It 
has a granular or coarse oolitic structure and may be soft and 
earthy or cemented into compact and tough masses. The deposits 
have been formed, without doubt, by precipitation from chalybeate 
waters which have come to rest in the depressions after circulating 
through the serpentine. An analysis of ore from Castleton Cor- 
ners, by D. J. Tysen, Jr., quoted by Dr. Britton,t shows the fol- 
lowing percentages : 










1. 15 









The presence of an appreciable amount of chromium is inter- 
esting. Abandoned limonite mines may be seen at Castleton 
Corners, Grant City and on the EUingwood road. 

Magnesite in massive or finely fibrous aggregates is found at 
Hoboken, New Rochelle, Pavilion Hill and other localities on 
Staten Island. Good specimens were collected from Hoboken, 
where it occurs quite extensively. Analyses of the mineral from 
this locality have given the following results : 

















1. 17 













Nos. I and 2 are by C. A. Joy. Ann, of the Lye. of Nat. Hist., 
Vol. VIII., p. 122. 

No. 3 is by Nuttall. Stll. Journal, Vol. IV., p. 17. 

Magnetite is common in the form of irregular grains or small 
octahedra. It is mostly if not entirely a by-product, separating 

t"The geology of Richmond Co.,N. Y.," Ann. of the N. Y. Lye, of Nat. HUt., 
VoL II. 

VOL XXII. — 2 8. 


out during the alteration of the primary components, more especi- 
ally amphibole and olivine. 

Quartz is found only along fissure or joint planes and on the 
surface of the serpentine. Good crystallized specimens were 
picked up at several localities on Staten Island where it occurs in 
company with the limonite beds. The crystals are small and show 
the ordinary prism faces terminated by =t R. 

Serpentine, — Various, more or less distinct varieties of minerals 
may be included under this caption, all of which are similar in 
composition and approach ordinary serpentine in other respects. 
Antigorite is a thin lamellar variety, with greasy feel, found at New 
Rochelle, Hoboken and Staten Island. Bowenite is possibly repre- 
sented by a massive, pale mineral found in small veins at New 
Rochelle. It has a hardness of 4.5-5 and resembles common opal 
in appearance. Splinters fuse to a white bead before the blow- 
pipe. In thin sections, it is seen to be built up of round or oval 
granules showing concentric structure. The centers are amor- 
phous or crypto-crystalline, while towards the periphery they are 
crystalline and polarize feebly. The outer layers are composed 
of minute fibers arranged radially and with parallel extinction. 
Unlike ordinary serpentine the axis of least elasticity of the 
fibers is at right angles to their length. Its chemical similarity to 
serpentine is shown by the following analyses : 








1. 17 











Total, 100.44 101.46 

Deweylite, an amorphous, resinous variety is reported by Prof. 
D. S. Martin as occurring on Staten Island. The mineral contains 
more water than ordinary serpentine; its formula according to 
Dana may be H^ Mg^(SiOJj+ 4H2O. Mannolite^ thin foliated 
serpentine with pearly luster and light green color is found at 

Talc is widely distributed, but rarely in quantity. Talcose bands 
intermixed with serpentine and chlorite may be observed at Pavilien 


Hill, Grant City and Hoboken. The Grant City locality is prob- 
ably the one referred to by Dana "*" as affording detached masses of 
foliated, snow-white talc. 

Hydrous Anthophylhte may be included in the list on account of 
its historical interest, the type locality being at Tenth Avenue. It 
was formerly supposed to be a distinct mineral species, but Grata- 
cap established its identity with tremolite in process of serpen- 

Origin of the Serpentines. 

The derivation of the Staten Island, Manhattan and New 
Rochelle serpentines from anhydrous, magnesian silicates in- 
cluding pyroxene, amphibole and olivine is established by the 
presence of these minerals in varying stages of alteration and by. 
the structural characters observable in most of the specimens. As 
to the 'nature of the parent rocks, however, further than their 
mineralogical composition, very little information is obtainable 
from microscopic study. The processes of serpentinization have 
nearly reached completion, while the original arrangement of the 
constituents has been obliterated by powerful dynamic forces. 
Nevertheless, there are a number of considerations brought out in 
the preceding pages, which, when compared with the results ob- 
tained in fields more favorable for investigation, indicate that the 
original rocks were of igneous character and allied to the pyrox- 
enites and peridotites, rather than that they were impure lime- 
stones, as has been believed by some geologists. These consider- 
ations may be stated as follows : 

1. The lenticular form of the areas and their relative elevation are 
characteristics usually associated with igneous intrusions. Lime- 
stones even when impure disintegrate rapidly and had they been 
subjected to the action of weathering and erosion it is certain that 
they would have been worn down to a much lower level, relative 
to the surrounding schists. In form the areas are analogous to the 
more or less serpentinized pyroxenites and peridotites that have 
been described from a great number of localities in the Atlantic 
coastal plain. 

2. Their position is difficult to explain by the sedimentary 
theory, as Dr. F. J. H. Merrill has remarked. Such a rapid and 
violent change in the conditions as would lead to the deposition 

*«« System of Mineralogy," 1900, p. 680. 


of basic magnesian rocks, when before and after them there were 
long periods in which strata of acid aluminous type were laid 
down, is not in accordance with our knowledge of sedimentary 

3. It is contrary to general experience that impure limestones 
yield by alteration a uniform mass of silicates and such uniformity 
is necessary to explain the sameness in character which the ser- 
pentines exhibit wherever observed. Throughout the extensive 
area of Staten Island, comprising a surface of 12-13 sq. miles, 
there appear to have been no silicates present other than pyroxene 
amphibole and olivine. 

4. Chromite has rarely if ever been noted as a constituent of 
serpentines whose origin can be traced to sedimentary rocks- 
From the observations of Voigt* it would further appear that this* 
mineral may be regarded as an absolute criterion in determining 
the genesis of serpentines, since it is a primary constituent of 
eruptive magmas. J. H. Pratt f has recently advanced this theory 
in connection with chromite deposits of North Carolina. 

5. The chemical composition of the serpentines is similar to 
that of pyroxenites. 

Age of Serpentines. 

It is impossible to fix any close limits upon the age of the ser- 
pentine bosses. The most that can be shown with certainty is 
that they are older than the Triassic sandstone and of more recent 
origin than the Manhattan schists which are now recognized as 
metamorphosed Hudson River strata. From present conditions it 
would appear that the original rocks were coarsely crystalline and 
had cooled at considerable depths. The parallel structure and 
frequent jointing show that they were under great pressure at one 
time, but there is no evidence of any re crystallization and they 
must have been unaffected by the dynamic forces which brought 
about the transformation of the sediments. The sequence of events 
was substantially as follows: {a) Deposition of sediments, (^) 
Metamorphism, {c) Intrusion of pyroxenites, [d) Orographic distur- 
bance, {e) Denudation and appearance of bosses in their present 
position. This is in agreement with the outline of the origin of 
the New Rochelle serpentine as given by Dr. F. J. H. Merrill.J 

* Zeits, /. prakiische Geologie^ 1894, pp. 384-393 
f TVans, Am, Inst. Min.^ ^f^g't Feb. 1897. 
% Op, cit. p. 42. 






Cycle VI. 
Let Figs. 3 5 and 36 be the diagrams of the cycle. 




no. 35. 
Heat being added isopiestically 

FIG, 36. 

.•.7;-7;(i + ^) = 7: 

p " 

«'» = »„ 7- = ^'«^ 


V = IK =s VX 
r a 

The point C lies on the adiabatic through A, hence 

^* ^""yvj ^""Xvi) 

a \ * a 

^ xy 

V V'^ 


• T = 





The perfect gas law is seen by inspection to apply 

* Continued from page 366, Vol. 22. 








E= I- 


^f - f » - f a = ^p log. f 


.-. Rip » C, log, X 

M.E.T. = \ 

^, log. ^ 

R^^^V —V ss V (x— i) 

M.E.P. -/ 



^, = A-A=a(i-^.) 

M.E.V.= / ^--'iJ 






Cycle VII. 
Let Figs. 37 and 38 be its diagrams. 

























■ — p 



FIO. 37. 

FIG. 3& 

For B as before we have 

^6 = 



A = 






The point 

C lies 






B and is 


to the 


-.»n(5;)--„-(^;)> (5) 










^'» - ^'a^ 

^„^ ^6 

^',/ *" ^c 

7; 7; 

7; " t: 


T, = - - 

■^ X 










W^ H,- H,^ H,- C;r^x{^^^-< [i --^ (II) 

E^i-") (12) 

/?^ » Cp log, X as in VI. (13) 



R.^v —V 











A -A 

R^^T,-T^ T„{x - I) as in VI. 

Cycle VIII. 
Figs. 39 and 40 are its diagrams. 














— — n 



no. 39. 

r\Q. 40. 

We have for B 


T, - Tx 



The isothermal through A intersects the adiabatic through B to 
determine C, 

From the adiabatic 


From the isothermal 

re J*a ,., 



7'/ = V yxy 

o a 


a r 




By substitution 

'• ^'r - ^^a^-*- 


M a 


^t - ^«(f » - fa) = 7;<:, log. ;ir 

W ^ H, - T^C^Xog, X 

£=i - 



M.E.P. -/ 

^i - T„C^ log, jr 

■^1— 7;ciog, 4: 







M.E.T. = 

M.E.V. -/ _ 

i) o« 

Cycle IX. 

Let Figs. 41 and 42 be its diagrams. 




FIG. 41. 

FIG. 42. 

Up to the point C the results of VII. may be assumed. 




The point D lies on an adiabatic through A and is subject to the 




'^i > ^'» 




-„f m" 

A^ ^" 

H r 

A ~^ 




1 *"l 

R^ r= Cp log ^ ;r as before 


•E.T. = J ( 

<^p log, X 


^, = ^.-^a=^a[^(^'')^"-l] 




['ft F - ■] 



if,-N,^N,-cj^x(M'-:(,-i) (7) 






n /^a Jtd * a 


M.E.V. =/- 




/?r=7;-r = 7;(4r-i) 





Cycle X. * 

In this cycle as in the last four heat is added at atmospheric 
pressure, then follows adiabatic expansion after which heat is ab- 
stracted according to some law as yet undefined. Adiabatic com- 
pression completes the cycle. As the law of abstraction of heat 
is as yet undefined we cannot, of course, derive formulae for the 
cycle and will leave its discussion as we did Cycle V. 

We might have derived formulae for the imperfect carrying out 
of cycles VI., VII., VIII. and IX. but they are of such slight im- 
portance in practice that it did not seem desirable. 

Besides the twenty-two cycles considered there may be others 
due to the combination or differentiation of these typical ones, but 
the object of this paper will be best accomplished by a study of 
types, he non-typical or synthetic cycles having been omitted. 
The method of study here set forth being of universal application 
to all possible cycles will furnish means of reaching a clear under- 
standing of any of the unconsidered cycles should need arise. 

The results obtained will be compared in the next paper. 


Of the many cycles considered we will choose for comparison 
only those that might be called the perfect cycles because accurately 
defined and these are Cycles I., I. C, II., II. A^. II. C, III., III. C, IV. 
IV. C. The atmospheric cycles are of comparatively little impor- 
tance and will be neglected in the comparison. We will take up 
each variable separately and study its value in the different cases 
by formula and by calculated examples expressed in curves which 
are then the graphical formulae. The curves given are approxi- 
mately correct and as the same approximation will probably main- 
tain for all the cases the curves will serve as well for comparison as 
if absolutely exact. Two cases of each are given, one with com- 
parison 2: I and one with 10; i. Call the atmospheric values 

A» ^«» ^«- 

Temperatures after Addition of H^, B. T. U. 


I.,I.C 7:=7„^=7;(n--^^-) (I) 




II., II. A.. II. c t; = T,x= '^A^^ ct) 



IV., IV. c 


• loe aoo 900 

(ICAT 0*iTi A9f ep 


= ^^(' + o:) 

r* h 





Using axes of T^ and /^j we see these are all straight lines 
passing through the axis of temperatures at 7"^ above the origin 
except in cycles (I., I. C) where the intersection is at 7^. These 
lines are inclined to the axis of H and make with it an angle a 
such that in 

I., I. C, II., II. A, II. C tan a = 





and in 

tan a' = 

while IV., IV. C are lines parallel to axis H^ 

These lines are shown in Fig. 46 for two compressions. 



I., I. c 

Pressures after Addition of H^, B. T. U. 

a=a^-a(i + ^^) 



HCiPCr <)NiT> A^pc^ 

II., II. A, II. C 

A = A-^=A^I 


A = A 

IV., IV. C 

* —A 1 


v* b 








Equations (7), (8) and (9) are all straight lines, (9) being parallel 
to axis //, while (7) and (8) are inclined. Equation (10) is an 
exponential curve sloping down to the right and concave up and 
asymptotic to axis of H as can be seen from the derivatives 










These formulae are given in Fig. 47 for the two cases. 

Volumes after Heating bv //j, B. T. U. 

'«* 8e» BOO 4«o eoo «oe 700 

►tCAT <)KiT> A9?cp. 


I., I. C Z' : 

II., II. A, II. C v^ = t'„ 

III., III. C V, = 7., Y = z', ( I + -^^ ) 

t *6 


IV., IV. c 

c ft b 




3r2.PA 21lC2 



Formula (13) is a straight line parallel to H^ and is always less 
than (14) which is similar but cuts axis of V^ at a point v^ higher 
than v^^. Equation (15) is a straight line inclined to H^ Equa- 
tion (16) is an exponential curve cutting axis V^ at point I^, it is 
concave up and slopes up to the right as is shown by the derivatives 






dH, {C-QT, 




1 \ ^p ^vj ■* 6 

These curves are shown in Fig. 48 for the two cases. 


Temperature After Expansion. 









I. c 





•00 -900 4oe eeo «oo 700 



r a 

T.= T 

a a 




II. A t,^tX^t\i+"'^) (22) 









T, = T„ (23) 

7; =7:. (25) 

T, = T^e'-' = Te'"'^ (26) 

T, = 7; (27) 

Curves (19) and (21) are similar in form, cutting axis 7^ at dif- 
ferent points, however, and having different slopes. It is easily 
seen that (21) is always greater than (19), also that (22) is greater 

than (21) since 

< I. 


Both (22) and (24) are straight lines, but they have different 
slopes though intersecting axis T^ at same point 

(tan J)„. ^. = -" = ^^-. (28) 

(tan <J),„. = „-',-^ (29) 

whence (22) is always greater than (24). Equation (26) is an ex- 
ponential cutting 7^ axis at 7^, it is concave up and slopes up to 
the right since 

-^ = - i - e^'''"'" (30) 

" S-"^ (31) 

dH;' r-'c;T, 

These curves are shown in Fig. 49 for the two cases. 

Pressures After Expansion. 


I- A = A. (32) 




n=,7 M.T5 AWejl 


II. A 

P,~pX~f.(l + (--^ 

II. C 

p. p. 


A- A 


, A A 

'' „.A /. . //, 




" <:,T, I 



IV. C 

A = A 

JL * a 




Equations (32), (34), (37), (39) are identical and represent a 
straight line parallel to axis //,. Curve (35) is a straight line in- 
clined to Hy All the others are concave up sloping down to the 
right ; their relative positions are seen in Fig. 50 for two compres- 

Volumes After Expansion. 









leo aoo 900 'leo 900 600 700 400 900 too9 

dCA"^ 'Mit; t^^t^ 




I. c V, = t'„;^^ = ^o ( ' + c f )'"' (42) 

II. 7'^ = z/„^ r = z;^ ^ I + ^^^J V (43) 

II. A v^ = v^ (44) 

II. C T', = vX'-^' = ^'„ ( I + /f )^"^ (45) 

r-* 6 

III. ^'. = ^'„>^ (46) 

III.C ^„ = ^„>'^'' = ^'„(i + /f)^-' (47) 


IV. t.,= z,„.''-> = z,/.-6 (48) 


IV. C 7', = v^c' = '^/(^^p-'-') '•^ (49) 

These curves will admit of considerable discussion, but the curves 
of Fig. 51 show at a glance all we wish to know in general. 

Heat Discharged or Abstracted. 


I- H, = CTlx\-i) (50) 

I.C /f,= 7;c-„iog.x=7;c.iog.(i + ^^) (51) 

II. H,^CJlX-.-i) (52) 

II. A H,= C,TSX-i)^^}^ (53) 

II. C ^.= C-.7;iog.(i-^^) (54) 

r^ 6 

III. H,= C^UY-i)^^, (55) 

III. C H,= CJ„ log. (i + ^^ ) (56) 

IV. H, = C^Ue^-^ - I) = Cjy^" - I) (57) 



IV. C 


Equations (33), (55) and (58) are identical, that is, these three 
cycles will discharge the same amount of heat and have the same 
efficiency ; moreover this efficiency will be independent of every- 
thing but the compression. These three cycles have, further, 
a common property not seen by the formula, but from their defini- 
tions each receives and discharges all its heat according to the 
same law. 

Cycle II. A receives all heat at constant volume and discharges 
all at constant volume. 

Cycle III. receives all heat at constant pressure and discharges 
all at constant pressure. 

Cycle IV. C receives all heat at constant temperature and dis- 
charges all at constant temperature. 

A consideration of the above would seem to warrant the prop- 
osition : 

When all the heat is discharged according to the same law un- 
der which it was received then the cycle will have an efficiency 
independent of everything but the previous compression and will 

be given by 


£= I - 


We may remark here that as IV. C is the Carnot cycle we can 
state that Cycles II. A and III. have the same efficiency as the 


3E e 


>»• 100 too 400 SCO 000 TOO aeo BOO leoe 


H»M Ortit> AJ'PE^. 



Carnot Cycle with same previous compression. This is an im- 
portant supplementary to the old theorem that the Carnot Cycle 
has the highest efficiency for its temperature range. 

The relation between the other values of H^ are best shown 
by the curves of Fig. 52 by implication. The quantities: Pres- 
sure range, Volume range, Temperature range do not need sepa- 







HWCT <^»»IT> ^Wcp 

rate sets of curves as we can get a fair idea of the values from an 
inspection of the previous curves. If, however, any case seems to 
call for an exact solution it can be obtained by a simple substitu- 
tion in the formula; already given. 



Mean effective pressure, volume and temperature, however, are 
important values and not easily located relatively from the formulae. 
Figs. 53, 54 and 55 show these curves as calculated for two cases 
of compression. It may be here remarked that in the case of 
Cycle IV. when the compression is 2:1 only about 44 B.T.V. can 
be added to i lb. air and with a compression of 10:1 about 282 
B.T.V., this is why the curves end abruptly at these values of H . 











«oo TOO aoo «oft io*e. 

A thorough discussion of the equations derived while important 
and leading no doubt to many new and useful results would be 
very long and would extend beyond the limits set for this paper 
which had for its object rather the exposition of the method of 
procedure than a thorough application of that method. 



Besides the complete discussion referred to there is another im- 
portant point of view to be taken of these formulae — that of inter- 
pretation with respect to operating engines ; this is also reserved for 
later treatment. 


r\c#s|4 crrcCriVc Tcr|Pn»V^TOi\c> 


Fellow in Mineralogy, Columbia University. 

In the number of crystal forms and variety of their combina- 
tions calcite is unsurpassed among minerals. For such a mineral 
a list in which are brought together the letters and symbols of all 
the forms with their interfacial angles is exceedingly useful. 
Irby's* list does well for its time, but he uses no letters and gives 

*"0n the Crystallography of Calcite." (naug. Dis., Bonn, 1878. Abstr. in Zeit., 
^^ysf., 3, 610, 1879. Other lists published since this are: Goldschmidt, « Index der 
Krystallformen der Mineralien," 1886 Ces^ro, Ann. Soc, G, Belg, Mim., 16, 165, 
1889. Goldschmidt, « Krystallographische Winkeltabellen,'' 1897. 


one half the true interfacial angles instead of the normal angles 
which are obtained directly by the reflection goniometer. Besides 
many new forms have been made known since that time. 

This paper is largely a compilation using Irby's list as a basis 
The angles have been taken from this list and from the papers of 
the various authors. Other angles have been calculated by the 
writer but there yet remain blank spaces for some of the angles. 
Taken in connection with Irby*s paper it serves as a reference to 
the first record of each of the forms in the literature. 

It may seem to some that since the appearance of Gold- 
schmidt's " Winkeltabellen " a paper of this kind is unnecessary, 
but Goldschmidt's system of angle measurement will probably 
not come into general uj>e for some time and besides in many 
cases it is more convenient to use the ordinary reflection goni- 
ometer than the two-circle goniometer. 

In a work of this kind errors have doubtless crept in. The 
author will be grateful for the mention of any such, great or small. 


The following formulae modified from Dana and Naumann are 
useful in calculation. 

r = 0.8543; 2^= 1.7086; r* = 0.7298; 2^^= 1.4596; 

4^2= 2.9193. 

Dihexagonal Prism \hkio\. 
COS X (ax,al) = -^^j^ _^ ^ _^ ^^^^ . 

Pyramid of the Second Order {h- h-zli-l]. 
cos Y (terminal) = j., -^' ■^,f-z ■ 

cos Z (basal) = ,., - t«,.,. 

Rhombohedron \hohi], 

3/2 _ 2/^c^ 

cos ^(terminal) = —,h ^,0.2 

^ 3/ -I- 4//^ 


tan (0001 '\hohl) tan (0001 Ahohl) h 
tan (0001 A I oil) ~ .986447 ~ /■ 

Scalenohedron [hkll] 
cos XihkllfMukl) = ^^,^ ^^^. ^ ^-^^-^y . 

3/== + 2r{2/^ + 2hk — /•*) 
cos F(Aiz/ A .>/./) = --^^-.,^^^_ ^ _^ ^^^^- . 

cos Z{hkllf\khll) = - ,-_^^^^,_y_^y . 

To find the values of m and « in niRn from the measured angles, 
X\ F' and Z' (true interfacial, not the normal angles), the fol- 
lowing equations serve : 

^. „, . ^, cos \X' + cos J y 

Given Jl ' and F', n = - -f ;jy7 -,,. 

cos \X' — cos J F' 

sin iZ' 
Given X* and Z', ^/ == 

2 cos^y?f' — sin |Z'" 
sin \Z' 

Given F' and Z' « = -; — i^^, , ,.,, 

sm ^Z' — 2 cos ^ F' 

tan ^Z' \/^ 

5: == cos f ; cot f = cot f x 0.20274 = w. 

When Z' cannot be measured its value may be obtained from this 


sinJ^' + siniF'=cos^Z'. 

The angle between any two faces P{fikll) and F{li'k*VU) may 
be found as follows : 

iW + 2c'{hk' -f kh' 4- 2IM + 2/^>&') 

"" ^ [3/'-f 4^'^+ >&" + hk)-\ [3/'' + 4^(//'' + ^'' +7/'/^)]' 
For special cases the formulae are simpler : 

/;. . X 3^-f 2r^(>t+ 2I1) 

COS (/ikl/ /\ 1011)= -r\ ^ ; 

^ v/ [3/H 4^(/^ + >^ + //>&)] [3 + 4^] 









1 M. 

M -f- 



w 43 


















I , 

4- ' 


; 8 






5» ' ^. 



«fe 8 


1 «^ 











1 J. 

o ?> 



•%•! 1 





















1 '"^ 





M 1 

w 8 

, t 1 1 


' « 


; « 






' + 
















"*' 1 












COS ihkil ^ 1010)= --. _ ^ ^ 

cos (hkll [\ 000 1 ) = -y=^. .: 

v/3/2 ^ ^c'iji'- + yt' + ///&) 

By the use of the following table a symbol given in one of the 
our usual notations may be changed to that of any other. 

Careful attention must be given to the signs. 

The following tables contain a list of all the forms observed on 
calcite crystals. A supplementary list includes uncertain and in- 
correct forms, but no attempt has been made to eliminate all of 
these from the list proper. 

List of Forms. 



1 />. 





I \ 







00 P2 




' lOIO 

00 R 

1 1 20 A '^'^^ 
25 17 



cagonal Prisms. 

*4 i 









16 6 






10 53i 
0001 A 




i "" 

of the Second Order 


29 40 





28 39 




44 55i 

41 20 





48 43 

44 84 



1121 i 


60 40 

51 10 

II < 





66 18 

54 30 



16. 16.32.9 


71 46i 

56 44 





73 4ii 

57 21 





75 55 1 

58 ij 





77 37 

58 28 






78 57 J 

58 47 






81 40} 

59 18 





83 19 ; 

59 34 i 

These numbers are references to the literature given at the end of the paper. 
4. Schnorr, Neumark, 30. Also Palache, Lake Superior, 37. 
10. Palache, Lake Superior, 37. 

12 ** <' '* <* 

18. Rogers, Frizington, 39. 



After the enumeration of the prisms, pyramids and rhombohe- 
drons, the scalenohedrons of the two prominent zones (0112.1011. 
1120) and (1011.0221.1120) are given. Then comes the positive 
scalenohedrons and the negative scalenohedrons not included in 
the above mentioned zones, arranged for convenience of reference 
according to the value of ;// in the Naumann symbol mRn. A 

Positive Rhombohedrons (hohl). 






0001 A 





1 28i^ 




*20 ' 


87 39i 1 





87 35 1 

ti9 5o 




87 2ii , 1 



20.0.20 I 



87 7i , 1 

119 44 

*24 \ 

U' \ 

19.0. I9.I 


86 56I ] 






86 46^ ] 





16.0. 16. 1 


86 22^ ] 



s • 



13.0. 13. 1 


85 32i ' ] 






8444 J 

ti9 10 


r- 1 



1 10^ 


84 12} ' ] 

118 59f 

*30 , 




9^ ; 

83 34i ; 1 







81 45i ' J 

117 59 




81 6^ i ] 

117 39 


/• , 


1 6iP 

8o24i ] 

117 17 





' y^ 

79 33i 1 

116 47i 


"■ 1 


I 5A' ; 

78 32 I I 

[16 9 


w • 



: 4^ 

75 47 1 

[14 10 






71 19} ] 

no i5i 






67 56 ; 1 







63 7 J 

[oi 9 

40 ^ 





44 36i 

74 55 





, JA" 







29 24i 






26 15 

45 3 




, «^ 

21 32 

37 4 






13 51 ' 


20. Bodewig, Matlock, 7. 

21. Hobbs, Wisconsin, 28. 

22. Sansoni, Badenweiler, 20. 

23. ** St. Blasien, 20. 

24. Goldschraidt substitutes this for 4- \%R given by Foullon, Leoben, 14. 
28. Thiirling, Andreasberg, 15. 

30. Sansoni, Blaton, 1 3. 

32. HOfer, Rauris, 23. 

39. Palache, Lake Superior, 37. 



Negative Rhombohedrons (ohhl). 









II 9i 





19 18 




19 3 






21 32 

37 4 




01 1 3 


26 15 

45 3 







28 29 



V • 





52 21 


C- ; 









35 23 

60 12 


O.I I. FT. 14 


37 47 

64 5 







64 53i 






68 53i 



43 8i 



K ' 


01 f I 


44 36i 

74 55 




46 22 







48 25i 








49 48^ 

82 5oi 









51 28 

85 16 


> ^' 





87 10 






54 5i 

89 5 


, 0.13. 13.9 









55 57 





57 loi 



r • 



58 2i 

94 341 


i 0553 







59 55 

97 4i 







97 58J 



61 3i 

98 33i 






63 7 

loi 9 





65 441 

104 i7i 





67 I 





, 28. 







, Mtinstertl 

lal, 20. 




5, 32. 



Rhisnes, 1 




K)n, Norbei 

¥» 24. 



i, Andreasb 

►erg, II. 


















Des Cl< 

oizeaux 6. 

Also Sansoni ll 

[ and Penfield and Ford, 38. 



similar order is used in the uncertain fonns. The letters in the 
column marked •* Gdt." are those used by Goldschmidt in his 
** Index der Krystallformen der Mineralien." In the next column 
" D " are the letters used by Dana in the System of Mineralogy, 
6th edition, and when enclosed by parentheses those used by other 
authors. Then follow the symbols of Bravais-Miller and Nau- 
mann. The interfacial angles given for the scalenohedrons are X 
Y, and Z, those over the acute polar, the obtuse polar and the 
basal edges, respectively. 

The forms with the number starred are those not mentioned by 
Irby, most of them having been made known since the publication 
of his paper. At the bottom of the pages where the starred num- 
bers occur are given the observer and locality of the form with a 
number referring to the literature at the end of the paper. 

Negative Rhombohedrons (ohhl). — Continued, 




Brftvais- Miller. \ 



Naumann. j 


0001 A 



V" ' 


0552 \ 

\R ' 

67 56 





O.II.II.I4 , 

-y^ \ 

69 46 1 

[08 42 





-3^ ' 

71 I9J ] 

[lo 15J 




0.16. 16.5 

-v^ ' 

72 13 ] 

[II 4 





—i/i , 

7351 J 

fi2 34i 





—4/1 1 

75 47 J 

[I4 10 





77 18 1 

[15 18 


^m ' 



-SiP ; 

7832 1 

16 9 



d 1 

0881 , 


8247 ' 






-9^ 1 

83 34i J 



' 2- ' 

2 \ 


liiV , 

8444 1 

[19 10 




0.13. 13. 1 

—13^ ' 

85 32h J 





0.14. 14. 1 


8551J ' 

[19 29 




0.17. 17. 1 


8635 1 





— 20A* , 

87 7i 1 

:i9 44 



1 , 


87 56 1 




' 1 

— 36A' 

8826 1 

:i9 54 


Naumann, 3. 






Sanson i, 

Andreasberg, II. 


Hofer, Rauris, 23. 


Des Cloizeaux, 6. 


Farrington, Joplin, 



Thiirling, Andreasberg, 15. 


Johansson, Norberg, 24. 

























« : 
' u : 

i X: 




w : 


e : 
















4.7. II. 18 






II. 6. 17. 23 


73- 10. 13 


7.2.9. II 






8. 1.9. 10 

9. 1. 10. 1 1 

13. 1. 14.15 


17. 2.19.15 

19.3. 22.16 


94. Moesz, Kdr5smezd, 34. 

95. ** " ** 

97. Palache, Lake Superior, 37. 

99. Moesz, K6r6smez6, 34. 

loi. Morton, Banile, 12. 

103. Gonnard, Couzon, 31. 
105. ** ♦» *' 

no. Melczer, Budapest, 29. 

112. Sansoni, Andreasberg, 1 1. 

118. Zepharovich, Bleiberg, 8. 

121. Sansoni, Andreasberg, ii. 

124. Thtirling, *« 15. 


- \ii\ 

- JA'V 

- iiPS 

- \Rl 

+ A^9 
+ 1'«^7 
+ lA-S 

+ i^V 

+ A^V 
+ i^3 
+ A^i 
+ i^i 
+ \R^ 

+ \R\ 











o / 

39 42 
ZZ 22 

35 16 
32 49 

36 47 
27 40 

34 4i 
36 21J 

39 ai 
39 26 
41 55 
44 32 
46 7 

49 22J 
51 59 

54 7 

55 54 
57 23 

59 46 
61 35 

63 oj 

64 10 

67 10 

76 46J 

77 2i 
77 24i 
n 34 
77 39 

77 54 

78 2j 

o / 

9 44 

19 20 

17 25J 

24 2 

21 41 

27 58 

25 23J 
24 oi 

22 I 
21 21 

20 36J 

18 49i 
18 i^ 
16 oi 
14 23 

13 3i 

" 57i 

10 13 

9 3ii 
8 23 

7 29i 
6 46 

4 52J 
6 28 

8 24} 

10 15 

11 21 

12 o 

14 27 

16 42^ 



o / 
24 43 

24 < 

25 57 
24 40 
23 53i 

8 54 
8 18 









09 39 

09 15 

08 55 

08 23 

07 59 
07 40 
06 52 
94 48J 

9« 43 
88 46^ 

87 I 
85 59 
82 3i 
78 24i 











No. Gdt. D. BravaU-Miller. 


; A*l/A 



! A/i/. 



*I25 21. 5. 26.16 



17 15 

77 3 

126 F'.n 4153 


78 4} 

18 7 

76 8 

*I27 J: 19.5. 24.14 

+ A'V 

; 78 5 

19 5 

74 34 

♦128 , II. 3.14.8 


1 78 4 

19 47 

73 26 

129 G: \ (X^) 7295 


78 3 

20 44 

71 53 J 

130 I/: X \ 3142 


' 77 49 

24 10 

66 I5i 

131 /: , 5273 1 


77 7i 

28 52i 

58 26i 



76 44 

30 49 

55 4 

*I33 1 


76 18 

32 35 

52 7 

*I34 ! 1 (*) I5.7f2.8 


76 4 

33 24i 

50 43 

135 A': , 7/ 2lSJ_ 


' 75 22 

35 36 

47 li 

♦136 ' 


, 75 6 

36 23 

45 40 

*I37 , ' i 25.13.P.12 


1 74 56 

36 53 

44 48 

138 L: 


74 42} , 

37 29} 

43 45 

*I39 c: 1 , 9-5.I4-4 


! 7^ ^* , 

39 6 

40 54 

140 M: ^ (7/,) 7.4. ly 


' 7340 ' 

40 4 

39 II 

*I4I I9.II.30.8 


' 73 28 

40 31 

38 23 



t 73 16J ; 

40 57 

37 38 

143 ' ^'' 

T . 5382 


i 72 54i 

41 46 

36 8J 

143 0: 

8.5. I3J 


1 72 12J 

43 13 

33 3^\ 

*I44 P' 

17.11. 28.6 


71 12 

44 28 


145 ^: JK 1 325i_ 


1 70 59 

45 32 

29 15J 

146 1 ^i , 19. 13.32.6 


: 70 26J 

46 29 

27 30 

147 , ^: io.7._i7.3 


' 69 57 1 

47 19 

25 56} 

♦148 s: 1 


' 69 30 , 

48 3 

24 32J 

149 ^ S: 1 1.8. 19.3 , 


' 69 5 

48 42i 

23 m 

*I50 7 : 1 


68 42i 

49 18 

22 9 

151 T: 
125. Sch 

f ' 4371 


68 21 

1 1 

49 50 

21 .7} 

norr, Neumark, 30. 

127. Sansoni, Andreasberg, II. 

128. Ces^ro, Engis, 16. 

132. D'Archiardi, Montecatini, 35. 

133. Schnorr, Neumark, 30. 

134. '* «* " 

136. Sansoni, Schapbachthal, 20. 

137. ♦* " " 

139. ** Andreasberg, II. 

141. ** Montecatini, 17. 

142. Schnorr, Neumark, 30. 

144. Palache, Lake Superior, 37. 

148. Sansoni, Andreasberg, 11. 


u it (( 








No. Gdt 

. D. 















51 9J 

18 32 





+ i?9 

66 42i 

52 II 

16 30 



1 "1 

6.5. II. I 

+ i^ii 

65 35i 

53 40 

13 32 


K : 

+ ^Y 

65 29 


13 I4J 

156 w 

: IV 


+ i^I2 

65 10 

54 »3 

12 25 




7.6. 13. 1 

+ ^13 

64 47i 

54 4oi 

II 28 


15. 13.28.2 

+ ^14 

64 28 

55 4 

10 39 

*i59 ! / 


17. 15.32.2 


63 56J 

55 42J 

9 I9i 

160 Y 

. 1 

9.8.17. 1 

+ ^17 

63 43 

55 58 

8 47 

*i6i 1 


10.9. 19. 1 

-f iVi9 

. 63 20 

56 24 




21.19 4^*2 

4- Rto 

63 II 


7 28 

152. Sansoni, Andreasberg, 1 1. 
155. Rogers, Colorado, 39. 
158. Stober, Framont. 25. 

159. Sansoni, Andreasberg, ii. 

161. D'Archiardi, Montecatini 35. 

162. Sansoni, Arendal, 22. 







. D. 








/ • 

59 21 






22 55i 

88° n 






56 26 

27 21 

89 40 





+ J^5 

51 45 

zz 50 

86 40 


c i 



51 53i 

38 18} 

80 4 




55 4 

35 54 

79 13 





62 37 

30 7 

77 35 




4. 10.14. 9 


68 6 

25 53 

76 46 





72 15 

22 40 

76 22 






78 4i 

18 7 

76 8 




81 58 

15 4i 

76 II 






12 54 

76 19 





88 26 


76 39 




94 28 

5 16 

77 28 




96 17 

12 13 

63 19 





93 54 

16 48J 

57 27 


\ B 


91 4ii 

20 40 

52 3oi 






87 50} 

26 44i 

44 4ii 



\ Ii 



80 i^ 

VI 30 

30 39 



1 3581 

— 2i?4 

75 30 

45 6 

23 I3i 

163. Moesz, Kdr5smez5, 34. 
170. Sanson! Andreasberg, 1 1. 

174. Irby gives as doubtful. Goldschmidt includes it in his list proper. 

175. ** ** ** 

176. Gonnard, Couzon, 31. 
VOL. XXII.— 30. 








Positive Scalenohedrons Not Included in the Foregoing 


1 1 


1 X 



No. ( 

adt. D. , Bravais-MUler. 








59 45 






+ i>py 

28 10 

92 49 




48 45 

17 16 

III 30 

*i84 ! 

roi («) ; 


+ i^y 

62 37 

35 23 

68 17 


j 1 

+ 4^13 


53 59 

21 I 

i86 ^1 1 1 I2.8.20.7 


+ iJfs 

67 1 

43 "i i 

46 7 

*i87 j 

b 1 1 20. II. 31. 15 

+ *^v 

65 s^ 

35 35 

61 8 


N \ ' 12.4. 16. 1 1 

+ A^2 

66 17 

21 36 

81 3 


Z , 16.4. 20. 15 

+ I^J 

. 71 10 

16 44 

86 41 


M 1 

+ t^3 

72 22^ 

34 20J 

55 20 


a ' 1 

+ f-ffV 



*192 1 

/: , 

+ I^V 

76 22J 

34 16 ' 

46 22i 


16. 10. 26. 5 

+ f^y 

73 4} 

43 41J 

29 19 

*I94 ! 



+ m 

77 30 


40 16^ 

*I95 , 

, ('^0 

19. 10. 29.6 

+ f^V 

77 44 

, 38 33i 

ZZ 28 


6^ 1 1 10 4.14.3 


+ 2>?* 

84 21 

31 16 

38 49 


T {FY 42^1 

+ 2X3 

80 ij 

' 37 30 

30 39 

♦198 ; 

T' I3-7-20.3 

+ 2i?y 


1 39 I7i 

27 14 


U! (Q) 1 

+ V^2 

88 48} 

2658 ' 

42 12 


6' ,(/-) 

+ i^2 

89 19J 

27 6i 

40 46i 


n: 1 i 

+ 3i?V 

108 30 

9 4. 

53 34 


1 (m) ' 5271 

+ 3^1 

88 51 

31 46J 

IZ 22 

*204 \ 




+ 3^V 

82 32 

36 10 

28 9 


ch, Cumberland 

, 33 

; also Moesz, 




Thtirling, Andreasberg, I 



Palache, Lake Superior, 



Johansson, Norberg, 24. 


Palache, Lake Superior, 



(( <( i< 



(( t( i( 



(( i< (( 



i( it << 



Observer not known. Goldschmidt includes it in his " 



Johansson, Norberg, 24. 


Sansoni, Blaton, 13 ; Cesiro, 

Rhisnes, 19. 


Ces^ro, Rhisnes, 19. 


Palache, Lake Superior, 



«( ^^ li 



i( t( (( 



Melczer, Budapest, 29. 



»iro, I 

Ihisnes, 19. 



Positive Scalenohedrons Not Included in the Foregoing 

Zones. — Continued. 





No. , Gdt. D. 










*205 , (n) 63.28.91. II 


84 II 

34 Zl 

28 58' 

*2o6 ' A {'i/^^) 





38 53 

*207 ! 

-f i^V 

92 46 



39 13 


15-4. 19.3 


95 2 



41 54 

*209 r 


-f 4i^» 

*2io A 

13. 1- 143 

-r 4^J 


*2II , 2 

' ^ 

+ 4^1 

♦212 e 

-f 4iVV 


*2I3 ' * 

9. 1. 10.2 

-f 4i?J 


*2I4 A 



+ 4/?f 


215 Si 

(jc^)\ 14.2. 16.3 

+ 4^J 

103 50J 



51 47 

216 33: 

{x) 5ibi 

+ 4^*1 

99 50 



46 41 

217 3! 

V 62b I 

+ 4^2 

91 3 



35 52 


(0 , 15.7-22.2 

4- 4^y 

219 ^\ 


+ 4^*3 

81 20 



24 21 


205. Mel 


4 4^V 

iczer, Budapest, 29. 

2c6. Sansoni, Andreasberg, II. 

207. " Freiberg, 27. 

208. ** ** ♦* 

209. Palache, Lake,Superior, 37. 

210. ** " '* *' 

211. •* ** *' ♦« 

212. ** ** ** '* 

213. ** ** " ** 

214. ** ** •* " 

218. Cesiiro, Rhisnes, 19. 




(( t( 



Negative Scalenohedrons Not Included in the Foregoing 















- i'i^4 


52 57 


31 52 

85 55 


a \ 

- i^i 

62 26 i 

34 12} 

69 7i 





- i^4 

64 43} 

37 28 

62 10 





- i^5 

65 25i 

42 14 

51 30 





- i^7 

65 24 

47 48i 

38 I 



, iii) 


- i^9 

65 44i 

50 51 




9. 1 1. 20.4 

, - l^»o 


51 ZZ 

27 7 



, - i^'3 

63 57 

53 58} 

21 I 




- 1^7 

67 4} 

48 57i 

39 35 





- i^i 

72 40 

29 25 

64 21 




4 8 12.5 

- J^3 

72 22} 

34 20} 

55 20 

• 232 


- »/"/ 

73 25i 

30 49 

54 38 


, (^) 

- ?^v 

70 39 

6 38 

79 0} 




- im 

34 35 




75 56 

14 8 





' -i?? 

78 3 

20 44 

71 53i 


1 -A>V 


21 45 

67 46 





— i?2 

77 49 

24 10 

66 I5i 






77 7i 

28 52} 

58 26i 



5. 11. 16.6 

1 " 

76 18 

32 35 

52 7 




1 231 


75 22 

35 36 

47 li 



1 -/-y 

73 40 

40 4 

39 II 



, (*) 


1 -JfS 

70 59 

45 32 

29 16 



, - mi 

79 38 

8 4 

50 23 




79 46 

22 10 

67 4 




1 - !^l 

81 38* 

18 48} 

70 24 

221. Palache, Lake Superior, 37. 
229. Sansoni, Andreasberg, 11. 

232. Schnorr, Neumark, 30. 

233. Ces^ro, Rhisnes, 16. 

234. Palache, Lake Superior, 37. 

235. Sansoni, Cto, 22. 

237. H5fer, Rauris, 23. 

238. Sansoni, Andreasberg, II. 
240. Palache, Lake Superior, 37. 
242. Gonnard, Couzon, 31. 

244. Palache, Lake Superior, 37. 

245. *« *« ** •« 

246. Sansoni, Andreasberg, ii. 



Negative Scalenohedrons not Included in the Foregoing 

Zones. — Continued. 



1 ^ 








1 A/H/a 


81 25 


1 khil 



20 29} 

67 47i 







24 53i 

60 54 





79 3» 

28 30 

55 " 




78 I 

31 31 

. 52 23 


y (C) 


82 so 

19 2 

68 26} 



I. II. 12.8 



6: (0) 



83 9\ 

21 51} 

62 52 


D : 


81 15 

30 25 

54 I 




79 44 

35 loj 

49 15 


m; {R') 


77 30 

36 28J 

40 16} 



- k^V 

72 38 

44 44 

26 53l 





85 loj 

19 29 

64 27} 




80 57 

30 12 

48 4 





31 30 

18 6 




81 52 

25 10 



»: {T) 


86 48 

17 34 

65 46 



4. 20. 24. 1 1 


66 39 

23 4li 

66 52 



1. 10. £7.6 


56 4 

8 4 

76 54i 


u; ; 


65 8} 

15 5 

68 48 





83 41 

28 58} 

46 56 


-Hm , 

84 3 

28 28 

47 29 



. -f^V 

91 I 

14 28} 

66 14^ 



-nm ; 


Q (-V) 

- j^f ■ 

93 17 

20 57 

69 8 


V (V) 





89 iij 

24 46 

43 21 


3.24.27 7 

- SA-? 

104 44J 

II 22 

54 I 


a- {N) 


96 26J 

53 58J 

21 I 





85 22 

26 37 

36 29} 


2: (^) 



103 6 

15 I 

47 59 





82 43 

24 14 

38 II 





74 32i 

16 52 

44 57J 


3i ,(«) 

1.9. 10. 1 


108 42 

10 21 

50 57 



— I2if j 64 24 

261. Gonnard, Couz 

7 54 
on, 31. 

52 9 


'. Thilrling, 

Andreasberg, 15. 


>. Palache, . 

Lake Superior, 37. 

263. Mclczer, Buda 

pest, 29. 


. Sansoni, i 

Andreasberg, li. 


<( (( 

f < 


5. Pal ache, 

Lake Superior, 37 

268. Schnorr, Neun 

lark 30. 


(( ^^ (( 

269. Ces 

kvOf Rhisne 

s, 19. 


(< t( (< 

271. Gonnard, Cou2 

:on, 31. 


». Sansoni, ^ 

Andreasberg, ii. 

272. Palache, Lake 

Superior, ^ 




(( <« 

277. Melczer, Buda] 

jest, 29. 


1. Observer 

not known. 


ft << 



►. GOE 


Gjuzon, 31. 

280. Sansoni, Freib 

€rg, 27. 



List of Uncertain Forms. 


Letter. Bravais-Miller. 




K' 1 5490 


Bournon, I. 



Des Cloizeaux substitutes J/*2 for this fonn . 




CesAro, Rhisnes, 19. 


1 9092 




14.0. 14.3 


Des Qoizeaux, 6. 


+ Y^ 

Dana, 5. 


, 13.0. 13-4 



^ 8083 


Penfield and Ford, Cayuga Sp'gs,N. Y. ,38. 







i 30.3.10 



0.1. 1. 10 

1 _ 



K ' OII4 




Bournon, I, 




Hausmann, 3. 


' 0667 


Brunlechner, Bleiberg, 26. 



1 — fiP 


1 0. 10. 10.7 




Morton, Oto, 12. 





0. 10. 10. 1 


. Sansoni, Andreasberg, 1 1. 


1 0.16. 16. 1 






Stober, Reichenweier, 25. 






—40 A', Rhisnes, 19. 




Gonnard, Couson, 31. 




Hauy, 2. 



' Gonnard, Couzon, 31. 


13. 2. 15.17 




+ ^H 




CesJiro, Engis, 16. 









— 2A"y 

Ces^o, Engis, 16. 


4. 10. 14.3 

— 2iP} 

Sansoni, Blaton, 13. 




for , '3.^9.32.3 


Sjogren, Gestrikland, 10. 


( 1 


H ll (< 




Morton, Arendal, 12. 




<< i( (< 




+ i^23 

Des Qoizeaux substitutes for -j-Ji?23. 


; 14.9.23. 15 






List of Uncertain Forms. — Continued. 








Des Cloizeaux substitutes +^-^23. 




Kemp, Port Henry, 21. 






Des Qoizeaux, 6. 








Ces^o, Rhisnes, 19. 




Ces^upo, Rhisnes, 19 (should be included 
in the list proper). 










+ ?>P3 



loi. 70.171. 34 


Ceslux), Rhisnes, 19. 






Sansoni, Blaton, 13. 



64.40. 104.21 


Ceskro, Rhisnes, 19. 



27. 14.41. 12 


<( << (( 


104. 13. U7. 187 


Kemp, Port Henry, 21. 





Ceskro, Rhisnes, 19. 




(i (i ti 




«i <( (1 



37.16.53 15 


(( (( (( 





(( <( *i 





+ V^H 

(< «< (( 




it H (< 








Ceskro, Rhisnes, 1 9. 


+ §i?2 

Des Qoizeaux substitutes for it -\-^R2. 




H <f «( 


1 2.4. 16.3 : 

+ 1^2 

** ** +fi^2. 





Cesiiro, Rhisnes, 19. 




Sansoni, Blaton, 13. 




Cesilro, Rhisnes, 19. 


Irby substitutes for H-y/^JJ. 


. 551. 171. 722.100 


for this +V^ff. 



15.4. 19.2 


Des Cloizeaux substitutes + lOj^f for this. 







14.4. 18. 1 


Des Cloizeaux substitutes for -j- V-^if* 



+ I2^V 

Cesiro, Rhisnes, 19. 





ti (( ti 




151. 154.305.27 


Miers, Egremont, 18. 





List of Uncertain Forms. — Continued. 







Sansoni, Andrea.sberg, 1 1. 


















1 ^ 


Des Cloizeaux substitutes for a form given 
by Levy. See Irby, 1. c, p. 57. 








I4. 26.40. 21 











, ;r' 

611 17.7 


Sansoni« Blaton, 13. 



3.8. 1 1.7 


Ces^ro, Rhisnes, 19. 



— JiP2 

Sansoni, Andreasberg, II. 





Sansoni, Andreasberg, li. 



Zippe substitutes for this — f^V — 

10 1 







7. 17. 24. 1 1 





1 p 


Sansoni, Blaton, 13. 










Thllrling, Andreasberg, 15. 




Des Cloizeaux substitutes — ^2 for this. 



6- 33- 39. 26 


Sansoni Andreasberg, 11. Ces^ro sub- 
stitutes — y ^§f for it. 




Cesiro substitutes for — |J^y. 








Des Cloizeaux substitutes for this — I^V . 





Sansoni, Andreasberg, ii. 





Sansoni, Andreasberg, II. Ces^ut) substi- 
tutes — |fi^f J for it. 


Ces^ro substitutes for — Ji^V- 



10.16.26 5 


Sansoni, Andreasberg, II. 




Sansoni, Blaton, 13. Ces^Ut) substitutes 
— Ji^y for this. 





Zippe substitutes for — |^y. 





1 r 


Ces&ro substitutes for — H^if* 




CesAro, Rhjsnes, 19. 


: i 


Dana, Bergen Hill, 5. 





List of Uncertain Forms. — Continued. 








- H^V 1 


- I^«, 

Zippe, given by Des Qoizeaux. 



- \m\ 

Des Qoizeaux substitutes this for — ^^|. 




r «^i 


1 "^ 


- \m 



5.i3»8 5 

- \f^\ 

Des Qoizeaux, 6. 



- %Rz 

Ces^ro, Rhisnes, 19. 



- \R\\ 

Des Qoizeauz, 6. 



- "^m 

Rath, Lancashire, 9. 



- VA'H 

Ces&ro, Rhisnes, 1 9. 



- 3^V 

<( it (i 



- V^f5 



- 6A'J 

Des Qoizeaux, 6. 



16.24.40. 1 

— 8ie5 



- loA't 

Sansoni, Andreasberg, 11. 


1 ft 

1. 16.17. 1 

- ^SR\l 

Sansoni, Andreasberg, 1 1. 



- 29i^H 

Des Qoizeaux substitutes this for 




Des Qoizeaux substitutes — 29^}^ and 
Ces^o — 66i^JJ for this. 


- 66A»e 

Cesiro substitutes this for — ^\^R\\. 



7. 168. 175. 1 



The abbreviations used are those given in Dana's 6th edition, System 
of Mineralogy. The first reference given is that of the abstract (in some 
cases the original article) of the article in the Zeit, /. Kryst. u, Min, 

1. Boumon. Traits de chaux carbon. 1808. 

2. Haiiy. Traits de Mineralogie 2"* Ed. Paris, 1822. 

3. Naumann. Fogg. Ann. 14: 235. 1828. 

4. Hausmann. Handbuch der Mineralogie, 2d edition. 1847. 

5. Dana. System of Mineralogy, 5th edition. 1868. 

6. Des Cloizeaux. Manuel de Mineralogie, Vol. 2. 1874. 

7. Bodewig. Groth's Min. Samm. Univ. Strass., p. 124. 1878. 

8. Zepharovich. 5: 269. Jahresber d. Ver. Lotos. 1878. 

9. Rath. 6: 540. Ber. Med. Ges , p. 28. 188 1. 
10. Sjogren. 8: 651. G. For. Forh., 6: 550. 1883. 

Att. Ace. Line. Mem. (3), 19: 450. 1884. 
Kongl. Sc. Vet. Akad. Forhandl., 8 : 65. 

Bull. Ac. Belg. (5), 9: 287. 1885. 
Vh. G. Reichs., p. 149. 1885. 


Sansoni. 10: 545. 


Morton. 11 : 319. 



Sansoni. 11 : 353. 


Foullon. 12 : 526. 


15. Thiirling. 15: 413. Jb. Min. Beil.-Bd., 4: 327. 1885. 

16. Cesaro. 13: 431. Mem. Acad. Belg., 38 : i. 1886. 

17. Sansoni. 18: 81. An. Ace. Torino, 23. 1888. 

18. Miers. 19: 904. Min. .,8: 149. 1889. 

19. Cesaro. ao: 283. Ann. Soc. G. Belg. M^m., 16: 165. 1889. 

20. Sansoni. 19: 321. Giorn. Min., i : 299. 1890. 

21. Kemp. 20: 416. Am. J. Sc., 40: 62. 1890. 

22. Sansoni. 20: 597. Giorn. Min., i: 137. 1890. 

23. Hofer. 24: 431. Min. Mitth., 12: 487. 1891. 

24. Johansson. 24: 138. G. For. Forh.,'41 : 49. 1892. 

25. Stober. 24: 629. Abhand. z. geol. Spec. Karte von Els. Loth. 

5: I. 1892. 

26. Brunlechner. 25: 432. Jahr. d. nat. hist. Museums Klagenfurt. 

22. 1893. 

27. Sansoni. 23: 452. Giom. Min. 5 : 72. 1894. 

28. Hobbs. 25: 527. Bull. Univ. Wisconsin, Sci. Ser. i: 115. 


29. Melczer. 30: 182. Foldt. Kozl. 26: 79. 1896. 

30. Schnorr 30 : 660. Wissensch. Beilage z. Programm des Real- 

gymnasiums zu Zwickau. 16 ss. 1896. 

31. Gonnard. 31: 69. C. R. 122, 348. 1896. Bull. Soc. Min., 

20: 18, 330. 1897. 

32. Cesaro. 31: 89. Mem. Acad. Belg. 53: 68. 1897. 

33. Butgenbach. 31: 183. Ann. Soc. G. Belg. 24: 66. 1897. 

34. Moesz. 31: 318. Foldt. Kozl. 27: 495. 1897. 

35. D'Archiardi. 31: 403. Att. Ace. Tosc, Proc. Verb. May 9. 


36. Farrington. Publications Field Columbian Museum, Geol. Series. 

I : 232. 1900. 

37. Palache. Geol. Surv. Mich. 6: 161. 1900. 

38. Peniield and Ford. Am. J. Sc. (4) 10 : 237. 1900. 

39. Rogers. Am. J. Sc. (4) 12. 43, 44. 1901. 


New York, May 23, 1901. 





By henry S. MUNROE 

Further experience in the laboratory with the classifier (Fig. i) 
described in the April number of the School of Mines Quar- 

FlG. I. Scale, }i^^=i^r 

Fig. 2. ys^^ = i^^. 

TERLY has developed the necessity for the change in design shown 
in the accompanying cut (Fig. 2). 

In the original form the material to be classified was fed through 
a stop-cock funnel. Experience has shown that this method of 


feeding limits seriously the capacity and efficiency of the apparatus. 
When the feed is rapid the weight of the column of sand and water 
in the tube c increases, causing the water to rise in the funnel tube 
until equilibrium is restored. The rising of the water in the funnel 
tube in turn checks temporarily the feed of ore and the column c 
thus becomes lighter causing a rapid downflow of water in the fun- 
nel tube, and an increased feed of ore until an equilibrium is again 
reached. In the meantime the large amount of ore thus fed into 
the classifier finds its way into the tube c, increasing once more 
the weight of that column, which produces a back current in the 
funnel tube as before. The classifier thus becomes intermittent 
and irregular in its action, at least so far as the upper part of the 
tube c is concerned. The lower part of the tube d is not affected. 

Further with fine material the back pressure in the funnel tube 
tends to arrest the flow of ore and choke the tube, and makes it 
difficult to restore the feed, continual clearing of the funnel tube 
with a long wire being necessary. 

These difficulties have been overcome in the new design (Fig. 
2). The funnel has been discarded and the second Erlenmeyer 
flask is now connected directly with the apparatus by a short 
length of rubber tubing of large diameter. A short brass sleeve 
inside the rubber tube makes a tight joint within the mouth of the 
flask, while at the lower end of the tube is large enough to slip 
over the mouth of the bulb tube d as shown. A screw pinch- 
cock permits the flow of sand to be controlled and adjusted. 

As the sand flows out of the upper flask water ascends through 
the connecting tube to take its place. This continuous ascending 
current keeps the sand loose and tends to promote a uniform feed. 
The feed is not affected in any way by the density of the column 
in c. If the material becomes compact or the feed is choked 
water may be pumped into the flask and the flow restored by 
gently squeezing the rubber tube. By pinching the lower end 
tightly with the thumb and fingers of one hand, before squeezing 
it above with the other, the sand may be loosened even when quite 
solidly packed. To permit this pumping of the water there must 
be some air in the top of the flask, which can be admitted, if 
necessary, through the side inlet. 


A CORRECTION — Thomas Egleston Memorial. 

In the article upon the late Professor Egleston in this Journal, Vol. 
XXL, p. 197, there appears the following quotation from a letter 
written by a fellow student at the School of Mines in Paris. 

** He did not take a regular course but followed certain lectures 
only — notably those of Professor de Senarmont on mineralogy, of 
Elie de Beaumont on geology and of Professor Bayle on palaeontol- 
ogy. His specialties from first to last, were mineralogy (notably 
on the crystallographic side) and mining. The late General Vinton 
was a classmate for, I should think, less than a year. Neither he nor 
Professor Egleston graduated ; they took only an eclectic course, 
with government permission to attend lectures in given depart- 

From papers recently submitted by the family of Professor Egle- 
ston it appears that he pursued the entire course of the school from 
1856 to i860, was ranked second among the foreign students and 
received at the completion of his course the certificate then granted 
to foreign students signed by the director (de Senarmont) ; this cer- 
tificate being the equivalent of the " brevet superieur " established 

in 1890. 

Alfred J. Moses. 





Metallic Sodium as a Blowpipe Reagent, Parsons (y. Am, Chem. 
Soc, XXIIL, 159). A small piece of sodium hammered flat is covered 
with a ]x»vdered oxide or a mineral to be tested, and the combination 
kneaded into a pellet with the knife blade. Then on placing the pellet 
on charcoal and applying the flame of a match, a momentary flash en- 
sues, and reduction is accomplished. By application of the blowpi{>e 
flame, the coatings, etc., characteristic of the metals present may readily 
be obtained. 

Preparing Normal^ Semi-rormal, etc.. Acid Solutions, Meade {J, Am, 
Chfm, SoCi XXIIL, 12). The method consists in weighing out the 
requisite amount of pure crystallized copper sulphate, and electrolyzing, 
which leaves in the solution the exact correspoading amount of H3SO4. 
The Cu is deposited on a cathode of copper foil. For tenth normal 
H2SO4 weigh out 12.487 gms. pure CuSO^. yHjO, dissolve in 750 cc. 
of water and electrolyze with a current of 1.5 amp. for about 8 hours. 
Transfer to a liter flask, and with the washings of the beaker and foil, 
bring the volume to i liter. For a normal solution, take 124.87 gms. of 
the copper salt and electrolyze for 12 to 18 hours with a current of 2.5 
amp., the other details being as before. Rigid tests proved the solutions 
to be absolutely accurate. 

Ferro-Silicon and Silico Spiegel, Ibbotson and Brearley (C N.^ 
LXXXIL, 269). Total Carbon is best determined by direct ignition of 
the powdered sample in a stream of O, Graphite by treating 2 to 3 gms. 
of the sample with 70 to 100 cc. HNO3 (Gr. 1.2) heating nearly to 
boiling, and adding HF cautiously (a few drops at a time) until the ma- 
terial is entirely decomposed. Filter on asbestos, wash with NaOH, 
then with HCl followed by water, and burn in a current of O. 

Silicon, Contrary to the usual statements these alloys can be attacked 
successfully by a mixture of 50 cc. HCl with 10 to 20 cc. HNOg (on 2 
gms.). Boil to decomposition, add about twice the volume of water, and 
filer at 01 ce^ wash with dilute HCl, ignite and weigh — allow o.i per 
cent, for SiOg which has gone into solution. Boiling after dilution causes 
more Si02 to dissolve. 

Manganese. Dissolve in HNO3 with aid of HF as above, and apply 
Reddrop's (Mignot) method (sodium bismuthate, H^Oj and colorimetric 
comparison) (vide Quarterly, XXII. , 94). 

Tungsten in Steels. Bagley and Brearley (^Ch, N., LXXXIL, 270). 
Schoffels method consists in treating the substance with neutral 2NH4CI, 
CuCIg, and determining W in the insoluble portion. If CI is present 
this method is not applicable without elaborate manipulation. By add- 
ing one third its volume of cone. HCl to the solution used this complica- 
tion is obviated without loss of W (vide Quarterly, XXL, 282, 283). 


Separation of Tungsric Acid. Ruegenberg and Smith {Jour. Chetn. 
Soct XXII. , 772). Digestion with HjSO^ of sp. gr. 1.378 was found 
to dissolve M0O3 or FejOg completely from their mixtures with WO3. 

Molybdenum in irons. — Dohler ( CA^w . Z/., 1900, 537). 10 gms. of 
steel or more of pig iron are dissolved in 100 cc. HNOg and evaporated 
to dryness, to separate SiOj. Take up with about 100 cc. IlCl and evap- 
orate to about 10 CO., adding water from time to time to carry off excess 
of HCl etc. Dilute, filter and bring the bulk of the solution to a liter, 
warm to 80° and pass HjS for an hour. Allow to stand 12 hours, 
then filter, and wash with cold HjS solution acidified with HCl. 
Make sure as to complete precipitation by evaporating the filtrate about 
one half, and passing HjS. Extract the precioitate with warm (NH^^jS 
solution. Bring this solution of M0S2 "P ^^ boiling heat and precipitate 
by dilute HCl. Heat until no odor of HgS in perceptible, filter through 
a weighed filter, dry at 120° to constant weight. This precipitate 
contains S ; a portion of th s precipitate well pjlverized is heated to 
rednes-s in a current of H until its weight is constant. This affords pure 
M0S2 (containing 60 per cent. Mo) from which the entire amount of 
Mo may be calculated. The author claims that although tedious, this 
method is the most trustworthy of any yet described. 

Analysis of Uranium and Vanadium Ores. Fritchle {Eng. and Min. 
Jour,^ Nov. )o, 1901). Digest in a flask for about an hour, at a gentle 
boil, 0.5 gm. of the well pulverized ore in 10 cc. of HNOj, then dilute 
with 10 cc. water, and neutralize with saturated solution of NagCOj, 
finally adding 5 cc. in excess ; then 20 cc. of a 20-per-cent. solution of 
NaOH. Boil slowly for half an hour, filter off, and wash with NaOH 
solution. Vd goes into solution, Fe and U remain in the precipitate. 
Dissolve the precipitate in 20 cc. of hot HNO3 (^ • dilute with 40 cc. 
water, add NH4OH until a slight permanent precipitate forms, then add 
40 cc. of a freshly prepared saturated solution of (NH4)2C08. Heat 10 
just short of boiling for a few minutes to dissolve U, leaving Fe, filter and 
wash. Convert the U to sulphate by evaporating with HjSO^ to fumes, 
dilute, reduce with Al foil, and titrate with standard KjMnjO^. De- 
termine Fe by obtaining in H^SO^ solution, reducing and titrating in 
the same way. 

Determine Fe by obtaining H2SO4 solution, reducing and titrating in 
the same way. 

The solution of Vd mentioned above might be used, by conversion to 
sulphate, reducing with Al, and titrating with K2Mn20g, but on account 
of the amount of salts it contains this method is less convenient than 
treating a separate portion of the ore, converting to sulphates by evapora- 
tion with H2SO4, reducing and titrating all three together, determining 
the Vd by deducting the K2Mn208 attributable to U and to Fe. 

Vd and U should be treated in hot solution. The end reaction for Vd 
is a little slow. 

Fe standard x 120/56 = U. 

Fe standard x 51.C/112 = Vd. 

Delicate Reaction for Manganese. Marshall (C W., LXXXIII., 76). 
A sulphuric or nitric solution containing Mn shows the permanganate 
coloration distinctly if potassium persulphate (K2S2O8) and a drop of 
AgNOg is added. The production of the color is uncertain unless a 


minute amount of Ag sale is present. The method might be utibzed 
quantitatively for coloriraetric determinations. 

Manganese as Phosphate, Bottger {Berichte, XXXIII., 1019). By 
adding 5 to 10 times as much NH4CI as will suffice to form a double 
chloride with the Mn present, and a slight excess of ammonia, the pre- 
cipitate readily comes crystalline on heatinii:. It should be washed with 
cold water containing minute amounts of ammonia to avoid any re-solu- 

Manganese or Cobalt as Phosphate, Dakin {Zts, f. Anal. Chem,, 
XXXIX , 784). Mn or Co may be accurately estimated as phosphates 
if (i) sufficient excess of (NH4)2HP04 is added (about ten times the 
weight of the metal present), and (2) the precipitate thoroughly washed 
with a i-per-cenr. solution of (NH4)2HP04. The latter salt is finally 
removed by one or two washings with dilute alcohol. Of course the pre- 
cipiiates should be crystalline before filtering. They retain one molecule 
of water when dried at loo*^ to 105°, but the determination is accurate, 
whether weighed as MnNH^PO^.HjO or CoNH^PO^.HjO or when 
ignited and weighed as the pyrophosphates. 

Qualitative for Nickel in Presence of Cobalt, Browning and Hart- 
well {^Am. J, Sct.f X., October, 1900). Modification of Clarke's method 
with ferricyanide and ammonia. 

Dissolve not over o. i gm. of the salts in about 5 cc. of water, add a 
little of saturated solution of alum, neutralize free acid by ammonia, and 
then make faintly acid with acetic. To this add about 0.5 gm. ferricya- 
nide crystals, shake to dissolve, and to effect the reaction with the Ni and 
Co ; then add 5 cc. strong ammonia and filter. To the filtrate (which 
should have no reddish color) add a piece of NaOH or KOH about the 
size of a pea and boil. The appearance of a dark coloration or of a 
black precipitate indicates nickel. The test was found to be very delicate. 

Arsenic — Marsh test, Estcourt ( Ci^. N,^ LXXXII., 287). It is noted 
that the presence of small amounts of SO 2 prevent the deposition of the 
As mirror in t-^e tube through which the H is passing, when heated by 
the flame. This fact served to completely mislead the author when 
making some tests for arsenic in beer. HgS, which may be produced in 
small amounts by the action of HjSO^ on Zn, may also produce a sim- 
ilar effect (ib. 295). A. H. Allen suggests the addition of a little Br water 
to destroy sulphites, and subsequently adding a little CujCl^ to reduce 
the As to arseniouj. In the report of commission of Manchester Brew- 
ers Association the Reinsch test is preferred. 

Arsenic — Marsh Test, — Effects of Selenium and Tellurium, Beny [J, 
S, C. /., XX., 322). Se and Te may occur in the substances tested in which 
commercial sulphuric acid has been used to prepare them. Se especially 
which simulates As in the tube deposit and in the spots on porcelain, 
though its color is usually more reddish than that of As. The interposi- 
tion in the train of a bit of cotton wool wetted with lead solution was 
found to absorb the HjSe and HjTe. 

Lead in Potable Waters, Bellocq (/. Pharm, Chim, [6], XIII., No. 2). 
Add to I or 2 liters of the water 5 to 10 cc. cf NH^OH solution of ZnO, 
and leave quiet for some hours. Decant the clear solution, and filter the 
remainder ; when dried, the precipitate is readily detached from the filter. 


Treat it with HC2H8O2 containing a little NH^CgHjO^ and test the 
solution with a drop of KjCjO^. 

Estimation of Mercurous Mercury, Peters {Am. J, Sci,y IX., 401). A 
volumetric process which affords accurate results consists in precipitating 
the nitrate solution cold with a slight excess of standardized (NH4)2C204 
and titrating the excess of oxalate in the filtrate. The precipitate may be 
used for gravimetric estimation if thoroughly dried over HjSO^. 2 to 
5 per cent of HNOg (Gr. 1.15) in the solution does not interfere with 
the determination of mercurous in presence of mercuric. The presence of 
12 to 20 per cent of the mercury in mercuric form does not materially in- 
terfere with the accuracy of the mercurous estimation. 

Nitric Acid in Combination with Alkalies, Perman (CA. iV., 
LXXXIII., 193). An alkaline nitrate when heated with a sulphate of 
which the base is other than an alkali, will lose NjOg completely. The 
presence of the red fumes will serve as a qualitatitive test. A quantita- 
tive determination of N2O5 in alkaline nitrate has been found practicable 
by mixing with anhydrous alum and heating gently — finally to dull red- 
ness and weighing. The heat applied must not be too strong or the 
AI2 (804)3 "^^y ^ decomposed. Presence of chlorides vitiates the results. 

Oxygen in Commercial Copper. Lucas {Bull. Soc. Chim. [3], XXIII., 
No. 24). This method consists in melting the Cu with Sn in the electric 
furnace in a current of CO, collecting the CO 2 and weighing. 

Dissolved Oxygen in Water. Letts and Blake {Chem, News^ 
LXXXII. 163). A separator funnel (globe stopcock) is filled completely 
with the water to be examined. Then withdraw a measured amount, 
and add a measured volume of standard FeSO^ solution (delivered in 
the bottom). Then add above, ammonia to fill the globe completely, 
insert the stopper, and mix by reversing the globe once or twice. Let 
stand 15 minutes. Then invert the globe and fill the tube (now upper- 
most) with a mixture of H2SO4 and water 1:1. On opening the stop- 
cock the acid flows in, and in the course of a few minutes dissolves the 
precipitated iron hydrates. Then run the solution into a porcelain dish 
and titrate excess of ferrous salt remaining, by standard H2Mn20g or 
K2Cr207. The latter gives the best results with seawater and sewage 
effluents. Convenient amounts to use are about 325 cc. of the water, 5 
cc. of FeSO^ (about 48 gms. per liter), and 2 cc. strong ammonia, the 
total capacity of the globe being about 332 cc. 

Water which Attacks Lead. Ackroyd (Chem, News, LXXXII., 162) 
notes that moorland waters which show an acidity equivalent to 0.5 part 
sulphuric acid per hundred thousand dissolve enough lead from the pipes 
in an ordinary house to cause plumbism. The acidity, due to carbonic 
and peaty acids, is determined by titrating 100 cc. of the water with 
N/100 alkali, using phenolphthalein as indicator. 

VOL. XXII.— 31. 



Liens — Attorney's Fees — Mines and Minerals. 

Where the evidence failed to show that a mine was being worked un- 
der a contract by which C. was in any way liable for the labor performed 
or material furnished therefor, but clearly established that it was being 
operated by two others under a recorded lease, id which C. had no inter- 
est, and such labor, etc., were furnished for their exclusive use and bene- 
fit, C.*s interest in the mine was not subject to a lien therefor, though he 
owned a one- fourth ioteresc in the mine, and was cognizant of the work, 
and to fcome extent, while employed as bookkeeper for the lessees, took 
part in its direction in their absence, and gave orders for merchandise in 
their name, but failed to notify the parties dealing with them of his true 
relation. Davidson et aL vs. Jennings et aL^ 60 Pac. Rep. (Colo.), 354. 

State Land Board — Lease of Coal Land — Renewal — Powers 

OF Board. 

Sess. Laws 1887, p. 328 (2 Mills' Ann. St. § 3634), gives the state land 
board full power to lease any portion of the state land at a specified rental 
on the valuation to be fixed, subject to a provision giving them unlimited 
power as to the terms of leasing of coal land, or the time for which the 
demise shall run ; and Sess. Laws 1895, c. 87 (3 Mills' Ann. St. § 3636), 
provides that no lease of state land shall be for a longer term than 10 
years. Held, that the latter prohibition did not apply to a lease of coal 
land, so as to effect the power of the board under the former statute. 

Sess. Laws 1895, c. 87 (3 Mills' Ann. St. § 3636), provides that no 
lease of state land shall be for a longer term than 10 years, and that a les- 
see may renew the same in the following manner : At any time within 30 
days preceding expiration, he shall notify the register of his desire, and, ' 
if the state board and he agree as to the value of the land, a new lease 
may issue, bearing even date with the expiration of the old one, and on 
like conditions ; provided, that the former valuation shall not be decreased 
without the parties* consent, and that the board shall not be prohibited 
from leasing any of the state lands to such parties as shall secure the great- 
est annual revenue. Held, that the provision that when a lease expires 
the holder may renew at any time withm 30 days next preceding, if he 
notifies the register of his desire, was not a limitation on the powers 
granted the board, which may or may not renew the lease, whether the 
lease be or be not filed within such time, but a restriction or an obligation 
on the lessee to file his petition within such time, and which would de- 
prive him of any equitable right to insist on a renewal if he failed to do so. 

None of the limitations provided for by such statute affect the |X)wer 
of the board to lease coal land or such time, and for the payment of 
such royalty, as it may determine, as specifically provided by Sess. Laws 
1887, p. 328 (2 Mills' Ann. St. § 3634). 

A mining company leased a section of coal land from the state, ad- 
joining its mines, and prosecuted its work with reference to its ultimate 
development. It had run its up:er levels to strike the upper vein on the 

* Prepared for The School of Mines Quarterly by Andrews & Murdoch, Berrien 
Springs, Michigan. 


leased land in question, but it was found to be too thin to be worked 
profitably. The working on the second vein, which was much thicker, 
and of good quality, was being extended in the direction of the land 
when its lease was about to expire. Thereupon it made application to 
the state land board for a renewal, and on its suggestion surrendered the 
original lease ; and the board proceeded, under ths statute, and in the 
full exercise of its authority, to consider whether the new lease should be 
granted, and on what te^ms, and ordered that on the surrender of the old 
lease a new one should be granted for i o years, at a minimum royalty of 
1 200 ; otherwise, on the usual terms — their record reciting the number of 
the old lea«e. The company paid its half year's royalty by transmitting 
its check for the amount, with the recording fees, which was accepted by 
the proper officer of the board, and never returned or offered to be re- 
turned until several months afterwards, ffeidy that the contract for the 
lease was complete, and the rights of the parties vested thereunder, so as 
to entitle the company to call on the board for the execution of a written 
lease as evidence of the contract into which it had entered. 

The affidavit cf the general manager of a mining company in support 
of its application for the renewal of its lease of a section of state coal 
land, adjoining its mining properly, stating his belief that, if there was 
any workable coal, it was impracticable to work it from the surface at any 
point of the section, but only from adjacent ground, is not a statement of 
an absolute fact, on which the land board had a right to rely in determin- 
ing whether it would extend the lease or grant a new one, but an opin- 
ion merely, affording no evidence whereon a claim of fraud could be 
predicated as a ground for refusing to execute the lease agreed on. 

Colorado Fuel & Iron Co. vs, Adams et ai,, 60 Pac. Rep. (Colo.) 367. 

Oil Leases — Construction — Rights Granted — Right of Forfeit- 
ure — Effect of Penalty for Nonperformance of Condition. 

By a course of decision in West Virginia which has established a rule 
of property, it is settled that an oil and gas lease in which the so e com- 
pensation to the lessor is a share of the product is not a grant of property 
in the oil or in the land until oil is actually produced, but merely of the 
right of possession for the purpose of exploration and development ; and 
there is always an implied, if not an expressed, covenant for diligent 
search and operation. 

A different rule of construction obtains as to oil and gas leases from 
that applied to ordinary leases or to other mining leases, and owing to 
the peculiar nature of the mineral, and the danger of loss to the owner 
from drainage by surrounding wells, such leases are construed most 
strongly in favor of the lessor. 

Where an oil and gas lease by which the lessor is to be compensated 
solely by a share of the product contains a proviso requiring the lessee to 
commence and complete a well on the property within a specified time, 
such proviso and Ihe time of its performance are of the essence of the 
contract, and it constitutes a condition precedent to the vesting of any 
estate in the lessee, without regard to the grammatical construction of 
the instrument. When the lessee makes no attempt to comply with such 
provision, aad evidences no intention to do so, at the expiration of the 
time stipulated the lease becomes forfeitable, at the option of the lessor, 


although by its terms it is for a definite term of years ; and, being in 
possession, the exercise of such option is sufficiently evidenced by the 
lessor's execution of a new lease to another party. 

A provision in an oil and gas lease, by the terras of which the lessor is 
to be compensated solely by a share of the product, that, in case of the 
failure of the lessee to comply with a condition requiring him to complete 
a well on the property within a stipulated time, he shall pay a forfeiture 
of I50, must be construed as providing a penaUy intended to secure the 
performance of such condition, and not as an alternative condition \ and 
where the lessee makes no attempt to fulfill the condition, and has no in- 
tention of doing so, he cannot, by a tender of the penalty, retain the lease 
in force until the expiration of its term, and thus secure an option on the 
property for speculative purposes. When, by his failure to comply with 
the condition, further performance of the contract becomes optional on 
his part, it is also optional on the part of the lessor. 

Huggins et al, vs, Daley, 99 Fed. Rep. (U. S.) 606. 

Master and Servant— Master's Violation of Statutory 


Hurd's Rev. St. 1889, c. 93, § 6, provides that a sufficient light shall 
be furnished at the top of every shaft of a mine to insure the safety of 
persons getting on or off the cage ; and section 8 requires that the top of 
every shaft shall be securely fenced by gates properly protecting such 
shaft ; and section 14 provides that any injury arising from failure to 
comply with the provisions of the act shall give a cause of action. Held^ 
that where a mine owner had erected above the opening of a shaft an un- 
inclosed frameworit of timber, supporting a structure called a ** tipple 
house,'* to which the cage could l3e hoisted, and through which the day 
shift entered and left the shaft, but it was customary to let off the night 
shift at the surface of the ground, and plauniff's intestate, being one of 
the night shift, and endeavoring to alight, fell into the sha t and was 
killed (it being dark at the surface, and there being then no fence there), 
it was proper to submit to the jury the questions whether the surface or 
the tipple house was the top of the shaft, within the meaning of section 8, 
and whether the proximate cause of the injury was the absence of a light. 

When an employe of a mine, endeavc rmg to alight from the cage at 
the surface opening of the shaft, fell into the shaft and was killed, the 
opening of the shaft being enveloped in darkness by reason of no suffi- 
cient light being furnished, as required by Kurd's Rev. St. 1889, c. 93, 
§ 6, contributory negligence of the deceased could be no defense to the 
mine owner in an action for damages for deceased's death, under chapter 
93, § 14, providing that any injury from failure to comply with the pro- 
visions of the act shall give a cause for action. 

Kurd's Rev. St. 1889, c. 93, § 6, provides that a sufficient light shall 
be furnished at the top of every shaft of a mine to insure the safety of 
persons getting on or off the cage ; section 8 requires the top of every 
shaft to be securely fenced, to protect the entrance to the shaft ; and sec- 
tion 14 declares that, for any injury to person or property by willful 
failure to comply with the provisions of the act, a cause of action shall 
accrue. Held that, where a declaration against a mining company alleged 
that plaintiffs intestate came to his death by defendant's willful omission 


to comply with sections 6 and 8, it was proper to refuse to allow officers 
of defendant to testify that they intended to comply with the statute in 
good faith, since the word ** willful," as employed in the declaration, did 
not involve a charge of wrongful intent, but only that the omissions were 
conscious acts, of the mind, and not from mere inadvertence. 
Odin Coal Co. vs, Denman, 57 N. E. Rep. (Ill.)> ^92. 

Mines and Mining — Construction of Lease — "Mine Run*' Coal. 

A mining lease stipulating for the payment to the lessor of a royalty of 
five cents per ton of 2,200 pounds on all lump coal mined, and five cents 
per ton on all ** mine run '* coal of 3,000 pounds mined, does not entitle 
the lessor to a royalty on screenings, consisting of nut and slack coal left 
after the lump coal is separated therefrom, as the words **mine run" 
coal mean the coal as it comes from the mines. Hardin vs. Thompson, 
57 S. W. Rep. (Ky.) 12. 

Mining Claims — Mistake in LcJcation — Extralateral Rights. 

Where by mistake a mining claim is located across, instead of along 
the vein passing through the location point, the rights of the locator are 
governed by the facts as they exist in regard to such vein. His side 
lines, as located, become end lines, and he is not entitled to any extra- 
lateral rights thereunder, although another vein, extending transversely 
to the one intended to be located, may have its apex inside of such sur- 
face lines. Cosmopolitan Min. Co. vs. Fooie et al.y loi Fed. Rep. 
(U. S.), 518. 

Coal Leases — Liability for Assessment — ** Reprises.'* 

The lessee of coal lands is liable for an assessment for sewer, his cove- 
nant being to pay a stipulated rent ** clear of and above all taxes and re- 
prises," the word "reprises** meaning such deductions for burdens inci- 
dent to the land as such as are required to be made from gross income in 
order to ascertain the clear or net profit. President, etc., of Delaware & 
H. Canal Co. vs. Yon Storch, 46 At. Rep. (Pa.), 375. 

Mines and Mining — Certificates of Location — Description — 


Gen. St. § 2400, allows the filing of an additional certificate of the 
location of a mining claim where the locator apprehends that his original 
certificate is defective, or desires to change his surface boundaries, or take 
in part of an abandoned claim. H^idy that where an original certificate 
sufficiently located a discovery shaft, and an additional certificate gave a 
description by meters and bounds starting with the discovery shaft, and 
running to posts in a mound of stone, the claim could be identified with 
certainty from both ceitificates taken together, and hence such certificates 
were admissible to establish the locator's claim, though neither, taken 
alone, sufficiently described such claim. 

A description in a certificate of the location of a milling claim which 
locates the mine a certain distance from a patented mine is sufficient, 
since such patented mine will be presumed to be a natural object or per- 
manent monument, until the contrary appears. 


A description in a certificate of the location of a mining claim that the 
mine is located in a certain section on top of the mountain south of Dew 
Drop gulch is sufficient, since such gulch is a natural object or permanent 
monument. Duncan vs, Fulton e^ a/., 6i Pac. Rep. (Colo.), 244. 

Mine Explosion — Duty as to Ventilation of Mine and Keeping it 
Clear of Standing Gas — Effect of Act of Congress. 

The place of an explosion in a mine, and its cause, and what, if any, 
negligence the owner is guilty of, are questions for the jury, when the evi- 
dence offered requires their submission to the jury. 

Instructions as to the duty of a mine owner with respect to ventilation 
of the mine and keeping it clear from standing gas are erroneous, when 
they are so inconsistent with other instructions that they tend to confusion 
and misapprehension, and when they make his duty relative instead of 
absolute, as required by the act of Congress of March 3, 1891, making 
the test what a reasonable person would do, instead of the command of 
the statute. 

The duty of a mine owner as to ventilation of his mine and keeping it 
clear of standing gas is made imperative by the act of Congress of March 
3, 1 89 1, and the consequence of neglecting it cannot be excused because 
some workmen may disregard instructions. Deserant vs. Cerillos Coal 
Railroad Company, 20 Sup. Ct. Rep. (U. S.), 967. 

Trespass to Mining Claim — Quartz Claim — Quantity of Ore Nec- 
essary — Estoppel. 

Under Rev. St. U. S. § 2320, requiring the discovery of a vein or lode 
within a quartz claim before any right can be acquired thereto, it is 
enough to entitle the discoverer to protect his mining rights if ore or 
metalliferous rock be found in place sufficient to warrant a prudent man 
in spending time and money on it, though it may not contain ore in pay- 
ing quantities. 

Where defendant testified that plaintiff, the owner of a quartz mine, 
gave him permission to locate a placer mine on the same ground, and that 
he spent 1 1,1 00 on a debt and water right necessary to work it, but on 
cross-examination stated that he was not told by plaintiff to work any 
particular mine, and plaintiff denied having any such conversation with 
him, but stated he notified him to keep off the ground, plaintiff was not 
estopped from asserting his claim to such quartz mine. Muldrick ^/ a/, 
vs. Brown ^/a/., 6( Pac. Rep. (Or.), 428. 

Mining Claims —Forfeiture — Assessment Work — Location. 

Since Rev. St. U. S. § 2324, authorizes the expenditure required to be 
made on lode mining claims to be either in work and labor or improve- 
ments, it is sufficient for defendants, claiming under a relocation after 
an alleged forfeiture by plaintiffs, to allege as such forfeiture that plain- 
tiffs failed during certain years to perform iioo worth of work and labor 
on the claim, but they must also negative the expenditure of that amount 
in improvements. 

Under Comp. St. Mont. 1887, div. 5, § 1477, requiring locators of 
mining claims 10 make and file for record with the county re:order a de- 
claratory statement thereof on oath, it is not sufficient for defendants. 


claiming under a location after an alleged forfeiture by plaintiffs, to allege 
simply that they have caused a record notice of their location to be made, 
for this is a mere conclusion, and does not suggest that the notice was 
ever verified, or put on record in the proper county. Power et aL vs. 
Sla etai., 61 Pac. Rep. (Mont.), 468. 

Death — Miners — Party Entitled to Maintain Action. 

Under Burns' Rev. St. 1894, § 7473, providing that, for the death of 
a person killed by reason of the violation of the act requiring mine 
owners to furnish timbers for making safe the place wherein miners are to 
work, a right of action against the mine owner shall accrue to the widow, 
children, or adopted children, or to the parents or parent, or to any 
other person or persons who were before such loss of life dependent for 
support on the person or persons killed, for damages sustained by reason 
of such loss of life, the personal representative of a person killed by 
reason of the failure of a mine owner to comply with the provisions of 
the act cannot maintain an action therefor. Boyd vs, Brazil Block- Coal 
Co., 57 N. E. Rep. (Ind.), 732. 

Mines and Minerals — Deed — Reservation of Minerals — Adverse 
Possession — Nonuser of Right of Access to 

Land — Easement. 

The owner in possession of lands covered with limestone ledges rising 
above the natural surface of the ground, having some timber thereon and 
some tillable land, but the chief value of which was manifestly the stone 
thereon, conveyed the same by warranty deed, excepting and reserving 
from the grant ** all mines and minerals which may be found on the above 
piece of land, with the right of entering at any time " to dig and carry 
away the same. Heid^ that though to construe the exception as referring, 
under the title of minerals, to the rock, would practically destroy the grant, 
the exception comprehended the rock, and reserved the title thereto in 
the grantor. 

Where a grantor conveyed land containing limestone ledges, reserving 
from the grant the minerals in the granted premises, and the right to enter 
to dig and carry away the same, the grantor's heirs are not deprived of 
their title to the minerals by the fact that the grantees have been in pos- 
session for a period sufficient to acquire title by adverse possession, and 
occasionally burnt a little of the lime deposits on the premises. 

Mere nonuser of the right of access does not deprive the grantor or his 
heirs of such right, where there is no adverse user of the rights reserved 
by the grantee. Brady vs, Brady et aL^ 65 N. Y. Supp. 6a i. 

Mines and Minerals — Natural Gas — Ownership — Increasing 

Flow — Artificial Means — Injunction — Destruction of 

Common Supply — Constitutional Law — Due 

Process of Law — Protection of 

Common Property. 

Since the right to take natural gas exists commonly in the owners of 
the superincumbent lands, it does not become the property of any of such 
owners until reduced to actual possession. Hence objections based on 
the right of transportation and sale of such gas are not pertinent to a 



complaint to enjoin the use of artificial means to increase the flow thereof 
from gas wells. 

Since the right to mine natural gas, when left to the natural laws of 
flowage, exists exclusively in the owners of superincumbent lands, and 
such gas is not subject to public appropriation without their consent, they 
are entitled to preserve it from destruction, independent of the right of 
the state to regulate it, and may enjoin any act of one owner to induce an 
unnatural flow into or through his wells which tends to the injury or de- 
struction of the common supply. 

Since the right to take natural gas is common to all the surface owners, 
and the gas does not become the property of any such owners until reduced 
to actual possession at the surface, Acts 1891, p. 89, prohibiting the use 
of artificial means to increase the natural flow of gas from a well, is not 
unconstitutional, as a deprivation of property without due compensation, 
but is a valid regulation for the protection of common property from de- 
struction, recognizing qualified ownership therein. Manufacturers' Gas 
& Oil Co. et al. vs, Indiana Natural Gas & Oil Co., 57 N. E. Rep. 
(Ind.), 912. 

Mining Claims — Conveyance to Settle Disputed Boundary — Con- 
struction — Action for Conversion of Ore. 

In compliance with a decree for specific performance of a contract made 
between the owners of two adjoining mining claims in settlement of a 
suit brought to determine a disputed boundary between them, the owner 
of one claim conveyed to the owner of the other the strip in dispute, 
** together with all the mineral therein contained, together with all the 
dips, spurs, and angles, and also all the metals, ores, gold and silver-bear- 
ing quartz rock, and earth therein." Heid, that such conveyance had no 
other effect than to fix the surface boundary between the two claims in 
accordance with the original intention of the grantee, and did not de- 
prive the grantor of any extralateral rights under the ground so conveyed ; 
such boundary being a side line of its claim. 

A complaint which alleges that plaintiff is owner of a mining claim, 
and of all the precious metals contained in any vein or lode, through its 
entire depth, whose apex is within the surface lines of such claim, and that 
defendant, the owner of a claim which adjoins one of the side lines of 
plaintiffs, has mined and removed ore from a vein which has its apex in 
plaintiffs claim, is sufficient, in the absence of a demurrer thereto, to 
support a judgment in favor of plaintiff for the conversion of such ore, 
although it does not specifically allege the facts which show that the 
boundary between the claims of the parties is a side line of plaintiffs 
claim beyond which it has extralateral rights. 

Possession of the surface of a raining claim is possession of a vein or 
lode having its apex within the surface lines of the claim, although in ex- 
tending downward, such vein may pass beyond the vertical side lines of 
the claim, and will support an action of trespass for the removal of ore 
from such vein, beneath the surface of an adjoining claim. 

Where, in an action for the conversion of ore, an injunction has been 
issued, in compliance with which the defendant has stored certain ore 
theretofore mined, it is not entitled to have the value of such ore taken 
into account in reduction of damages unless it proves such value, and re- 


turns or tenders the ore to the plaintiff. Montana Min. Co., Limited, vs, 
St. Louis Min. and Mill. Co., of Montana, 102 Fed. Rep. (U. S.), 430. 


Code Civ. Proc. § 1731, authorizing a stay of execution pending an 
appeal from a judgment directing the delivery of possession of real estate, 
applies in case of appeal by defendant in ejectment involving an unpat- 
ented mining claim. State ex ret. Baker vs. Second Judicial Dist. Court 
of Silver Bow County et aL, 61 Pac. Rep. (Mont.), i582. 

Mining Lease — Forfeiture — Construction of Contract — Waiver 

OF Performance. 

A lease of lands for oil and gas purposes provided that same should be 
void if a well was not completed on the premises by lessee within two 
months from the date thereof, unless the lessor should thereafter pay 
monthly to the lessor $10 for each month's delay in completing said well. 
It was further provided that, if operations upon the well were not com- V 

menced within 30 days of the date of the lease, ^10 extra should be paid 
for the second month. Work on the well was not commenced for over a 
month after the date of the lease, nor complete within two months and 
eight days. After the expiration of the second month, the lessee paid 
the lessor the sum of 1 10, and again, twelve days afer the expiration of 
the third month, the lessor tendered a similar payment, which was re- 
fused. Held^ that the first ^lo payment by the lessee could not be 
claimed by the lessor as on account of the money to be paid extra for the 
second month, and that a lease made by the lessor to another party after 
the expiration of the third month, and when the well had not been com- 
pleted, under the plea that the first lease had been foifeited for non- 
payment of the sum agreed to be paid for delay, was void. 

The lessor having permitted the lessee to locate a well under the lease, 
and himself constructed a dam to obtain water with which ro drill it, soM 
and delivered wood for fuel to the contractor employed by the lessee to 
drill the well, and lodged and fed the men who were doing the work, 
after three months from the date of the lease had expired, and down to 
the very day that he made a ne\y lease to another party, thus inducing the 
lessee and those claiming under him to believe that he was satisfied with 
their efforts at development, and that it was not his intention to claim a 
forfeiture, he will not be permitted, in equity, to insist that the first lease 
was forfeited at the time the second was executed, even though grounds to 
declare a forfeiture existed. Duffield et al. vs, Michaels et ai,^ 102 Fed. 
Rep. (U, S.), 820. 

Water Courses — Rights of Millowner? — Mining — Pollution 

OF Water of Creek. 

Where there are two ore mills in operation on the sarre stream, the 
lower proprietor may be compelled to take some steps and be at more ex- 
pense thaa he would if he were the only proprietor on the stream ; and 
it is the duty of the upper proprietor to use great care and caution, and to 
take such action as will avoid, as far as possible, any injury occurring to 
the lower proprietor by the flow of tailings from his mill. 

In making preparations to start up a mill for the reduction of gold and 




silver ores, which had been idle for a number of years, plaintiff found the 
water at the settling tank, from which the supply of water for operating 
the mill was obtained, contained a quantity of tailings, and that after 
being cleaned out the tailings and debris from above came into the tank 
in such quantities as might, if continued, prevent the mill from being 
run. Thereupon plaintiff demanded of defendant, who was operating a 
mill above the plaintiff's on the same creek, that he prevent his tailings 
from polluting the water, and was given assurance that by the time he got 
ready to start up this would be done. At the time this suit was com- 
menced defendant was not discharging tailings or sediment into the 
stream sufficiently to interfere with plaintiff's using its waters for his mill, 
and had constructed a series of reservoirs by which he impounded the tail- 
ings, and prevented any injurious matter flowing from his mill into the 
creek. Held^ that plaintiff was not entitled to an injunction restraining 

! defendant from polluting the stream. Otaheite Gold & Silver Min. & 

^ Mill Co. vs. Dean, 102 Fed. Rep. (U. S.), 919. 

Mining Claim — Correction of Calls. 

A patent to a mining claim will not be reformed in equity, so as to 
make the calls therein correspond with alleged monuments located on the 
ground, where it appears that for about 16 or 18 years preceding the fil- 
ing of the bill the alleged monuments were not in place, and there is as 
much doubt as to where the monuments were first located as there is 
whether the course is correct. Thallman et ai, vs. Thomas, 102 Fed. 
Rep. (U. S.), 935. 

Mining Claim — Notice of Location — Description — What Sufficient. 

The law will not hold the locator of a mining claim to a strict and 
technical observance of the statute in respect to the .terms of his notice, so 
long as he substantially complies with its requirements ; and if it appears 
that the location was made in good faith, and by any reasonable con- 
struction, in view of the surrounding circumstances, the language em- 
ployed in the description will impart notice to subsequent location. 
Wells vs. Davis ff a/., 62 Pac. Rep. (Utah), 3. 

Water Rights — Appurtenant to Land — Placer Mining. 

Where a third party, who owned land on which there was a placer 
mine, used water from a certain ditch to operate the mine as long as the 
land remained in his possession, such use made the right to appropriate 
water appurtenant to the land, and it passed with the land under a deed 
thereof. Mattis z^j. Hosmer ^/ ^z/., 62 Pac. Rep. (Or.), 17. 

Mines — .Abandonment — Placer Claims — Location Notice — 

Record — Evidence. 

Defendant M. and C. purchased certain mining claims on December 
28, 1896, and on the 30th went to examine them. No assessment work 
had been done since January i, 1896. Disappointed in the claims, they 
decided to abandon them, but, without leaving the spot, and within 10 
minutes of the alleged abandonment, relocated the claims in the name of 
defendant B. C. testified that **we" had men at work there on January 
22, 1897. M. testified that he "was there in possession" of the claims 


in January, February, and April, 1897. There was no evidence that B., 
who lived in New York, was ever informed that the claims had been re- 
located in his name, or that any work was being done for him, or that 
M. and C. were his agents. M.*s answer disclaimed all interest in the 
premises, and B.'s answer was verified by M., not as B.'s agent, but as 
one of the defendants. Heid, that a finding that there was no abandon- 
ment by M. and C. on December 30th was justified. 

Notices of location which state that the undersigned had located the 
ground for borate mining purposes, and describing the claims as 1,500 
feet long by 600 feet wide, are good notices of location of placer claims ; 
there being no difference between placer and lode locations in this regard. 

Notices of location of a mining claim which describe it as bounded on 
the east by the H. mine, and as being a quarter of a mile south of the 
B. road and about three miles east of C, sufficiently describe the claim 
with reference to natural monuments, since notices of location are liber- 
ally construed. 

CJnder Rev. St. U. S. § 2324, requiring locations of mining claims to 
be distinctly marked on the ground, so that the boundaries can be traced, 
and that records of mining claims shall contain the name of the locator, 
the date of location, and such description of the claim with reference to 
some natural monument as will identify it, it is not necessary that the 
record shall show that the claim is marked on the ground. 

Objection to the admission of the record of mining claims of a special 
district in evidence, without proof that there was such a district and a 
custom requiring a record, and that the book came from the proper cus- 
tody, is not well taken, where the pleadings admitted that the claims were 
in the district, and the complaint alleged, and was supported by proof, 
that G. was the recorder of such district, and G. was called by both 
sides to produce the record of their location notices. McCann et al., vs. 
McMillan et aL^ 62 Pac. Rep. (Cal.), 31. 

Mining — Notice of Claim — Sufficiency — Description of Claim. 

A final notice of the location of a mining claim which fails to mention 
either the state or county of the purported location, but which refers to 
the preliminary notice posted by the locators as required by law, and 
recorded in the proper county, which latter notice named the county in 
which the claim is located, is a sufficient compliance with Rev. St. U. S. 
§ 2324, requiring all records of mining claims to contain such a de- 
scription of the claim located, by reference to some natural object or 
permanent monument, as will identify the claim, and also with St. 1897, 
p. ar5, requiring the certificate to be recorded to state a description of 
the claim, defining the exterior boundaries as they are marked upon the 
ground, and such additional description by reference to some natural 
object or permanent monument as will identify the claim. 

The description of a mining claim in a notice of location by metes 
and bounds, with reference to permanent stone monuments erected at the 
four corners and at the center of the east and west end lines of the claim, 
is sufficient, under St. 1897, p. 215, requiring a description which de- 
fines the- exterior boundaries as they are marked upon the ground, and 
such additional description by reference to some natural object or per- 
manent monument as will identify the claim, especially as against a sub- 


sequent locator who finds the bedding and tools of the previous locators 
upon the claim, and an employee holding possession for them. Talmadge 
et al, vs. St. John ^/<j/., 62 Pac. Rep. (Cal.), 79. 

Mines and Mining — Overlapping Claims — Necessary Work — 


Plaintiff located a raining claim, and thereafter defendant entered on 
the land, claiming it as vacant mineral land, in that plaintiff had failed 
to do the amount of work required by law for five years previous. 
Plaintiffs owned a claim that overlapped the claim in dispute, and on the 
land common to both claims had done work in the way of tunneling 
largely in excess of that required by law, with the manifest intention of 
striking the lode of the claim m dispute. HM, that a finding that the 
work performed could not be a benefit to the claim in dispute, and that 
plaintiffs were not entitled to the same, was erroneous, since the court 
could not be permitted to substitute its judgment as to the expediency of 
the methods employed for developing a mine for that of the owners. 
Mann vs, £ublong ^/ a/., 62 Pac. Rep. (Cal.), 120. 

Mines and Mining — Adjoining Claims — Injunction. 

Complainant and defendants were joint owners of a group of mining 
claims, and defendants were owners of an adjoining claim. A tunnel had 
been run through the group owned jointly, for the purpose of working 
that property ; and defendants e> tended the tunnel into the property 
owned by them, and the use of the tunnt-1 by defendants in working their 
property resulted in the exclusion of complainant from the tunnel. Ife/d, 
that compUinant was entitled to an injunction restraining defendants from 
excluding her from the tunnel. People ex reL Breene et aL vs. District 
Court of Lake County e/a/., 62 Pac. Rep. (Colo.), 206. 

Mining Clalms — Determining Rights of Claimants — ^Jurisdiction 

OF Equity. 

Plaintiffs executed a bond and lease of a mining claim to defendant, 
who was a'so a part owner, under which defendant went into exclusive 
possession. He subsequently took proceedings to forfeit the interests of 
plaintiffs for refusing to pay their share towards assessment work, which 
they claimed he had undertaken by the bond and lease to do at his sole 
expense. After such proceedings for forfeiture, defendant applied for a 
patent, and obtained a receiver's final receipt in his own name. J/e/^, 
that the rights of plaintiffs in the claim were not matters which could be 
adjudicated by the land department of the United States, but were within 
the jurisdiction of a court of equity, and that plaintiffs were entitled to 
maintain a suit for their deiermmation without waiting until a patent for 
the claim had been issued to defendant. Malaby vs. Rice e/ a/., 62 Pac. 
Rep. (Colo.), 228. 

Mines and Mining — Failure to Pay Laborers in Lawful Money — 

Special Legislation. 

Under Ky. St. § 2739a, only such persons or companies engaged in the 
mining industry as are unable to pay their employes in lawful money at 
the time required by the statute are given 15 days to execute duebills, and 


therefore, where the employer is able to pay, an iodictment against him 
for failing to do so may be returned before the expiration of the 15 days. 
The fact that the statute applies only to persons or companies employ- 
ing ten or more persons does not render it unconstitutional as special 
legislation, as the classification is natural and reasonable, and consistent 
with the end sought to be accomplished by Const. § 244, which provides 
that all wage-earners ''employed in factories, mines, workshops or by cor- 
porations shall be paid for their labor in lawful money." Commonwealth 
vs. Hillside Coal Co., 58 S. W. Rep. (Ky.), 441. 

Title and Conveyances. 

In the year 1865, Lady W., the plaintiflPs predecessor in title, sold and 
conveyed to J. C. & Sons, Limited, at the price of £28,500, all the 
mines, veins and seams of coal, iron, and other ores under certain heredi- 
taments known as the **S. Estate,'* containing 1,178 acres, or there- 
abouts, together with full power to make, sink and work all such pits, 
shafts, drifts, levels, * * * and do all such other acts as might be deemed 
necessary or convenient for working, searching for, getting and raising 
the same, and to do all other works and things, which might be nec- 
essary or convenient for the purposes aforesaid, or any of them, doing as 
little damage as might be consistent with the due and proper carrying 
out of the said works. The deed contained a covenant by the company 
not to sink any shaft, or drive any level, which shouM or might weaken 
or damage any building on the vendor's lands, and to make satisfaction 
to be assessed as therein mentioned for the spoil, or damage, or injury 
done to, in, and upon the said lands by the machinery for getting and 
raising the said coal, or by the sinking or working of any pit or shaft, or 
otherwise by virtue of or under the powers aforesaid. The defendants in 
this action were lessees of J. C. & Sons, Limited. They were working, 
on the long- wall system, some of the deeper seams of coal under the S. 
estate, and had caused some subsidence of the surface in places, but no 
substantial damage was proved. The plaintiff brought this action for an 
injunction to restrain the defendants from working the coal so as to cau«e 
subsidence of the surface. The defendants claimed a right to cause sub- 
sidence so long as they worked in a proper method. There was a con- 
flict of evidence whether it was possible to work the coal at all without 
causing subsidence, but it was in effect admitted that it was not possible 
to do so at a profit. HM, that the grant of the minerals did not ex- 
presly or by necessary implication give power to let down the surface, 
and the plaintiff was entitled to the injunction. Westmoreland vs. New 
Sharlston Colliery Co., 80 Law T. (N. S.), 846, affirming (Eng., 1899) 
6z J. P., 198. 

Mines and Minerals — Reservation in Lease. 

An acception of reservation contains in a lease of ** mines and min- 
erals*' under lands demised \nc\\xdtSt prima facie, d\\ those substances 
otherwise falling under the definition of minerals which have a use and 
value of their own, either for the purpose of sale, or for other purposes * 
independently of, and separately from, the use of the rest of the soil, 
whether capable or not of being worked for commercial profit. 

A power subsequently contained in the lease, for the lessor and those 


claiming under him << to drain, get, and dispose of the said mines and 
minerals at his and their free will and pleasure, nevertheless not breaking 
open or doing any damage to the surface of the land, * * * or the 
buildings thereon, and making reasonable compensation * * * for all 
damage thereby occasioned,*' does not restrict the meaning of the words 
" mines and minerals '* to such substances as can be worked for commer- 
cial profit. Johnstone vs. Crompton, 68 Law J. Ch. 559 (Eng., 1899), 2 
Ch. 190, 81 Law T. (N. S. 165, 47 Wkly. Rep.), 604. 

Mining I^ease. 

The law that, when a mine is once open, the sinking a new pit on the 
same vein is not necessarily the opening of a new mine, is inapplicable to 
a case where two portions of a settled estate are separated from each other 
by a strip of land belonging to a different owner, although one continuous 
seam of coal runs under the whole. If, therefore, the seam of coal has 
been worked under one portion of the settled estate in the lifetime of the 
settler, but not under the other, the unworked portion cannot be held an 
'* open mine,** even when the tenant for life has acquired the coal under 
the intervening strip. The court, in sanctioning a lease of the unworked 
portion, will consequently direct that three-fourths of the rent must be 
set aside and invested, according to subsection 3 of section 4 of the 
settled-estates act of 1877. In re Maynard*s Settled Estate, 68, Law J. 
Ch. 609 (Eng., 1899), 2 Ch. 347, 81 Law T. (N. S.) 163, 48 Wkly. 
Rep. 60, 63 J. P. 552. 

Operation of Mines — Statutory Regulations. 

If the operator of a coal mine employs an examiner holding a certifi- 
cate from the State board of examiners authorizing him to act as such, 
and the examination of the mine is made at the time required by law, it 
will constitute a compliance, so far as the operator is concerned, with the 
provisions of section 4, c. 93, 2 Starr and C. Ann. St* 2,719, requiring 
inspection of mines to determine their safe condition. Kellyville Coal 
Co. vs. Hill, 87 (Ill.)> App. 424. 

Mines and Mining — Rights and Liabilities Incident to Working. 

Hurd's St. c. 93, § 44, provides that every laborer or miner who shall 
perform labor in opening and developing any coal mine, including sink- 
ing shafts, constructing slopes and drifts, mining coal, and the like, shall 
have a lien on all the property of the mine to the value of such labor. 
Held, where a bill was filed to have a lien declared against the property 
of a mining company, and it was nowhere averred in the bill that the 
work and labor performed were in ** opening and developing** a coal 
mine, no lien could be had. Borders vs, Uhe, 88 (111.)* App. 634. 

Mines— Abandonment — Relocation — Validity. 

Where defendant purchased a mining claim on December 26, 1890, and 
thereafter abandoned it because unable to do the assessment work, and 
defendant's son relocated the claim as an abandoned claim on January 30, 
1891, giving the date of discovery as December 20, 1890, the relocation 
was invalid, and did not prevent a location by others. 

Where defendant purchased a mining claim, and shortly thereafter 


abandoned it because unable to do the assessment work, it was error to 
refuse to charge that defendant intended to abandon his rights under his 
purchase. ^'^'^ J 

Defendant purchased a mining claim December 26, 1890. and shortly 
thereafter abandoned it because unable to do the assessment work. De- 
fendant's son relocated the claim January 30, 189 1, as an abandoned lode, 
giving the date of discovery as December 20, 1 890, and then conveyed to 
defendant, who claimed solely under the relocation until after plaintiff 
had located an interfering lode. Heid, that defendant could not recall 
his abandonment, and claim that the relocation was to protect his rights 
under the original claim. Niles vs. Kennan, 62 Pac. Rep (Colo.), 360. 

Mines and Minerals — Trespass — Injunction — When Justified. 

Where, in an action to enjoin defendants from trespassing on a mining 
claim and destroying plaintiffs end gates, ditches, etc., the only evidence 
of trespass was that on one occasion one of defendants came and com- 
menced digging on a ditch, and refused to go when ordered, saying that 
he was going to pull up plaintiffs gates and take the water they were 
using, the complaint was properly dismissed, since an injunction to re- 
strain trespass on a mining claim will not issue on proof of a single act of 

In an action to enjoin trespass on a mining claim, proof that defen- 
dants are insolvent is not of itself sufficient to justify granting an injunc- 
tion. Parker vs. Furlong et aL, 62 Pac. Rep. (Or.), 490. 

Mines and Minerals — Counterclaims — Subject of Suit — Independ- 
ent Trespass — License — Easements — Advprse Claim 
Against United States — Parol License. 

Plaintiff brought an action for damages, and to enjoin defendants, who 
owned a placer mine just above plaintiff's, on the same creek, from flood- 
ing debris down on plaintiff's claim in the operation of defendants' mine. 
Defendants filed a counterclaim for damages occasioned by water being 
backed up on their claim in consequence of a dam built across the creek 
below defendants* mine by plaintiff. He/d^ that the counterclaim was 
properly struck out on motion of plaintiff, since such damages were in no 
way connected with the subject of the suit, and also arose from an inde- 
pendent trespass. 

Where plaintiff sued for damages and to enjoin defendants from flood- 
ing debris down on plaintiffs mining claim in the operation of defend- 
ant's mine, which was situated just above plaintiffs, on the same creek, 
an answer which alleged that plaintiff leased his mine to defendants, in 
consideration of which they dug ditches, constructed flumes, and put in 
sufficient pipes to operate it, was properly struck out, on motion, as 

Where, in an action for damages occasioned by defendants flooding 
plaintiffs placer mine with debris from defendant's mine, which was sit- 
uated just above plaintiffs on the same creek, defendants defended their 
right on the ground of adverse user before plaintiff obtained title to his 
claim from the United States, but did not show that such adverse user 
existed for 10 years subsequent to the acquirement of plaintiff's title, such 
use constituted no defense to the action, since there can be no adverse 
user as against the United States. 


In an action for damages occasioned by defendant flooding debris down 
on plaintifTs placer mine from defendant's mine, which was situated just 
above plaintiff's mine, on the same creek, defendant claimed to do so by 
virtue of a license from plaintiff, and alleged that on the strength of such 
license he had made valuable improvements on his own claim, but did 
not show that he paid anything for the license, or that he had not been 
repaid for his improvements by the operation of his own mine. Held^ 
that the license was revocable, and did not constitute a defense to the ac- 
tion. Miser vs, O'Shea et aL^ 62 Pac. Rep. (Or.), 491. 

Master and Servant — Relations of Parties — Injury of Servant 

WHEN not in Line of Duty — Negligence — Unsafe 

Premises — Injury of Licensee. 

A coal miner who, during the noon hour, while not engaged in work, 
goes to a different part of the mine, for the purpose of visiting with an- 
other miner, is not, while so absent, engaged in the line of his duty, so as 
to impose upon the employer the duty of a master to see that the entry 
through which he passes from and to the part of the mine where he is 
employed is kept in a safe condition for his passage. 

Where the miners in a coal mine, ^ith the knowledge and implied con- 
sent of the owner, are accustomed to use the passages or entries in the 
mine as a place for congregating or passing to and fro during the hours 
of recreation, it is negligence in the owner to introduce and extend along 
such an entry an electric wire which is dangerous to the life of those who 
come in contact therewith, without properly insulating or inclosng the 
same, or giving notice of the danger to those who, he should reasonably 
apprehend, are likely to be brought in contact with it, and such neg- 
ligence will render him liable for the death of a miner who, in the accus- 
tomed use of the premises, and without knowledge of the danger or 
negligence on his own part, is killed by coming in contact with such 
wire. Ellsworth z^x. Metheney, 104 Fed. Rep. (U. S.), 119. 

Coal Mine — Lease of Adjoining Coal — Authorization of Tres- 
pass — Lessor's Liability — Measure of Damages. 

Where defendant, the owner of a coal mine and of the surface of ad- 
joining lands, leased his mine, and included therein the coal underlying 
the adjoining tract, to which he had no title, and in pursuance of such 
lease the tenants mined coal under the adjoining tract, and defendant re- 
ceived a royalty thereon, he is liable in damages for the tenant's trespasses, 
having specially authorized them by his lease. 

Where, through inadvertence, defendant, the owner of a coal mite, 
caused coal underneath an adjoining tract, to which he had no title, to be 
mined by his lessees, and received a royalty thereon, the measure of the 
damages is the value of the coal at the pit mouth, less the cost of its eleva- 
tion to the surface, and, if loaded on railroad cars, the value after loading, 
less the cost thereof, and not the value of the coal in its un severed condi- 
tion in the mine. Donovan vs. Consolidated Coal Co. of St. Louis, 58 
N. E. Rep. (III.), 290. 

Oil Lease — Damages. 

Where a decree in a suit by the lessor against the lessee in an oil lease, 
providing that the lessee shall have no rights in the premises unless he 


sinks a well at a certain place, and appointing a master to ascertain dam- 
ages for failure to sink the well before, is slightly modified on appeal, and 
case is remanded to the court below for further orders, to give effect to 
the judgment of the appellate court, but no reference is made to the 
provisions for ascertaining damages, the same are not affected by the 
judgment on appeal. 

Defendant leased from plaintiff land, with the exclusive right to oil, for 
a royalty of one-eight thereof. He had also a lease of adjoining lands, 
and on them sunk a paying well near plaintiffs land, draining the same, 
and refused to sink a well near there on plaintiffs land. Thereafter, 
when the oil was nearly exhausted, he forfeited his lease from plaintiff 
under a decree that he should do so or sink a well. Held, that plaintiffs 
damages were one-eighth of the oil produced by the adjoining well up to 
the time of the forfeiture of plaintiffs lease. Kleppner vs. Lemon, 47 
At. Rep. (Pa.), 353. 

Mines and Minerals — Contracts. 

Defendant proposed that if plaintiff would purchase certain mining 
claims through one A., and procure a conveyance to himself or to A., to 
be held in trust for defendant and the plaintiff, then in that event plaintiff 
should have a certain interest in all claims that might be thereafter located 
by defendant ; but no time was fixed within which such claims should be 
bought, or within which defendant should commence to locate claims. 
Held, that the acceptance of the proposition by A., who had not been 
consulted when it was made, was essential to give it the effect of a con- 
tract, and hence plaintiff was not entitled to any interest in a claim located 
by defendant before A. consented to make the purchase. Ranahan vs. 
Gibbons etaL^ 62 Pac. Rep. (Wash.), 773. 

Mining Claim — Locators — Citizenship — Rights in Mining Ground 

— Not AFFiCTED by Co-Locator Being and Alien — Location 

of Mining Ground — ^By Citizen and Alien Jointly — How 

Far Valid — Consent to Boundary of Mining Ground 

— Part Owner Cannot Bind His Co- Owner — 

How Long Necessary to Bind 

Adjoining Owne • s. 

When it appears in a protest and adverse suit that one of two locators 
of a mining claim was a native-born citizen ; that the other, although of 
foreign birth, had resided in various States of the Union for many years ; 
that he had served as a soldier in the volunteer army of the United 
States from 1863 to 1866, and had been honorably discharged therefrom ; 
that he had exercised the right of franchise, and had voted at several 
territorial elections, and in 1887 had taken and ' subscribed an oath to 
the effect that he was over 21 years of age, a resident of Utah, a natur- 
alized citizen of the United States, and that he would support the consti- 
tution thereof ; that he was awarded and drew a pension from the gov- 
ernment for disabilities received while in the military service, a finding 
that the first party was a citizen and the second party a naturalized citi- 
zen will not be disturbed. 

The rights of a citizen locator of mining ground and his subsequent 
grantees cannot be affected by the fact that his co-locator was an alien. 

VOL. XXII— 3a. 


If a citizen ard an alien jointly locate a cbim not exceeding the amount 
of ground allowed to one locator, such location is valid as to the citizen, 
or to one who has declared his intention to become such , and a convey- 
ance by him, through an alien, to another citizen, conveys a complete 
title to the claim located, provided all other provisions of the law were 
complied with, and there be no intervening rights. 

An alien who has declared his intention to become a citizen by enlist- 
ment in the United States army under section 2166, Rev. St. U. S., may, 
under the provisions of section 2319, locate mineral lands upon the unoc- 
cupied public domain. 

The consent of an owner of an undivided interest in mining ground to 
the establishing of a certain line as a boundary cannot bind his co-owner, 
not consenting. 

Although when adjoining owners and their predecessors in interest oc- 
cupy land to a given line, and treat such line as a boundary between their 
respective lots for 20 years, neither can thereafter claim beyond such 
line; and although a parol agreement, long acquiesced in, to settle a 
boundary between adjoining owners, being the result of an honest attempt 
to fix the true boundary line, according to which the parties and their 
prtdeccssors to have actually occupied and made improvements with ref- 
erence thereto, though the time has not yet been sufficient to establish a 
bar under the statute of limitations, will work an estoppel, yet a recent 
parol agreement between parties fixing the boundary line between un- 
patented mining claims is void under the statute of frauds, and, under the 
circumstances shown, could not bind the government. Strickley et ai. 
vs. Hill, 62 Pac. Rep. (Utah), 893. 

M'NiNG Claims — Side and End Lin is — Secondary Veins — Extra- 
lateral Rights — Vein Crossing Common Side Line at Angle. 

Where the end lines of a iriining claim have been established, they re- 
main the end lines as to all veins found within its surface boundaries. 

When a secondary or accidental vein crosses a common side line be- 
tween two mining claims at an angle, and the apex of the vein is of such 
width that it is for a given distance partly within one claim and partly 
within the other, inasmuch as neither statute nor authority permits a di- 
V sion of the crossing part of the vein the rights of the parties will be de- 
termined by the priority of location, and the entire vein considered as 
apexing upon the senior location until it has wholly passed beyond its 
side line, without regard to the direction in which the vein dips. St. 
Louis Min. & Mill. Co. of Montana vs, Montana Min. Co., Limited, 
104 Fed. Rep. (U. S.), 664. 

Mines and Minerais — Entry — Cancellation — Relocation — Boun- 
daries — Visible Vein — Monuments— Impracticable Location — 
Suitable Marking — Abandonment of Location. 

B., as amicus curies ^ filed an adverse to C.'s entry of a mining claim; 
and the government land department, in deciding on such adverse, found 
that no discovery had been made on such claim on October ii, 1893, 
and that the shaft therein had been dug but 25 feet deep, and entered 
judgment canceling the entry. B. thereafter performed work in develop- 
ing the claim, and C. filed a second application for a patent, but not 


basing the same on his former entry, ffe/d, that the findings and de- 
cision of the land department were not res Judicata in favor of £., and 
therefore not admissible in evidence in his action in the district court in 
support of his adverse to such second application, as such findings and 
decision amounted to no more than a nonsuit of C.'s application, and oc- 
curred in a proceeding in which B. neither had nor asserted any bene- 
ficial interest. 

B., as amicus curia ^ filed an adverse to C.'s entry of a mining claim ; 
and such advene was susta'ned by the government land department, and 
the entry canceled. In such proceeding C. testified that he had made no 
discovery on the claim, that he had dug a shaft but 25 feet, and that he 
had done no work on the shaft during the year in which such proceedings 
were had. After the decision B. developed the claim, and, when C. 
filed application for a patent, brought action m support of his adverse to 
such application, alleging that he had performed such work in reliance on 
C.'s testimony, but failing to allege that such testimony was given to de- 
ceive B., and with the intention that he should act thereon. Hcld^ that 
B.'s allegations did not show an estoppel in pais, and a demurrer thereto 
was properly sustained. 

Rev. St. U. S. § 2320 provides that no location of a mining claim 
shall be made until discovery of the vein or lode within the limits of the 
claim located. Plaintiff filed his certificate of location of a mining claim 
in May, 1892, but did not make a discovery until April, 1894, at which 
time he relocated the premises. Before such discovery, defendant made 
a discovery and location on the premises. Held, that plaintifTs discovery 
would not relate back to the time of the filing of the original location, 
so as to cut out intervening rights, and hence such original certificate of 
location was not admissible in evidence as against defendant. 

2 Mills' Ann. St. § 31 S3, provides that where one or more posts which 
must be erected to mark the surface boundaries of a mining claim *' fall 
by right on precipitous ground, where the proper placing of it is imprac- 
ticable or dangerous to life or limb,'' it shall be proper to place such post 
at the nearest practicable point. The corner of a mining claim fell on the 
top of a railroad embankment la or 15 feet in height. Heidy that such 
pomt was not an impracticable or dangerous one, within the meaning of 
the statute, and that it was proper to instruct the jury that in the absence 
of a showing that it was impracticable to place the post at that point, or 
that it would be interfered with by the passage of trains, it was the duty 
of the claimant to erect the same at its true point. 

Under 2 Mills' Ann. St. § 3153, providing that where a corner of a 
mining claim falls on precipitous ground, where it is impracticable and 
•dangerous to life and limb to place it, such post may be erected at the 
nearest practicable point, and '< suitably marked to designate the proper 
place," a post set away from the true southwest comer of a claim, and 
marked ** W. C. 4 — 9,005," did not indicate by either course or distance 
where the true comer was 10 be found. 

On a contest as to priority of location of a mining claim, while the 
jury were reviewing the disputed premises, a guide selected by defendant, 
and who had surveyed the premises, pointed out to the jurors the place 
where one of the corners of the claim were located. On the return of 
the jury, and before argument, the matter was called to the attention of 
the court, and the jury were cautioned to disregard the statement of the 


guide. Heldy on appeal by plaintiflf, that the supreme court would not 
assume that the conduct of the guide was prejudicial, in view of the 
court's instruction. 

In a contest over priority of location and discovery of a mineral vein, 
the jury was sent to view the premises, and all except two of them went 
into the shafts dug by the respective contestants. The bailiff in charge 
stated to the jurors that those who did not care to go down the shafts need 
not do so. Held, that in the absence of a request to instruct the jury as 
to the manner in which they were to view the prem ses, or the selection of 
guides by the parties to point out such features as it was desirable for the 
jury to see, there was no such misconduct on the part of the jury as to 
warrant a new trial. Beals vs. Cone et aL, 62 Pac. Rep. (Colo.), 948. 

Mines and Minerals — Ownership. 

An amended complaint in an action to quiet title to an interest in a 
mining claim was filed several months after the filing of the original com- 
plaint, and alleged that plaintiff then was, and at all times therein men- 
tioned had been, the owner of such interest ; there being no other time 
mentioned in the pleadings, and no demurrer thereto. There was a find- 
ing that defendant conveyed the interest by deed of date prior to the 
filing of the original complaint, and that the plaintiff practiced no fraud 
in obtaining such deed, which finding was not questioned on appeal. 
Held, that the finding sustained an averment of ownership prior to the 
time the complaints were filed. 

Action was brought to quiet title to an undivided two-thirds interest in 
a certain mine ; the claim being founded on a contract whereby plaintiff 
was to have such interest, and was to deliver one-third of the output to 
defendants. The decree adjudged plaintiff the owner in fee of a |-J- in- 
terest, and that defendants had no title thereto, and enjoined the latter 
from setting up any claim to such interest adverse to plaintiff; similar 
provisions being decreed as to the interests of defendants. Held, that the 
decree did not cut off defendants from claiming one-third of the minerals 
under the contract, since the interests as adjudged were undivided, the 
owners becoming tenants thereof in common. 

Where an undivided interest in a mining claim is conveyed in consider- 
ation of the grantee's agreement to work the mine and deliver one-third 
of the minerals to defendant, and do certain other acts in the future, their 
performance not being made a condition subsequent, a mere failure to per- 
form on the part of the grantee does not constitute a failure of consider- 
ation, so as to warrant cancellation of the deed. 

Where an action is brought to quiet title to an undivided interest in a 
mining claim, such interest having been conveyed in consideration of a 
certain contract, whereby plaintiff was to work the mine as he saw fit, and 
deliver one-third of the output to defendant, plaintiff cannot be required 
to work the mine, as a condition on which his title will be quieted, since 
his contract is incapable of being: specifically enforced. Downing vs, 
Rademacher et aL, 62 Pac. Rep. (Cal.), 1055. 

Mines — Laborers — Liens — Fireman — Filing. 

Where a co-owner of a mining claim authorizes his co-owner and others 
to place mining machinery upon their claims and to do mining thereon, 
and the foreman in charge of such mining employs laborers to work upon 
such claims, who perform labor thereon, of which facts such co-owners 


have notice, and fail to post the notice as provided for in section 2226, 
Comp. Laws, 1897, such laborers are entitled to a lien upon the entire 
claim or claims, and may file and foreclose a lien for the value of the 
labor performed thereon. 

Under such circumstances, the foreman in charge of the mining done 
upon such claim is the agent of the owners thereof, under section 2217, 
Comp. Laws, 1897, notwithstanding such foreman was employed by one 
of the co-owners. 

A lien filed against several mining claims is not void because the 
amounts against each claim are not segregated in the lien, but, where 
other liens exist at the time, such lien is postponed to other valid liens. 
Postfix. Fleming ^/ a/., 62 Pac. Rep. (N. M.), 1087. 


Tunneling, A Practical Treatise by Charles Prelini, C.E., with ad- 
ditions by Charles S. Hill, C.E. 150 diagrams and illustrations. 
Pp. XV and 307. D. Van Nostrand Company, New York, 1901. 
Price ^3.00. 

In his preface the author comments upon the fact that there are ''but 
two books on tunneling in the English language, neither of which he 
could recommend as text-books for his pupils." The works referred to 
are those of Drinker and of Simms. Both are extremely valuable as 
reference books, but far too ponderous, expensive and diffuse to be used 
advantageously by students during the period of their school training. 
The writer of this notice can appropriate to himself the expression quoted 
above, and is glad to welcome Mr. Prelini's book. Any engineering sub- 
ject, like tunneling, the details of which fall almost entirely outside of the 
observation of the average student, presents serious difficulties to instructor 
and student alike. The student is usually hampered by want of that 
modicum of preparatory acquaintance with the subject, and with engineer- 
ing matters in general, which would enable him readily to grasp principle 
and detail, while the instructor frequently labors under a keenly felt lack 
of adequate illustrative diagrams and models. 

The work before us is well designed to serve as a text-book, and we 
think will be found useful also by practicing engineers. It is concise, 
orderly in its arrangement, and in the examples quoted is something bet- 
ter than a historical narrative. The author has endeavored to preseot an 
epitome of current practice, and has not entered into a confusing mass of 
detail. He has refrained, too, from giving undue space to certain mon- 
umental tunneling operations which are interesting mainly because of the 
abnormal conditions under which they were carried out. 

Beginning with various preliminary considerations, as to location and 
survey of the center line, and the determination of the form and dimen- 
sions of cross- section, the author next proceeds to a brief discussion of 
drilling machines and explosives used. Jn Chap. IV. the general methods 
of excavation are considered, for soils of different kinds. Following are 
taken up successively: the temporary support of tunnel excavations; 
modes of transport of the excavated material; centers and molds for 
masonry linings ; timber and masonry linings ; tunneling in firm rock, 


with a number of examples ; representative power plants for tunnel work ; 
tunneling through soft ground ; special methods for treacherous ground ; 
open cut methods, with recent examples, including eight pages on the 
New York underground rapid transit railway ; the several methods of sub- 
marine tunneling, with a number of good examples of recent work ; acci- 
dents and repairs during and after construction; relining timbered 
tunnels with masonry; ventilation and lighting of tunnels during con- 
struction ; general figures of cost and speed of work. It will be seen from 
the above that the author treats only of railway and other tunnels of large 

There are several matters in the treatment of which we think the author 
has erred on the side of excessive brevity : for example, in describing the 
actual work of drilling, blasting and handling the rock, in ordinary tim- 
neling ; the driving of the advance heading, as distinguished from the en- 
largement to full cross-section ; figures of speeds and costs might have 
been given with advantage in somewhat greater detail, with comparative 
figures relating to heading and enlargement. Some of the illustrations are 
rather imperfect, as for example the four on p. 86. We would suggest, 
also, that in a future edition, the author would do well to insert references 
to the technical periodical literature dealing with the various subdivisions 
of the subject. 

The book as a whole is excellent, and we think it will prove to be a 
valuable aid for all who have to do in any way with this branch of en- 
gineering. R. P. 

The Chemical Analysis of Iron^ a complete account of all the best known 
methods for the analysis of Iron, Steel, Pig Iron, Iron Ore, Limestone, 
Slag, Clay, Sand, Coal, Coke and Furnace and Producer Gases. By 
Andrew Alexander Blair. Fourth Edition. Philadelphia and Lon- 
don, J. B. Lippincott Company, 1901. 8vo. 319 pages. 

A fourth edition of this useful work has become necessary to keep pace 
with the changes constantly occurring in the requirements of iron works 
laboratories, and the introduction of improved methods. 

In preparing this edition, the author has rewritten many parts, and 
throughout has substituted names for formulae (except of course in equa- 
tions) and has also made the nomenclature uniform, l)esides omitting the 
marginal notes. 

Among the changes in the text may be noted the entire abandonment 
of the use of ammonium copper chloride in carbon determinations, the 
corresponding potassium salt having been adopted ; the omission of 
Stead's alkaline method for colorimetric determination of carbon ; and 
omission of determination of antimony in coins. 

The new features are numerous, and carefully selected. We may enu- 
merate the most important : 

Bamber's method for sulphur in pig irons (p. 67). 

Handy's for phosphorus by acidimetric titration of the molybdate pre- 
cipitate (p. 104). 

Vanier's use of a centrifugal in the Deshayes method for manganese 
(p. 122). 

New form of apparatus for the determination of carbon by fusion with 
potassium bisulphate (p. 133). 

Use of a ** dummy** counterpoise absorption apparatus in the deter- 
mination of carbon by weighting as CO3 (p. 148). 


Job and Davies method for carbon in irons (p. 160). 

Notes on the repeated use of potassium copper chloride solutions 
(p. 164). 

Modifications in the details for determination of titanium (p. 179). 

The ether method for separation of niokel, inserted as an appendix in 
the third edition, is here introduced in its proper place (p. 186) and we 
have also the application of the ether method in the separation of chro- 
mium (p. 189) and of aluminum (p. ipi). 

An entirely new section describes methods for the determination of 
molybdenum in irons and in ferro-molybdenum (p. 199). 

A new chapter is also devoted to methods for the analysis of ferro- 
chrome, fcrro-silicon and ferro-titanium (p. 205). For most of the de- 
terminations, the material is to be attacked by fusion with sodium peroxide, 
which reagent however, is not described in the chapter on reagents, an 
omission which should be remedied, as the peroxide is prone to deteri- 
orate, and unless kept with great care, may fail to serve its purpose. 

The methods for the analysis of coal and coke are those given by the 
Committee on Coal Analysis of the American Chemical Society (p. 276), 
their report being substituted bodily in place of the text given in previous 

It will be seen that the book has undergone a most thorough revision, 
and in its new form, is well brought up to date. E. W. 

Qualitative Chemical Analysis. By Albert B. Prescott and Otis C. 
Johnson. Van Nostrand Co., New York, 1901. 8vo. 420 pages. 
Price, ^3.50, net. 

This fifth edition of this well-known text-book on qualitative analysis 
deserves a special review, as it is in reality a revised and enlarged edition. 
All the characteristic features of the preceding edition have been retained, 
but a great quantity of new material has been added and the old corrected. 
The order of treating the subject, if intended for a student without much 
previous training in chemistry, is certainly open to criticism ; for to the 
beginner the reactions of mercury, arsenic, antimony, etc., and their sepa- 
ration are not encouraging. 

In the introductory part short accounts of the periodic system and the 
ionization theory, etc., are given, as in most of the recent text-books, but no 
further use seems to be made of any of the principles involved later in the 
book. These were better omitted entirely in a book on qualitative analysis 
in which the periodic system is not utilized, and neither ionic or reversible 
reactions indicated. Much of the new material deals with the rarer ele- 
ments, and this is a most welcome addition, as they are daily becoming 
of greater industrial importance. 

As a whole the book by enlargement has changed in character ; it is 
a carefully compiled and condensed small encyclopedia of analytical and 
inorganic chemistry, containing a tremendous amount of information 
logically arranged and concisely stated, but does not attempt to explain 
analytical reactions from the standpoint of modern chemistry. It does 
not tend to give one a broad or comprehensive view of chemical reactions, 
but on the other hand it contains more useful chemical facts than any 
volume of equal size. 

The book is excellently printed and tound, and seems very free from 
typographical errors. E. H. M. 

lf% Jena Normal if% 

C. A. F. Kahlbaum's Chas- Schleicher & 

ChCRieally Pure Organic SchUell'S 

Inoraaiilc ChemlcaU. C. P. FILTERPAPER 


18th St and 3rd Avo.. Nsw York. 

■■raattaa *«» ManuFAeTuMtiit o* 

Chemicals t Chemical Apparatus 


SCHOTT & GENOSSEN, Manufacturers of the famous Jena Glass 

C. A. F. KAHLBAUM, Chemkals- 


CARL ZEISS. Microscopes. 

FRANZ SCHMIDT & HAENSCH, Polariscopes and Spectroscoiies. 

DKSMOUTIS & CO., Purest Hammerert Platinum. 


GREINER & FRIEDRICH'S, Fine Stopcock Ware. 

MAX KOHL, Physical Apparatus. 

TENDER. Reagent Bottles with Black Indelible Letters. 

KAVALIER'S Famous Bohemian Glassware. 

The Above (inns ire koown to furnish the best goodi in theii respective lines 
We ue Iheir repreientKti>e« for the United Stale* ind CanulB ind carry a consiitei 
able nock of IhcM goodi on hand. Kohl'i Appanlni and Stender's Bottlci arc 
nporled to order only. 

Jena Glass the Olass of the Fntnre 

Neumann Bros. 

Book Binders 

6 and 7 EAST 16TH ST., 



Books bound in any style. 
Single volumes or in quantities 

Woodbridge School, 


School of Mines Preparatory School. 

417 Madison Avenuie, 

Between 48th and 49th Streets, NEW YORK CITY 

Twentieth Year Begins October ist, igoi. 

THE school is well equipped with phy-sical and ^chemical laboratories, in which the 
students are required to perform a complete set of experiments illustrative of their 

recitations in physics and general chemistry. A special laboratory is devoted to 
qualitative chemical analysis for advanced students. 

Five hundred Students of Columbia School of Mines have been instructed in the 
Woodbridge School. Also a large number have been prepared for Massachusetts In- 
stitute of Technology, Stevens Institute, Sheffield Scientific School, Lawrence Scien- 
ific School, Troy Polytechnic Institute, Cornell University, and the Classical, Medical, 
and Law Departments of Harvard, Yale, Columbia and Princeton. 

A summer school for students who have failed in June begins August 13th and 
coaches men for the Fall examinations. All classes are limited to five. College men 
are coached in Freshman and Sophomore Mathematics and Quantitative Analysis. 

An advanced course for older students prepares them to <inter the second year class 
Last June one of our students received his degree in Electrical Engineering in 
three years.