Skip to main content

Full text of "Exploration and survey of the valley of the Great Salt Lake of Utah, including a reconnoissance of a new route through the Rocky Mountains"

See other formats


= o E : : 
me) 
rote > = > 2 
be ce = = - 
ps m ” aan w 
VOILALILSNI NVINOSHLINS sa lYyVvud riot BRARI ES  oMITHSONIAN 
iE = ; Z ae Z z 
Z : 8 = 3 
a % ay. n . 
oa 154 OQ 4 a O a 
a < a = = 
a = a 2 2 
SMITHSONIAN s 
me _ = ” 
Hu o uu us 
a. eee. ow aa pe 
x; = < ya <= 
oa = co e 0 
i © ~ oO ser 
od z a | Fred a 
JOILNOLILSNI NVINOSHLINS S31IYVYEII An 
ae z . z at 
= = = Ie - 
> be > - , Ed 
: = 7 i a 
ABRARIES SMITHSONIAN INSTITUTION NOILOLILSNI NVINOSHLINS Sadi yV. 
‘y ” = r wm = n Bil 
e = ES < =) 
er { Vif 
” 7p) es er n 
O ie ee 7 re) 
= ie z, ae =» 
ye % = = -fa a = = 
JOILNLILSNI_NWINOSHLIWS S23 hYVYditT LIBRARIES SMITHSONIAN INSTITL 
ra ™ Wf z wn” > . 
” te wn — Se 
= a = Oo Yin, = 
Cc bs ; = = Bis Vy EL, c 4 
= SN" a = or 7 
9 SX ae fe) —" fe) 
aad ee es | PA 
IBRARIES SMITHSONIAN INSTITUTION NOILNLILSNI NYVINOSHLIWS 
= i = a8 us: = od 
- js = ai \ . “ (aga 
Es Ps) > > ® se [sAS* 
i ¥ — ra aan — ie 
i 2 i 2 = Real 
a , a n se oo aie 
= ; 77) z wn . 
NVINOSHLINS sa iuvuad Mul BRARI ES SMITHSONIAN INSTITL 
= a = N e = z NG 
= S 2 FZ 5 WN. 
Nn y Ww) WY PT oe .\ 
— % = fe) = <\ 
ae = = = ty 
= a ay | ee 
SMITHSONIAN INSTITUTION NOILNLILSNI NVINOSHLINS S314V 
x. — om ps Wren WE oc 
< %, = <x om WS hg << 
co s co SoM! 5 
cae Zo O — Oo _ 2 
ie = - m a y 
—NVINOSHLIWS ~°3 luvug ia LIBRAR LES SMITHSONAN on 
2 Oo ce ee eis ae 


E al fy S (RO = GMD = SS 2 
ne ud => v . 
m7 A Uy = = ae 
- Gyp 7 FS & 
ae” 2 : mn 
aa op) _ (of) . 223 
Be PN! NVINOSHLINS SMITHSONIAN _INSTITUTI 
> wn” = ” => ss 
= = a = <t We 
= = oe =e ae ‘ww 
oD ny D ra ae 
i oo ae ae O sm 
= = = x bE 
@ = e a » 2B = . 
RAR | BS MITHSONIAN INSTITUTION NOILAIILSNI NYINOSHLIWS Sa luYuVvud 
im a 2 a oa. 8 
a. fie oc = oc 
oc or oc c oc 
5 = co m 
~ - ‘i . _ oO . 7 
. eal z< e — 
NLILSNIT NVINOSHLINS SAIYVYdIT LIBRARIES SMITHSONIAN” INSTITUTIC 
Bane: = id = : 2 
0 XY 5 Pe = ae 
> SA oe > >, 
~~ SS ae FG is 
4 VN Lise = = ou 
—_ w nae w 
SMITHSONIAN INSTITUTION NOILNLILSNI NVINOSHLINS S3IY4VvUug 
” = * 2) ma 02) ; 
. = <x = <x = 
; = z = 2 a 
IN a O eM EO = 
NS Wy) Ww oe Ce} SD 2] 
» XS oO =I. a cD , ae O 
KOZ E 2, x = 
ae eS = = 
MLILSNI_NVINOSHLINS S3IYVYSIT LIBRARIES INSTITUTIC 
ane 2 E hd > : ” Zz 
i =" xn “& a = a fr? 5 
mee oS OS 6 Be S 
” 2 me 4 2 a Z 
RARIES SMITHSONIAN INSTITUTION NOILNLILSN S3SIYVug. 
if z pa z 
. ie 2 cn = 
hc 5 Po) 5 > o) 
|, GY ~ E cs - 
INLILSNI_NVINOSHLINS S31uvuE IT_LIBRARIES, SMITHSONIAN — 
= = a < 
bs = 4 ig ~i z 
"hn wy D a a 
ys rs oO 5 = O ee 
| lp = = = <= = 
— ce a: “es: 
RAR | ES SMITHSONIAN INSTITUTION NOILOLILSNI_NVINOSHIINS S31YVY aI 
a. ts fd ee. ac 3 ASN a 
hey EDO) RR : 
ao 7 “ny S 
rt tip = cm = ie NY on 
ier " z at ‘> Soe 
ALILSNI_NVINOSHLINS 24 luVvad pe LIBRARI ES” SMITHSONIAN” INSTITUTIO 
| =a cee iy Sa gl Ps 


i 


we 


Tih 
RA FS 
Poa 4: 


. 7ei os 
ri 


oy 
: 7 


NNN NEN AICTE Ny RT ma MRCP YA OYTO NVA WT EE LONE EE — 


cP Mempeeey ge he SOTERA SY de q 


re 


$l | Div. Insects 
AS | 3, &. Nat Mus 
eee) fet, 2 
EXPLORATION AND SURVEY 
NALL EY 
GREAT SALT LAKE OF UTALL- 


A RECONNOISSANCE OF A NEW ROUTE THROUGH 
THE ROCKY MOUNTAINS. © 


BY HOWARD STANSBURY, 


CAPTAIN ‘CORPS TOPOGRAPHI CAL ENGINEERS, 
U. S. ARMY. 
Bin ony AVCHD coe 


PRINTED BY ORDER OF THE SENATE OF THE UNITED STATES, 


PHILADELPHIA: 
LIPPINCOTT, GRAMBO & CO. 


1852. Ee \ HSONIA ; 
MAR 1 1979 


LIBRARIES 


Bureau ToroGRAPHICAL ENGINEERS, 
19th April, 1852. 


SIR: 

I have the honour to submit a copy of CAPTAIN STANSBURY’S 
Report of his Expedition to the Salt Lake, called for by a resolu- 
tion of the Senate of the 12th of March, 1851. 

Respectfully, Sir, 
Your obedient servant, 
J. J. ABERT, 
Col. Corps Top. Eng’rs. 


Hon. C. M. Conran, Secretary Department of War. 


War DEPARTMENT, 
Washington, April 19th, 1852. i 
Hon. W. R. Kine, President of the Senate. 
SIR: 

In compliance with a resolution of the Senate, passed March 
12th, 1851, I have the honour to transmit herewith «A Copy of 
the Report of Captain Howarp Srtanspury, of the Corps of 
Topographical Engineers, of his Exploration of the Valley of the 
Great Salt Lake.” | 

I have the honour to be, 
Very respectfully, 
Your obedient servant, 
C. M. Conrap, 
Secretary of War. 


~~ 


Printed by T. K & P. G. Collins. 


“FR >’. t ; 


are fF ahi 


TSIM AWRY SG 


INTRODUCTION. 


In preparing this Report of the Exploration of the 
Valley of the Great Salt Lake, I have occasionally availed 
myself of the notes and journals of the other members of 
the expedition, where they tended to elucidate facts of 
which I was not personally cognizant. Generally, those 
here given are the result of my own observation. Itisa 
subject of much regret that the exigencies of the service so 
hastened our departure, as to give but little time for prepa- 
rations so necessary to the proper outfit of a party about to 
engage in an extensive exploration. The instruments that 
could be obtaimed upon so short a notice, were not of a 
character suited to the purposes for which they were 
required; and the want of such facilities proved the occasion 
of no little vexation and delay. ‘The pressure upon the 
Bureau of Topographical Engineers would allow of the 
detail of but a single officer to aid me—a force entirely 
inadequate to the satisfactory performance of the multi- 
farious and arduous duties required in the course of so 
long, and so widely-extended, an examination. The ill- 
ness of that officer, during the whole of the journey from 
the Missouri to Green River, deprived me altogether of his 
much-needed services, and threw upon myself, alone, the 
whole burden of that portion of the reconnoissance. 

In a part of the Survey of the Utah Valley, we were 
aided by Lieut. G. W. Howland, of the regiment of Mounted 
Rifles, who was detailed from the command at Cantonment 

3 


4 | INTRODUCTION. 


Loring, for the purpose; but who, before its completion, 
was required to rejoin his regiment, for service on the 
Pacific coast. 

In the Department of Natural Science, from my very 
limited time, I was not successful in securing the services 
of a competent assistant. Yet, although as much has not 
been accomplished as I had anticipated, it is hoped that 
some additional light has been thrown upon the Geological 
formation and Natural History of these almost unknown 
regions. The papers of Professors Baird, Haldeman, 
Torrey, and Hall, together with the analyses of Dr. Gale, 
will not be without interest to the lovers of science. To 
these gentlemen, and to Messrs. Girard & Peale, I am 
much indebted for the labours which, from a regard to the 
general interests of science, they have bestowed toward 
rendering the present report more complete and _ satis- 
factory. . 
In what has been said respecting the Mormon commu- 

nity, [ have endeavoured frankly to present the impres- 
sions produced upon my mind by a somewhat intimate 
acquaintance of a year’s duration with both rulers and 
people. The intelligence of their organization into a 
Territorial Government, had not reached the valley when 
we left it. How far the change in their relations to the 
country, may, as has been asserted, have revolutionized 
the feelings of the people, it is impossible for me to say. 
But no representations, that have yet been made public, 
have served in the least to alter my expressed opinion of 
their character for either love to the country, or loyalty 
to the government. Since the return of the expedition, it 
has appeared evident that the nature of the domestic rela- 
tions of the Mormons has been very generally misappre- — 
hended. It seems that the “spiritual wife system,” as it 
has been very improperly denominated, has been supposed 
to be nothing more nor less than the unbridled license of © 
indiscriminate intercourse between the sexes, either openly 
practised by all, or indulged, to the invasion of individual 


INTRODUCTION. \ 5 


rights, by the spiritual leaders. Nothing can be further 
from the real state of the case. The tie that binds a Mor- 
mon to his second, third, or fourth wife, is just as strong, 
sacred, and indissoluble, as that which unites him to his 
first. Although this’ assumption of new marriage bonds 
be called “‘ Sealing,” it 1s contracted, not secretly, but under 
the solemn sanctions of a religious ceremony, in the pre- 
sence, and with the approbation and consent of relatives 
and friends. Whatever may be thought of the morality of 
this practice, none can fail to perceive that it exhibits a 
state of things entirely different from the gross licentious- 
ness which is generally thought to prevail in this commu- 
nity, and which, were it the case, would justly commend 
itself to the unmingled abhorrence of the whole civilized 
world. ‘The recent acquittal of a Mormon Elder for shoot- 
ing the seducer of one of his wives, on the ground that the 
act was one of justifiable homicide, fully corroborates the 
truth of this remark, and shows in how strong a light the 
sacredness and exclusive character of such relations are 
viewed by the Mormons themselves. 

The route pursued by the expedition on its return, 
through a pass in the mountains hitherto unknown, will, 
perhaps, lead to further investigation of that remarkable 
depression lying between the Park Mountains and the 
South Pass. That a feasible route may be traced through | 
this depression has been satisfactorily demonstrated; and 
the saving in distance cannot but prove an object of im- 
portance, either in the establishment of a post route, or in 
the construction of a railway communication across the 
continent. The development of the inexhaustible mine- 
ral resources of the coal basin of the Green River valley, 
may be found, and at no very distant period, to go far 
toward lessening the obstacles which at present exist in 
the settlement of a country so destitute of other fuel. It 
is to be hoped that the government will not discontinue 
the further examination of these most interesting regions. 
_ From the heads of the Arkansas to the northern boundary 


6 INTRODUCTION. 


of the republic, hes a field possessing mineral and agricul- 
tural resources sufficient, were they more fully known and 
explored, for the sustenance of a population equal to that 
of the original thirteen states of the Union. Constituting, 
as it does, a sort of neutral ground between widely sepa- 
rated portions of this vast country, what can be more 
obviously desirable than that its character should be more 
fully known, its hidden sources of wealth developed, and 
rendered available to the enterprise of our ever advancing 
population ? | 

In conclusion, I take much pleasure in acknowledging 
the efficient and faithful services of my friend and assist- 
ant, Lieut. J. W. Gunnison, of the Corps of Topographical 
Engineers. To high professional skill, he added energy, 
judgment, and an untiring devotion to the interests of the 
expedition, which very materially contributed to its success. 
Whilst confined to winter quarters in Salt Lake City, 
he paid particular attention to the religious doctrines and 
practices of the Mormon Church, the results of which, as I 
understand, he is about publishing to the world. The 
subject will doubtless prove of great interest to the theo- 
logian, and, indeed, to all who have watched, with any 
attention, the progress, in this country, of the various ex- 
travagant theories, civil and religious, which form so 
marked a characteristic of the present age. 


CONTENTS. 


CHAPTER I. 
FROM FORT LEAVENWORTH TO FORT KEARNY. 


Fort Leavenworth—Cholera—‘“ Corralling” on the plains—Sauk Indians—Big 
Blue River—Emigrant graves—Pawnee horse-thieves—Independence road 
—California emigrants—Boston pack company—Dutch gold-hunter—Storm 
on the prairie—Game— Platte valley—Buffalo—Scarcity of fuel—Fort 
MR Ve Shits Sahin «Goan uals eruhaliaive.ca staid de's'n she tuhwaln ch noisheaaie hob pisiodahe Shiv velve stakes lage 


CHAPTER II. 
FROM FORT KEARNY TO FORT LARAMIE. 


Character of the Platte valley—Absence of timber—Fossils—Forks of the 
Platte—Action of water—Prairie-dog village—Buffalo meat—Fossil bones 
—Crossing of the South Fork—Sioux dead lodge—Ash Hollow—North Fork 
of the Platte—Fourth of July on the plains—Sioux burial—Indian girl— 
Inhumanity of emigrants—Sioux villages—Cholera—Sioux moving—Court 
House Rock — Lignite-Capitol —Chimney Rock — Scott’s Bluff— Fort La- 


CHAPTER III. 
FROM FORT LARAMIE TO FORT BRIDGER. 


Limestone quarry—Warm Spring—Destruction of property by emigrants— 
Gypsum—Deer creek—Coal—Crossing of North Fork—Coal on North Fork 
—Red Buttes—Abandoned wagons—Dead cattle—Gales on the plains— 
Poisonous alkaline springs and incrustations—Independence Rock—Devil’s 
Gate—Sweetwater River—Wind River Mountains—Saleratus Lake—De- 
serters—Cajfion of the Sweetwater—South Pass—Pacific Springs—Emigrant 
ruse—Fossil trees—Green River—Fort Bridger—Uintah Mountains—Road 
to Fort Hall—Road to Great Salt Lake...............scsececsscsccesesesonsesces sevens 


CHAPTER IV. 
FROM FORT BRIDGER TO THE CITY OF THE GREAT SALT LAKE, 


Reconnoissance between Fort Bridger and the Northern end of Salt Lake— 
Bear River—Medicine Butte—Pumbar’s Creek—Ogden’s Hole—Shoshonee 
7 


13 


31 


54 


CONTENTS. 


root-diggers—Pass through Wahsatch Mountains—Valley of the Great Salt 
Lake—Brown’s Settlement—Remarkable hospitality—Rumoured opposition 
of the Mormons to the exploration—Its causes and removal—Co-operation 
OUMEMG ANE ROTIGeS His ste vscddscestcc duvonenee'deutransioyvacertssmeen eet ote betede eeamteeene Page 


CHAPTER V. 


Reconnoissance from Salt Lake City to Fort Hall—Bear River—Valley of the 
Malade—Wahsatch Mountains—Trace from Sheep Rock—Hedspeth’s Cut- 
off—Valley of the Pannack—Wind-mill Rock—Port Neuf River—Valley of 
Lewis’ Fork of the Columbia—Fort Hall—Cantonment Loring—Return to 
Salt Lake—Reconnoissance of Cache Valley—Road from Fort Bridger 
throumh, this valbeyiies sss. dnadsccter sodebeteeds soisvcuwaks sanththashordedeen avons cievewa 


CHAPTER VI. 


Reconnoissance of the Western shores of the Great Salt Lake—Obstacles to be 
- encountered—Desert character of the country—Brackish water—Indians 
—wWater fowl—Salt and Sulphur springs—Salt Lake from Promontory 
Point—North Shore of the lake—Water levels—Large Inland sea—Desert 
plains — Scarcity of water—Suffering of animals — Magnetic iron ore— 
Ancient beach of the Lake-Island Mountains—Extensive mud plains—A 
Sunday’s march —Bivouac—Ancient crater—Field of solid salt—Pilot 
Peak—Shoshonee winter lodges—The seventy-mile desert—Road to Hum- 
boldt’s River—Another ancient crater—Abandoned wagons and emigrants’ 
‘‘cache”—Spring valley—Warm springs containing fish—Tuilla valley— 


South shore of the lake—Salt Lake City—Results of the reconnoissance... 


CHAPTER VII. 


Suspension of the survey for 1849—Progress of the work, and obstacles to its 
prosecution—Winter in Salt Lake City—Snows in the mountains—Sketch 
of early history of the Mormons—Exodus from Illinois—Journey through 


the wilderness—Mormon battalion—Arrival in Salt Lake valley—Founding 


‘of the city—Scarcity of food—Improvements and settlements—Establish- 
ment of:a provisional government—State of Deseret—Warm and hot springs 
—Irrigation—Public buildings—Currency—Connection of church and state 
—Revenue—Tithes—Character of ‘‘the President’”—Treatment of emi- 
grants—Manners and customs—Claims of the Mormon church—Priesthood 
—Book of Mormon—Spiritual revelations—Miracles—Power of the priest- 
hood—Polygamy—Sealing—Practical working of the system—Effects to 
be anticipated—Perpetual emigration fund—Route to San Diego—Propa- 
gandism—Sources of wealth—Agriculture—Soil—Area of cultivation— 
Lake Utah—River Jordan—Settlements within the Great Basin, and man- 
ner of forming them—Manufactures—Education—Territory of Utah—Ap- 
pointment of Mormons to office—Loyalty of rulers and people—Brigham 
Young—Indians—Mormon difficulties with the Utahs—Zoological collec- 


SF 


97 


CONTENTS. | 9 


CHAPTER VIII. 


Survey of Salt Lake—Mouth of the Jordan—Antelope Island—Formation of 
salt—Frémont’s Island—Ancient water levels—Egg Island—Carrington’s 
Island—Hat Island—Volcanic character—Bear River Bay—Mud Island— 
Larve of insects—Curious excrescences—Landing to encamp—A night on 
the mud flats—Alum Cliff—Stansbury’s Island—Black Rock—Flat Rock 
Point—Indian springs—Shoshonee Indians—Silk plant—Spring Bay— 
Larve beds—Bitumen—Density of the lake water—Gunnison’s Island— 
Water fowl—Eggs—Northern bay of the lake—Western shores—Fresh 
water—lIts scarcity—Dolphin Island—Field of salt—Perilous adventure— 
Strong’s Knob—Sand flats—Utah digger—A night voyage—Bay on Antelope 
Island—The dome—Completion of the survey of the lake—Triangulation— 
Salt water bathing—Termination of the exploration of the valley—Sum- 
mary—Hospitality of the Mormons............sccsceesseveseee sovensees soeeesees Page 151 


CHAPTER IX. 


Departure from Great Salt Lake City—Golden Pass of the Wahsatch moun- 
tains—Snows in the cafions—Camass Prairie—Heads of the Timpanogas— 
Proposed change in the location of the present route through the Wahsatch 
range—Storm in the mountains—Singular dyke—Red Fork of the Weber 
—Mormon emigrants— Echo Creek—Cache Cave—Yellow Creek — The 
Needles—Bear River—Coal, and mineral tar—Eastern rim of the Great 
Basin—Propesed route for Pacific railroad, from Fort Bridger to the valley 
of Salt Lake—Fort Bridger—Military post—Exploration of a new pass 
through the Rocky Mountains—Valley of the Colorado of the West—‘‘The , 
church”—Pilot butte—Charge of Indian warriors—Green River—Bitter 
Creek valley—Beds of bituminous coal—The ‘“ Haystack” —*“ Red Gate” — 
Muddy Creek—Trappers’ tale—Beaver dams—Dividing ridge between the 
waters of the Atlantic and the Pacific— North Fork of the Platte—In- 
dian forts—Cotton-wood parks—Medicine-bow Butte—Buffalo-butchering, 
‘‘mountain fashion” — ‘Running buffalo” —Laramie river—Alarm: of 
indians—Ogallalah Sioux—Language of signs—Sioux village — Buffalo 
feast—Laws of hunting—Black Hills—Heads of Crow Creek—Cheyenne 
Indians—Cheyenne Pass—Lodge Pole Creek—Successful result of the re- 
connoissance—Rail road to Salt Lake—Different routes to the valley of the 
Humboldt—Mormon route to San Diego—Route from Salt Lake City to the 
Humboldt—Chugwater valley—Fort Laramie—Fort Leavenworth.......... 217 


APPENDIX A. 


Tables of Measured Distances from the Missouri River at Fort Leavenworth, 


to Salt Lake City.......c...ssscce avers misia ebluikia sive cee alee las Sunatseiiee eis tees? asa) sie 270 
From Salt Lake City to the Missouri, upon the Return Route............ssecscees 278 
Hromsalt Lake City to Wort. Hall in, Oregons sinsesess ocsvnccsy ocesepvecinevnnrians saacci 293 


From Salt Lake City to the San Pete Settlement ...........cccsssscssneesseseceeesenes 


10 . CONTENTS. 


APPENDIX B. 


Latitudes and Longitudes of the principal Triangulation Stations.......... Page 297 
Tabulation of the Triangles developed in the Survey of the Great Salt Lake.. 298 
Table of Geographical Positions........ccccseeeseeeeee eaaisns sennineciowienekoeietceleeia eta 300 


APPENDIX C. 


ZOOLOGY. ; 
Mammals, by Prof. Spencer, F. Bairdss i sevnsccctctenscicescevesuceeh seuepeus sepeneeee 309 
Birds, by Prof. Spencer F. Bairdici.5..5:sswonstiodsesueuenavesvecweeuenssuthert caliente siete 314 
Reptiles, by Prof. Baird and Charles Girard..........00+sssccsecsecsenseesieneees seesiss 336 
Insects, Dy Prof. Haldeman... seninassecamendies atierelievees sieeve nina 'ciicist Wesel eeaneanaame 366 
On certain Insect Larve, by Titian R. Peale...........000 areepewoeds (tinieblels ejecta . 379 
APPENDIX D. 
BOTANY. 
Catalogue and Description of Plants collected, by Prof. John Torrey.......... . 383 


APPENDIX E. 
GEOLOGY AND PALHONTOLOGY. 


Letter from Prof. Hall of New York, on the Geology and Paleontology of the 
GOMBIARY faee acters eee cebtene tee cisinalsnieWcelar une nbesiaeels sa a. Wenbievinieninslsiis’ melee osu esletiaealscels seat 401 


APPENDIX F. 
CHEMICAL ANALYSES. 


Chemical Analysis of the Mineral Waters and other specimens collected, by 
Drs Die DG ale cu cipew cappnieaeastencs'escsiceecensue/noul's at snicsssaisha's orate tsa. asecieess nema 417 


APPENDIX G. 
Meteorological Observations ........ceccsscosece cones Giscsascscunseeedse cose eaten eeme cannes 423 


LIST OF ILLUSTRATIONS. 


View of Fort Utah, on the Timpanogos. 

Fort Laramie. 

Scene in the Black Hills—Bitter Creek Valley. 

Crossing of the Platte. 

Fort Bridger. 

First View of Great Salt Lake Valley, from a Mountain-pass. 
Panoramic View—Across Bear River Bay—Great Salt Lake. 
West End of Frémont Island, and Antelope Island from Alum Bay. 
View looking North-west from Promontory Point. 

Station—East End of the Base Line, Salt Lake Valley. 

Great Salt Lake City, from the North. 

Street in Great Salt Lake City. 

Hot Springs—three miles from the City. 

Bowery, Mint, and President’s House—Great Salt Lake City. 
Fort Utah, or Provaux City—Utah Valley. 

Utah Prisoners, under the Common Platform—Fort Utah. 

Old Elk and his Squaw—Utah Indians. 

Baron La Hontan’s Map of ‘‘The Long River.” 

Station, and Mass of Mica Schist—Frémont’s Island—Great Salt Lake. 
Cave on Frémont’s Island. 

East End of Frémont’s Island and Promontory Range, looking North. 
Landing to Encamp—Bear River Bay—Great Salt Lake. 

Valley between Promontory Range and Rocky Butte—Bear River Bay. 
Camp No. 4, near Promontory Point. 

Flat Rock Point. 

Part of Western Slope of Promontory Range. 

Gunnison’s Island—Eastern Shore. 

Gunnison’s Island and Station, from the North. 

Limestone Cliff—North End of Gunnison’s Island. 

View from Strong’s Knob. 

Carrington’s Island. 

East Side of Stansbury’s Island. 

Peak on East Side of Stansbury’s Island. 

Entrance to the Valley of the Weber. 


Valley of the Jordan, from the Mouth of the She-rente. 
11 


13 LIST OF ILLUSTRATIONS. 


y 
REPTILES. 

Siredon Lichenoides, Plate I 

Cnemidophorus Tigris, Plate II. 

Crotaphytus Wislizenii, Plate III. 

Elgaria Scincicauda—Plestiodon Skeltonianum, Plate IV. 

Sceloporus Graciosus—Uta Stansburiana, Plate V. 

Holbrookia Maculata—Phrynosoma Modesta, Plate VI. 

Phrynosoma Platyrhinos—Phrynosoma Douglassii, Plate VII. 

Phrynosoma Cornuta—Phrynosoma Coronata, Plate VIII. 


INSECTS. 


Labidus Saji—L. Harrisii—L. Melshajmeri—Euphoria Cernii—Cotalpa Grani- 
collis—Henous Techanus—Megaderus Corallifer—Cicada Striatipes—C. Rer., 
Plate IX. 

Zaitha Bifoveata—Cdipoda Corallipes—Ephippiger Trivavensis—Anabrus Sim- 
plex, Plate X. 


PLANTS. 


Strepanthus Crassicaulis, Plate I. 
Phaca Mollissima, Plate II. 
Cowania Stansburiana, Plate III. 
Spirza Dumosa, Plate IV. 
Heuchera Rubescens, Plate V. 
Chenactis Achillezfolia, Plate VI. 
Monothrix Stansburiana, Plate VII. 
Crepis Acuminata, Plate VIII. 
Amblirioa Pudicum, Plate IX. 


PALZONTOLOGY. 


CORALS—CYATHOPHYLLIDA. 
Faviphyllum? rugosum, (n. sp.) F. multilamella. F. Stansburii—Lithostrontion 


—(sp. indet.), Plate I. 
BRACHIOPODA. 


Terrebratula subtilita—Spirifer hemiplicata. S. octoplicata—S. triplicata, Plate 


II. 
Chonetes variolata—Productus costatus—P. Pome Hen tee, P. — (sp. indet.) 


Orthis umbraculum? Plate III. | 

. ACEPHALA. 

Avicula? custa. Tellenomya protensa—Cypricardia occidentalis. Allorisma ter- 
minalis, (n. sp.)—Nucula arata, (n. sp.) 

GASTEROPODA. 

Pleurotomaria coronula, (n. sp.) Euomphalus subplanus, Plate IV. 

Map of the Great Salt Lake, and adjacent country. 

Map of the Reconnoissance of the Country between the Valley of the Great Salt 
Lake and the Missouri River, at Fort Leavenworth. 


STANSBURY’S REPORT. 


CHAPTER I. 


FROM FORT LEAVENWORTH TO FORT KEARNY. 


Washington, March 10, 1852. 
CoLoNEL JoHN J. ABERT, 


Chief of Bureau of Topographical Engineers. 

’Srr:—I have the honour to submit to the Bureau of Topogra- 
phical Engineers the following report of the results of an expe- 
dition, organized in obedience to your orders of April 11, 1849, 
having for its object a survey of the Great Salt Lake, and an 
exploration of its valley. 

Your instructions required me to report to the commanding 
officer of the regiment of Mounted Rifles, at Fort Leavenworth, on 
the 10th of May following, and directed me to accompany those 
troops on their route to Oregon as far as Fort Hall, at which 
point I was to separate from the command and prosecute the 
examinations required. Owing to causes beyond my control, and 
of which the bureau is already aware, I did not reach Fort Lea- 
venworth until after the departure of the Rifle Regiment from 
that post, and was consequently obliged to make such change in 
my arrangements as the circumstances required. The necessary 
outfit and provisions were obtained from the proper departments 
of the army, and the party enlarged and well armed, to enable 
it to protect itself from any danger or depredation to which it 
might be exposed from tribes of roving or hostile Indians. I wish 
here to express my obligation to Colonel Sumner, the command- 
ing officer at Fort Leavenworth, and to the quartermaster, Major 
Ogden, as well for the prompt and efficient aid rendered by them 
as for the kind interest they evinced in the success of the expe- 
dition. very facility was cheerfully accorded, and every thing 
conducive to our comfort most liberally supplied. We were much 
delayed, however, by the heavy drafts made upon the resources 

, 13 


14 FROM FORT LEAVENWORTH TO FORT KEARNY. 


of the post for outfits and transportation furnished to several 
heavy trains for Oregon, New Mexico, and California, as well as 
by a panic occasioned by exaggerated reports of the existence of 
the cholera at the post; which caused the desertion of forty 
teamsters and mechanics in one night. Not a hand was to be 
hired, nor could the quartermaster furnish me with a single team- 
ster. I was consequently obliged to send an express to Kansas 
for the necessary additional force. 

Before leaving Fort Leavenworth, we were joined by a small 
party of emigrants for California, who desired to travel in our 
company for the sake of protection, and who continued with us 
as far as Salt Lake City. This proved a fortunate arrangement, 
since we thereby secured the society of an excellent and intelli- 
gent lady, who not only, by her cheerfulness and vivacity, beguiled | 
the tedium of many a monotonous and wearisome hour, but, by 
her fortitude and patient endurance of exposure and fatigue, set 
an example worthy the imitation of many of the ruder sex. 

The cholera had for a considerable time been raging on the Mis- 
sourl; and as we passed up, fearful rumours of its prevalence and 
fatality among the emigrants on the route daily reached us from 
the plains. On the day we left Fort Leavenworth, one member 
of our little party was carried to the hospital in a state of col- 
lapse, where he died in twenty-four hours. The only officer 
attached to my command had been ill for several weeks, with 
severe attacks of intermittent fever, which now merged into chronic 
dysentery, and he was, in consequence, unable to sit on his horse, 
or to do duty of any kind. These were rather discouraging cir- 
cumstances for an outset; but, at length, on the 31st day of May, 
our preparations being completed, we commenced our journey, my 
own party consisting in all of eighteen men, five wagons, and 
forty-six horses and mules; while that of Mr. Sackett, our fellow- 
traveller, contained six persons, one wagon, one travelling carriage, 
and fifteen animals. Lieutenant Gunnison, being too ill to travel 
in any other manner, was carried on his bed, in a large spring 
wagon, which had been procured for the transportation of the 
instruments. The weather, in the morning, had been dark and 
lowering, with occasional showers, but it cleared off about noon; 
the camp broke up; the wagons were packed, and we prepared to 
exchange, for a season, the comforts and refinements of civilized 
life, for the somewhat wild and roving habits of the hunter and 
the savage. My party consisted principally of experienced voy- 


EMIGRANTS TO CALIFORNIA. 1d) 


 ageurs, who had spent the best part of their lives among the wilds 
of the Rocky Mountains, and to whom this manner of life had 
become endeared by old associations. We followed the «‘emigra- 
tion road,” (already broad and well beaten as any turnpike in our 
country,) over a rolling prairie, frmged on the south with trees. 
The hills consisted principally of carboniferous limestone, in ap- 
parently horizontal strata, which in places formed quite prominent 
escarpments. Our first day’s journey was only of six miles; but 
we were now fairly embarked, and things gradually assumed the 
appearance of order and regularity. 

Although the route taken by the party has been travelled by 
thousands of people, both before and since we passed over it, I 
have thought that some brief extracts from the daily journals of 
the expedition might not be without interest; for, although nothing 
very new may perhaps be elicited, still it is not improbable that 
they will convey, to such as peruse them, a more correct idea of 
what the thousands have had to encounter who have braved this 
long journey in search either of a new home in Oregon, or of that © 
more alluring object—the glittering treasure of California. 

Friday, June 1.—Bar. at sunrise, 20.86; Ther. 63°. The 
road for the first few miles wound along the fence of what ap- 
peared to be a large, neglected Indian farm, following for about 
nine miles the dividing ridge between the waters of the Missouri 
and those of the Kansas, with deep ravines inclining to the north- 
east. The ridge terminated in a steep hill, at the bottom of which 
we found Walnut Creek, running to the south. By an _ escarp- 
ment on the west side, the rock was found to be of the same 
character as that passed over yesterday. At 34 o’clock, we de- 
scended by a steep and somewhat dangerous road, to the valley of 
a small and beautiful stream running north, upon the left bank of 
which we encamped, near the edge of a wood, fringing the stream, 
in which black-walnut, white-oak, and hickory predominated. A 
short distance from the camp, to the north, are high limestone 
bluffs, without trees. 

In the course of the afternoon we passed the i grea of 
a Mr. Allen, consisting of about twenty-five ox-teams, bound for 
the land of gold. They had been on the spot ae days, de- 
tained by sickness. One of the party had died but the day be- 
fore of cholera, and two more were then down with the same 
disease. Inthe morning early, we had met four men from the 
same camp, returning on foot, with their effects on their backs, 


16 FROM FORT LEAVENWORTH TO FORT KEARNY. 


frightened at the danger and disgusted already with the trip. It 
was here that we first saw a train “corralled.”’ The wagons were 
drawn up in the form of a circle and chained together, leaving a 
small opening at but one place, through which the cattle were 
driven into the enclosed space at night, and guarded. The ar- 
rangement is an excellent one, and rendered impossible what is 
called, in Western phrase, a ‘‘stampede’—a mode of assault prac- 
tised by Indians for the purpose of carrying off cattle or horses, 
in which, if possible, they set loose some of the animals, and so 
frighten the rest as to produce aw general and confused flight of the 
whole. To a few determined men, wagons thus arranged form a 
breastwork exceedingly difficult to be carried by any force of un- 
disciplined savages.—Occasional showers during the day. Hven- 
ing clear and pleasant, with a bright moon. Day’s travel, twelve 
miles. 

Saturday, June 2.—Bar. 29.17; Ther. 64°. The general course 
to-day has been north-west, over a rolling prairie country, in- 
dented by deep ravines, formed by numerous small streams flow- 
ing into the Missouri, which runs eight or ten miles to the north- 
éast. In crossing a steep ravine in the forks of one of these 
affluents of that river, a part of one of the wagons was broken, 
the repair of which occupied the remainder of the day, and obliged 
us to encamp on the left bank of the stream, the bluff of which 
was quite steep. Near the top of the bank was a stratum of shale 
about two feet thick; the overlying limestone being considerably 
undermined by disintegration: over the limestone was a layer of 
light-coloured friable sandstone. In the shale, no fossils were dis- 
covered, but the limestone contained stems of encrinites. The 
strata appeared to be horizontal. Grass and water are here very 
abundant, and fine springs are to be found on the south side of the 
stream, which is richly wooded. Day’s travel, seven miles. 

Sunday, June 3.—Bar. 29.01; Ther. 80°. Camp not moved to- 
day. The cliff on the north side of the creek was traced for 
about a mile up the stream. The shale continues horizontal. In — 
some parts it was dark, and apparently carboniferous, but no fos- 
sils were discovered in it. Above it the limestone was sandy and 
ferruginous, and the upper layers contained many fossils,—spirifer, 
productus, &c.,—mixed with small shells. The cliff was from one 
hundred and fifty to two hundred feet high, facing north-west. 

Monday, June 4.—Bar. 29.18; Ther. 65°. The road in the 
morning was very sinuous, from its following the crest of a high 


SAUK INDIANS. ; 17 


ridge to avoid the spurs and ravines on the right. We are now 
fairly on the broad open prairie; the air fresh, cool, and delight- 
ful; the view on all sides very extensive. In the afternoon we 
were met by a small band of Sauk Indians, who presented a paper, 
written by some philanthropic emigrant, representing that as we 
were now passing through their country, consuming their grass, 
water, and wood, (the latter of which was very scarce) they wished 
to receive something by way of remuneration, whether money, bis- 
cuit, (of which they are very fond,) or tobacco. They were rather 
a fine-looking body of men, and seemed quite peaceably disposed. 

They were evidently on the look-out for the different companies as 
_ they passed, with the purpose of levying contributions. They ac- 
companied us to camp, and received some biscuit and tobacco, 
with which they seemed well satisfied. 

The formation passed over to-day has been the same as that 
observed heretofore, except that the shales appear to be rather 
more predominant, and the limestone more ferruginous, and per- 
haps more siliceous. Ina deep ravine the shales were very evi- 
dent, being in some places washed out to a great extent from the 
overlying limestone, which presented large tabular masses, in plaée, 
in which no dip was discovered. 

Tuesday, June 5.—Bar. 29.17; Ther. 70°. The country tra- 
versed to-day has been principally rolling prairie, rising gradually 
for about six miles; our road, following the crest of a ridge, with 
heads of ravines from the north and the south interlocking, was 
rendered both crooked and undulating. After tracing this winding 
track for some time, we entered the main emigration road from 
Kansas. Up to this point the road has been very good—smooth, 
hard, and dry, and free from abrupt descents or ascents. The 
country around us is entirely destitute of trees; not so much as a 
twig is to be seen; all is bald, naked prairie, with sweeping undu- 
lations of the surface, as if a heavy ground-swell of the ocean had 
been suddenly arrested and converted, by the wand of some en- 
chanter, into fixed and solid soil. 

_ Rain came on about noon, with occasional showers, until nearly 
sunset, when it cleared off, with high wind from the south. Ferru- 
ginous and slaty limestones were occasionally exposed in the ra- 
vines, very fossiliferous, containing principally testacea. Some 
stems of crinoidez were also found in the rock, in place. Plover 
and prairie hens were now seen very frequently; the former, how- 
ever, for the first time, and very shy. In the afternoon, we met a 

9 


18 FROM FORT LEAVENWORTH TO FORT KEARNY. 


small party of travellers, with a sick man ina wagon. They proved 
to be returning emigrants, who, after proceeding as far as Fort 
Kearny, had lost heart, sold out all they had, (their flour and 
bacon at one cent per pound,) and were now slowly and sadly 
wending their way back to their homes. They assured us that 
many more were in the same melancholy case. Day’s march, 
fourteen miles. 

June 6.—Camp up by4A.M. Bar. 28.75; Ther. 70°. Wind 
south-east ; clouds heavy and threatening. It shortly commenced 
raining hard, and continued until nearly noon. The ground to-day 
has been strewed with pebbles of granite, quartz, and porphyry, 
and also with large blocks of porphyritic granite. On the tops of 
the hills, limestone again appeared; it was non-fossiliferous, and 
rather sandy. About five miles from camp we crossed a small 
stream, from which were procured some specimens of spirifer. 
Under this rock was a non-fossiliferous stratum and then shale. 
The upper stratum was not in place. In the afternoon we passed 
a melancholy memento of disappointed hope and blasted enter- 
prise—four freshly-made graves of emigrants, who had died by the 
way, and were here left on the wide waste, with not a name to pre- 
serve their remembrance. How different such a fate from the high 
and sanguine prospects with which they had set out ! 

In the evening a heavy thunder-storm from the south-east, with 
rain and violent wind. Day’s travel, twenty miles. 

June T.—Bar. 28.48; Ther. 68°. The travelling to-day is 
heavy, in consequence of the rains of yesterday. The road lies 
through a rolling prairie and upon a ridge dividing the waters 
of the Missouri from those of the Big Blue river, a tributary of 
the Kansas. Met a Mr. Brulet, a French trader, from Fort La- 
ramie, with a large train of wagons, laden with packs of buffalo- 
robes, bound for St. Louis. He had been forty days on the road, 
and had met not less than four thousand wagons, averaging four 
persons toa wagon. ‘This large number of emigrants appeared to 
him to be getting along rather badly, from their want of experi- 
ence as to the proper. mode of travelling on the prairies, to which 
cause much of the sufferimg experienced on these plains is doubt- 
less to be ascribed. We availed ourselves of his offer to  — 
back letters to our friends at home. 

In the course of the moraing, passed the fresh grave of a poor 
fellow whose last resting-place ‘he been partially disturbed by the 
wolves. They had burrowed a large hole near the head, which, 


RETURNING EMIGRANTS. ° 19 


however, had been subsequently filled up with sticks by some com- 
passionate traveller. It was an affecting object, and no good omen 
of what might be looked for, should any of us fall by the way in 
our long and arduous journey. Upon a ridge near our noon halt, 
was found considerable detritus of primitive rocks, scattered over 
the surface of the ground, and many boulders of granite. Above 
this lay the limestone, the lower strata of which appeared to be 
composed of honey-comb limestone; the upper strata were more 
sandy and without fossils. After a march of seventeen and three- 
fourths miles, encamped on the left bank of what our guide called 
Legerette Creek. The banks, at the crossing, were high and steep, 
and afforded some very good sections of limestones interspersed 
with shales. A road had been made, with no little trouble, by the 
emigrants, down the banks on each side, and the crossing was tole- 
rably good. ‘The stream is thirty feet wide by one foot deep, and 
flows with a bold and rapid current into the Missouri. The strata, 
exposed by a section at the crossing, the direction of which was 
S.S. H. and N. N. W., were limestones with strata of shales, con- 
formable, with a marked dip of 12° to the west, and containing a 
considerable number of fossils, productus, spirifer, &e. Our camp 
for the night is situated on the edge of a beautifully broad and 
level prairie, nearly elliptical in form, almost encircled by this 
lovely stream, whose banks are fringed with the richest foliage: 
noble old oaks, elms, and walnuts overhang the water, with a back- 
ground of rising hills covered with grass and flowers. 

Friday, June 8.—Bar. at sunrise, 28.79; Ther. 68°. Wind 
north-east, cool and delightful. A small party, with a single 
wagon, drove into camp just as we were leaving the ground. They 
had formed part of a company from St. Louis, had proceeded 
within sixty miles of Fort Kearny, but had quarrelled, and be- 
come disgusted with the trip and with each other, and had sepa- 
rated. These persons were on their return to St. Louis. They 
gave discouraging accounts of matters ahead. Wagons, they said, 
could be bought, upon the route of emigration, for from ten to fif- 
teen dollars apiece, and provisions for almost nothing all. So 
much for arduous enterprises rashly undertaken, and prosecuted 
without previous knowledge or suitable preparation! What else 
could be expected? The road to-day has been quite circuitous, 
ranging from 8. by W. to W. by N. We have been following the 
ridge dividing the two main forks of Legerette Creek, just above 
the junction of which we encamped last night. Extensive grassy 


20 FROM FORT LEAVENWORTH TO FORT KEARNY. 


slopes descended from the road on each side of us as we gradually 
ascended the ridge. From our elevated position, the course and 
windings of either branch of the stream could plainly be traced 
by the fringes of rich timber which clothed their banks, while the 
dull uniformity of the prairie was agreeably relieved by the ravines 
of numerous tributary waters, extending almost to the crest of the 
ridge over which we travelled, and wooded to their very heads. 
As we continued to rise with the country, the graceful undulations 
of the naked hills and hollows contrasted agreeably with the waving 
lines of hickory and oak that marked the course of each little 
rivulet until it joined its destined stream, and formed together a 
landscape which, for extent and rich picturesque beauty, cannot 
easily be surpassed. ‘The country generally begins to present a 
more sandy appearance than heretofore, and the rock to be com- 
posed of extremely friable materials, from the decomposition of 
which results much sand impregnated with iron and shale, together 
with a great deal of gravel. The examination of a ravine near 
the morning’s camp, the north side of which was very precipitous 
and about a hundred and fifty feet in height, exhibited a section 
from N. to 8. of the upper strata of rocks, which were in this di- 
rection horizontal, but with a dip to the west. The strata con- 
sisted of layers of shales, sand, and detritus of older rocks, mixed 
with sand—all very friable.. In the bottom, limestone again ap- 
peared. On a hill opposite, the limestone was found cropping out 
with a dip to the west. The country to-day has presented a differ- 
ent aspect from that heretofore passed over; being intersected by 
deep ravines, most of which are heavily wooded to near the summit 
of the ridge. We encamped at the head of one of these, in a 
handsome grove of timber, after a drive of seventeen and a half 
miles. 

Saturday, June 9.—Bar. 28.66; Ther. 68°. Morning cool and 
sky overcast. The road continued to ascend for a few miles, when 
we crossed the Big Vermilion, (a tributary of the Big Blue,) which 
heads a mile to the N. H. ‘The crossing is miry. In the after- 
noon, encamped on the right bank of the Big Blue, near a spring 
of fine water, on the margin of a level prairie, bordered with huge 
trees, under the welcome shade of which we pitched our tents after 
a fatiguing march of twenty-six miles. 

The stream is here about seventy yards wide and three feet 
deep, flowing with a bold current, and is tolerably well wooded. 
We found the trees and stumps on its banks carved all over with 


BIG BLUE—TRADER’S GRAVE. , 328 


the names of hundreds of emigrants who had preceded us, the 
dates of their passing, the state of their health and spirits, to- 
gether with an occasional message for their friends who were ex- 
pected to follow. Such a record, in the midst of a wide solitude 
like this, could not but make a strong and cheering impression on 
every new-comer, who thus suddenly found himself, as it were, in 
the midst of a great company of friends and fellow-travellers. 
On the left bank was the freshly-made grave of a French trader, 
whose name was well known to most of our voyageurs. It was 
heaped up with earth and covered longitudinally with heavy split 
logs, placed there to prevent the depredations of the wolves; the 
whole being surmounted by a wooden cross, with the name of the 
deceased and the usual significant abbreviation, IHS, carved 
rudely upon it. We had passed six graves already during the day. 
Melancholy accompaniments they are of a road silent and solitary at 
best, and ill calculated to cheer the weary, drooping wayfarers. Out 
encampment was pleasantly situated under the spreading branches 
of some large oaks, with a spring of pure, cold water near at 
hand—the latter an item which we soon afterward learned to value 
beyond all price. Just above us was a wagon with a small party 
of emigrants. ‘They had lost most of their cattle on the journey; 
and the father of three of them having died on the road, they, in 
conformity with his dying wishes, were now on their return to the 
settlements. A short distance beyond these, we found another 
small company, who had been encamped here for twelve days on 
account of the illness of one of their comrades. ‘They also were 
on their return. Had we been going out on a private enterprise, 
discouragements were not wanting as well from the dead as the 
living. 

Since crossing the Vermilion, the character of the country has 
changed from that of a high and rolling prairie to a comparatively 
flat and elevated plateau, with the drains much broader and not so 
depressed as heretofore. The soil is much deeper, the trees larger 
and more numerous, and the water cooler and more abundant. 
White sandstone, light-coloured shales, some flints, and a few fos- 
sils, were passed during the day. At the crossing of a small 
branch, about two miles before reaching the Big Blue, the rock 
exhibited a section from north to south, nearly horizontal, with 
perhaps a slight dip to the south. It consisted of white limestone 
and strata of flint, with some imperfect fossils. The general sur- 
face of the rock is worn into escarpments in the shape of bastions, 


22. FROM FORT LEAVENWORTH TO FORT KEARNY. 


with numerous terraces rising one above the other, having a strik- 
ing and picturesque effect. Some shales are interspersed among 
the chalky limestone. Here also were seen several small boulders 
of red granite. Some good-sized catfish were caught in the Blue 
by the men, during the night. 

Sunday, June 10.—Bar. 28.82; Ther. 70°. The camp rested, 
it having been determined, from the commencement of the expedi- 
tion, to devote this day, whenever practicable, to its legitimate 
purpose, as an interval of rest for man and beast. I here beg to 
record, as the result of my experience, derived not only from the 
present journey, but from the observation of many years spent in 
the performance of similar duties, that, as a mere matter of pecu- 
niary consideration, apart from all higher obligations, it is wise to 
keep the Sabbath. More work can be obtained from both men 
and animals by its observance, than where the whole seven days 
are uninterruptedly devoted to labour. 

Very early in the morning it was discovered that three horses 
from our herd, and one from a neighbourmg encampment, had 
been stolen during the night, and that so adroitly as not to occasion 
the slightest noise, although our animals were all picketed in the 
very centre of the camp and within a few feet of the tents. Search 
was forthwith made for the trail of the robbers, which was soon 
found, and ascertained to be that of Indians. Two parties of 
scouts were despatched in pursuit. In the mean time, the ammu- 
nition chests were opened, additional cartridges served out, arms 
examined and reloaded, the men practised in shooting at a mark, 
and every preparation made to guard against a repetition of the 
outrage. In a few hours one of the parties, under Archambault, 
the guide, an experienced hunter and mountaineer, returned, but 
without the animals. They had crossed the Blue at the ford, and 
followed the trail of the Indians about six miles, but here the lat- 
ter had recrossed, and taken to the hard open prairie, where all 
further trace of them was lost, and the pursuit in consequence 
given up. They found, however, the spot where the Indian party 
had encamped the day previous, marked by the fragments of an 
ox they had just stolen from a neighbouring train. They had 
taken two; one they had killed and devoured, leaving in their 
haste the yoke and hide of the slaughtered animal, together with 
a small portion of the meat, while they made off with the horses, 
and drove the other ox before them. ‘The robbers were Pawnees, 
and had evidently been watching ever since our arrival, as they 


PAWNEE HORSE-THIEVES. 23 


had selected the very best horses in both trains, all of which, to 
make the matter worse, happened to be private property. Effec- 
tive measures should certainly be taken to punish and thereby pre- 
vent the occurrence of these outrages by a band of savages, who, 
although receiving a large annuity from the national treasury, 
take every opportunity to prey upon those under the protection 
of the government. Several large catfish and some soft-shelled 
turtle were caught in the stream by the men. The rich bottom in 
the rear of the camp produces strawberries of fine quality in the 
utmost profusion; the men gathered them by hatfuls. ‘Two very 
large terrapins were also found here on the prairie. 

In the afternoon, the advance of a train from St. Joseph, be- 
longing to Messrs. Bissonet and Badeau, bound on a trading expe- 
dition among the Sioux, passed the camp and halted on the bluff 
beyond. Mr. Bissonet, who is an old trader and appears to be 
well acquainted with the country, informed me that the stream 
called by our guides the Legerette is in fact the Nemaha; and 
that the streams called by Fremont, Great and Little Nemahas - 
are the waters of Turkey Creek, and flow into the Blue to the 
north of the road. A section of about one hundred feet high, in 
a ravine on the south side of the river, showed the strata to be 
horizontal from north to south, with a hip of ten degrees to the 
west. The order of superposition was as follows :—Lower, most visi- 
ble, red clay and sand; gray shales; blue limestone; gray limestone, 
and flint; white sandstone. They all contained fossils except the 
clay. <A species of mallowand @nothera occurred on the bot- 
toms of the streams, with Digitalis and Loasanitida. Phlox, once 
abundant, is becoming scarce. 

Monday, June 11.—Bar. 28.56; Ther. 65°. At half-past five 
o'clock, a most violent storm of wind and rain set in, and raged 
with great fury for three hours. The tents were prostrated, and 
the baggage much wetted by the rain. Several large trees were 
blown down, and one fell across an emigrant wagon close by us. 
The owners, who had sought refuge in it from the tempest, narrowly 
escaped with their lives. About nine, it cleared, and the tents were 
raised to allow them todry. Hight miles from the Blue, we struck 
the emigration road from Independence. Here we found a com- 
pany of seventy or eighty persons, with some twenty wagons, ou 
their way to California, among whom I recognised several former 
compagnons de voyage on the Missouri. After crossing Ketchum’s 
Creek, encamped a short distance to the right of the road, 


QA ~ FROM FORT LEAVENWORTH TO FORT KEARNY. 


on one of the head branches of Turkey Creek, the channel 
of which appeared to be cut deeply into the detritus of a sand-— 
stone slightly ferruginous. Near the Blue, the highest rock ob- 
served was ferruginous sandstone, and during the day an imper- 
fect section exposed shaly limestone and white sandstone. Our 
course has lately been rising with the country, and we have been 
traversing a sort of plateau, having, however, no very marked 
ridges; but being intersected in all directions by ravines, the sides 
of which have but a moderate slope and little fall, with water 
standing in pools. Passed six graves to-day. Day’s march, se- 
venteen and a-quarter miles. 

Tuesday, June 12.—Bar. 28.64; Ther. 63°. Breakfast at four. 
In ten and a half miles crossed the west branch of Turkey Creek, 
and halted to noon on the bank of Wyeth’s Creek, six miles be- 
yond. The crossing here is bad and rocky, and the grass poor, 
having been eaten close by the trains which had preceded us. 
The afternoon was oppressively hot and close, the wind being 
from the eastward, with every appearance of rain. We have been 
in company with multitudes of emigrants the whole day. The 
road has been lined to a long extent with their wagons, whose 
white covers, glittering in the sunlight, resembled, at a distance, 
ships upon the ocean. We passed a company from Boston, con- 
sisting of seventy persons, one hundred and forty pack and riding 
mules, a number of riding horses, and a drove of cattle for beef. 
The expedition, as might be expected, and as is too generally the 
case, was badly conducted: the mules were overloaded, and the man- 
ner of securing and arranging the packs elicited many a sarcastic 
criticism from our party, most of whom were old and experienced 
mountain-men, with whom the making up of a pack and the load- 
ing of a mule amounted to a science. We passed also an old 
Dutchman, with an immense wagon, drawn by six yoke of cattle, 
and loaded with household furniture. Behind, followed a covered 
cart containing the wife, driving herself, and a host of babies— 
the whole bound to the land of promise, of the distance to which, 
however, they seemed to have not the most remote idea. To the 
tail of the cart was attached a large chicken-coop, full of fowls; 
two milch-cows followed, and next came an old mare, upon the 
back of which was perched a little, brown-faced, barefooted girl, 
not more than seven years old, while a small sucking colt brought 
up the rear. We had occasion to see this old gentleman and his 
saravan frequently afterward, as we passed and repassed each 


STORM ON THE PRAIRIE. 95 


other, from time to time, on the road. The last we saw of him 
was on the Sweetwater, engaged in sawing his wagon into two 
parts, for the purpose of converting it into two carts, and in dispos- 
ing of every thing he could sell or give away, to lighten his load. 

(Enothera, with its bright yellow flowers, was frequent in the 
landscape, both to-day and yesterday, with Amorpha and Artemi- 
_ sta. The prairie-rose is becoming quite abundant. 

Near Wyeth’s Creek, a section showed the upper rocks, near the 
top of a ridge, to consist of dark ferruginous sandstone, under 
which were white clayey shales, the soil being formed principally 
by decomposed sandstone. 

After travelling twenty-six miles, we encamped on the level bank 
of Walnut Creek—a tributary of the Little Blue, with a tolerable 
supply of grass and water. Near this encampment, a small section 
showed the formation to consist of red sandstone, apparently hori- 
zontal, very friable, and containing traces of organic remains, but 
none recognisable. Under the sandstone were traces of shale, 
light-coloured and very sandy. Black iron-ore was discovered in 
the sandstone. 

Wednesday, June 18.—About two o’clock in the morning, the 
camp was suddenly aroused by the bursting upon it of a most fu- 
rious storm. ‘The wind blew a hurricane, the rain fell in torrents, 
while the thunder and lightning were terrible and incessant. For- 
_ tunately the camp had been pitched in a sheltered spot, or 1t must 
have been entirely blown away by the tempest: as it was, the 
tents were prostrated by the wind, and preserved with much dif- 
ficulty. Our men were exposed to all its fury for several hours. 
At length, however, the sky partially cleared, but the lowering 
enemy seemed still to linger, as if meditating another attack. The 
morning proved exceedingly hot and close; the barometer con- 
tinued to fall. Our poor mules having been picketed within the 
lines all night, and consequently exposed to the storm, seemed 
dejected, tired, and hollow; and altogether the camp seemed 
weary and dispirited. The weather looked so very doubtful that 
we did not move until half-past two o’clock; the men being until 
then engaged in drying their bedding, which had been thoroughly 
soaked by the rain. An immense number of black beetles and 
other insects swarmed around the camp last evening. Attracted 
by the light, they annoyed us beyond measure, and could be heard 
all night, pattering against the tents like large drops of rain in a 
heavy shower. , | 


26 FROM FORT LEAVENWORTH TO FORT KEARNY. 


Amorpha, Commelina, and Artemisia are still abundant, but the 
phlox is disappearing. In the bottom of the creek a species of 
larkspur and wild-onion abound; our men used the latter freely, 
and we found them quite palatable in flavourmg our bean-soup. 
Shortly after resuming the march, we reached the dividing height 
between a tributary of Hmigrant’s Creek and Sandy Creek, another 
affluent of the Little Blue. The view here was extensive and very 
beautiful. The Sandy and its numerous small tributaries could 
be distinctly traced in all their sinuous courses, by their dense 
bordering foliage, now turned up by the wind, and contrasting 
strongly with the encircling bluffs that stood out against the sky, 
without a single bush or twig to relieve the sight. Indeed, the ap- 
pearance of a tree, in all these regions of naked prairie, is to the 
traveller a certain indication of the presence of water; and in con-- 
sequence, numerous tracks are to be seen, leading off from the main 
road to some sheltered and sequestered grove, affording the wel- 
come indulgence of shade, wood, water, and grass to the weary 
emigrant and his still more wearied beasts. Elsewhere, during the 
long fatiguing day, shade there is none; unless, indeed, it be be- 
neath his wagon, which to him is literally his home. In it he car- 
ries his all, and it serves him as tent, kitchen, parlour, and bed- 
room, and not unfrequently as a boat, to ferry his load over an 
otherwise impassable stream. Many have no other shelter from 
the storm during the whole journey, and most of these vehicles are 
extremely tight, roomy, and comfortable. Encamped after a short 
march of five miles, on the right bank of the Sandy. In the after- 
noon we met four men from Fort Laramie, who informed us that a 
large band of some six hundred Pawnees had become so very trou- 
blesome to the emigrants between the Blue and Fort Kearny, that 
a force had been despatched from that post to drive them off. A 
small party had also been discovered afew miles back, lurking 
under a cliff of rocks, a short distance south of the road, doubtless 
on the watch for another chance to steal our horses. We were on 
our guard, however. The camp was formed by drawing the wagons 
into a semi-circle, resting on the stream, with the tents pitched 
alternately between them, except those of the officers, which were 
placed on the bank and faced the enclosure. Within this all the 
animals were carefully picketed, and a strong guard, well armed, 
set at dusk. ‘These precautions doubtless saved us from molesta- 
tion. The formation to-day consisted of white and red sandstones, 
principally the latter, being evidently the sandstone formation 


LITTLE SANDY—-VALLEY OF THE LITTLE BLUE. 27 


above the carboniferous series. The white sandstone was very 
fossiliferous. 

Thursday, June 14.—Ther. at sunrise, 66°; Bar. 28.58. 
Heavy dew; wind N. W., quite fresh and cool. Leaving Little 
Sandy, the road follows the ridge between it and the Little Blue, 
over decomposed red sandstone. Halted to noon on the right 
bank of the Big Sandy, a tolerably bold stream when the water is 
high; but at present it stands only in pools, on a bed of white sand two 
hundred feet in width. The afternoon march was on the south side 
of the ridge, which forced us to cross the heads of numerous ravines 
and sma!] runs putting into the Blue, upon one of which we encamped, 
having accomplished twenty-one miles. We met to-day, for the 
first time, several new plants, indicating an approach to regions of 
a different and less fertile character than those we had hitherto 
traversed. The aloe and the prickly-pear were found in the sand- 
hills, as were the Commelina and the saxifrage. The prairie-rose, 
Amorpha, Hinothera, and Artemisiaabound. A blue lupine and a 
white mallow were also gathered. 

In the afternoon we met Major Belger, of the quartermaster’s 
department, on his return from Fort Kearny, with an escort of 
dragoons. He had fallen in with a band of five hundred Pawnees, 
who, however, did not attack him. He confirmed the rumour that 
a force had been sent from the fort after them. A fight had taken 
place on the north side of the Platte, between the Indians and 
two parties of emigrants, in which the former were defeated, with 
the loss of their chief, five others killed, and six wounded; the 
whites having one man wounded and a horse killed. A large 
number of plover were seen to-day. 

Monday, June 18.—Bar. 28.138; Ther. 86°. We have been 
travelling for the last three days up the valley of the Little Blue. 
Where we first struck it, the stream is eighty feet wide, apparently 
deep, very crooked, with a swift current. It is fringed, sometimes 
on one side, sometimes on the other, with a narrow belt of cotton- 
wood and willows. The valley presents a tolerably uniform ap- 
pearance, bounded by ridges, seldom more than a mile or two apart, 
the intervening bottoms sloping gradually down to the river. The 
grass 1s generally very abundant, and préle (the common scouring- 
rush) is found in great plenty. Our mules ate it with avidity. 

In the morning we passed a government ox-train, laden with 
provisions for the new post about to be established in the neigh- 
bourhood of Fort Hall. It consisted of thirty-one heavy wagons, 


28 FROM FORT LEAVENWORTH TO FORT KEARNY. 


four hundred oxen, (five or six yoke to each team,) and about forty 
men. At night the wagons are drawn into a circle, in the open 
plain, away from any covert, and chained together by the wheels, 
leaving a small space. The cattle are driven, after feeding, into 
the enclosure thus formed, when the aperture is closed for the 
night, and a guard set. A very formidable little field-work is thus 
easily and rapidly constructed. In the morning and middle of the 
day the cattle are turned loose to graze, and a day-guard is de- 
tailed for their protection. This is the general mode adopted by 
' travellers on the plains for the security of themselves and their 
property. 

The valley of the Little Blue has not presented any great novelty 
in the way of flowers. The only new plants met with have been a 
lupine, the flower of which, of a bright purple, rises directly from 
the root; the plant is totally leafless. A splendid variety of the 
mallow, of a bright carmine colour, its trailing stems sending up 
flowers in little patches of a few yards square, presented a rich and 
beautiful appearance, enlivening the monotony of the prairie by its 
brilliant hues. The aloe occurred in some places in abundance ; 
and there were a few cacti, and a species of a leguminous plant 
was met with, having a flower of a pale purple colour, resembling 
a vetch; also a species of pale blue digitalis. 

Bestercs being Sunday, was devoted to rest. Most of the 
people, however, availed themselves of the opportunity to take a 
hunt, as we had killed no game up to this time. In fact, we had had 
no opportunity, the game having been driven from the vicinity of 
the travelled route by the unintermitted stream of emigration which 
had already passed over the road. The result of their efforts was 
accordingly not very magnificent, the whole party bringing in only 
a duck, a musk-rat, a large snapping-turtle, and one miserably poor 
little antelope. The constant use of salt meat, without vegetables, 
had affected us all with a cutaneous irritation, to be allayed only 
by the use of fresh meat; and hence the arrival of this antelope, 
poor as it was, was hailed by our voyageurs with lively satisfaction. 
The little carcass was cut up and divided among the several messes, 
a portion being sent to our travelling companions; and it was 
amusing to see how soon every one was sedulously engaged in pre- 
paring this most welcome addition to our usual homely fare. The 
scene was picturesque: the camp-fires blazed in every direction, 
while around each might be seen a busy little group, boiling, roast- 
ing, and baking, in happy anticipation of their venison dinner ; the 


VALLEY OF THE PLATTE. 29 


mules, meanwhile, filled to repletion with the rich grass of the 
prairie, lay stretched and rolling upon the grass in lazy enjoyment, 
exempt for one happy day from the harness and the whip. 

After travelling up the Blue for about twelve miles, we left it 
and crossed the ridge dividing its waters from those of the Ne- 
braska or Platte River. The character of this ridge is that of an 
extensive level plateau, or table, with slight undulations: the 
soil is composed of sand and clay, having occasionally water stand- 
ing on it in pools, which, however, are dry most of the summer. 
On arriving at the western edge of the plateau, the country became 
more elevated, and presented a range of small hills of a sandy 
reddish clay, with a sharp outline toward the river, forming the 
«‘coast of the Nebraska,”’ and also constituting the bluff bounding 
the river valley on the south. From this elevated position the valley 
presented a lovely appearance. ‘The bottom was as level as a floor, 
covered with short fresh grass of the richest green, without a 
shrub or bush to interrupt the view. Beyond this verdant carpet 
of two miles in breadth, flowed the river of which we had heard so 
much, while a dense growth of large timber, covering Grand Island, 
which lay immediately before us, formed a fit framework for this 
lovely picture of calm and quiet beauty. 

Archambault, our guide, told me that the last time he had passed 
this spot, the whole of the immense plain, as far as the eye could 
reach, was black with herds of buffalo. Now, not so much as one 
is to be seen; they have fled before the advancing tide of emi- 
gration. Driven from their ancient and long-loved haunts, these 
aboriginal herds, confined within still narrowing bounds, seem 
destined to final extirpation at the hand of man. ‘The prairie 
bottom of the Platte is here elevated but a very few feet above the 
river in its present stage, which, however, is higher than usual. 
The appearance of the water is precisely that of the Mississippi and 
Missouri, of a muddy white, and its current is, like theirs, con- 
stantly boiling and eddying in restless turbulence. It is quite 
shallow, as its name, both in Indian and French, indicates, so that 
I found no difficulty in riding my mule over to the island, at the 
head of which we encamped for the night, after a march of thirty- 
two miles. 

In the course of the morning we passed the remains of a Pawnee 
village, recently abandoned. The band to which it belonged was 
probably the same before mentioned as having been driven off by 
the troops from Fort Kearny. Near it, several large mushrooms, 


30 FROM FORT LEAVENWORTH TO FORT KEARNY. 


the first we had seen, were found. Cacti were met-.with during 
the day, also the purple mallow, as well as a small red species, the 
perfume of which is very sweet, resembling that of heliotrope. 
Dwarf Artemisia was also abundant. The measured distance from 
Fort Leavenworth to this point, by the usually travelled route, is 
two hundred and ninety-six miles. 

Tuesday, June 19.—Ther. at 5 o’clock, 70°. Men and animals 
much fatigued by the journey of yesterday. We travelled up 
the Platte fifteen miles, and encamped within two miles of Fort 
Kearny, on the bank of the river, for the sake of water and grass. 
Wood for cooking could be procured only by wading the river, and 
bringing it from the opposite side on the shoulders of the men. 
After encamping, rode up to the fort, and called upon the com- 
manding officer, Colonel Bonneville, whose adventures among the 
Rocky Mountains are so well known to the world. He received 
us very courteously, offering us every facility in his power in 
furtherance of our progress. We remained at this post until the 
afternoon of the 21st, to recruit the mules, get many of them shod, 
and to procure such necessary supplies as could be obtained. The 
post at present consists of a number of long low buildings, con- 
structed principally of adobe, or sun-dried bricks, with nearly flat 
roofs; a large hospital-tent; two or three workshops, enclosed by 
canvas walls; storehouses constructed in the same manner; one or 
two long adobe stables, with roofs of brush; and tents for the accom- 
modation of the officers and men. There are stationed here two com- 
panies of infantry and one of dragoons. I was told that the hail- 
storms had been very frequent this season and quite destructive, 
cutting down the weeds and stripping the trees of their foliage. 

Lieutenant Gunnison being still quite feeble, and unable to ride 
on horseback, I purchased for his use a little spring-carriage, 
which had been left here by a party of emigrants. Such abandon- 
ments are very common; most of these sanguine and adventurous 
companies, by the time they get thus far, beginning to find out that 
they have started on their journey with more than they can con- 
trive to carry. In order to lighten their load, most of them dis- 

_pose of every thing they can possibly spare, and at almost any price... 
‘Flour and bacon, for example, had been sold as low as one cent 
per pound; and many, being unable to sell even at that price, had 
used their meat for fuel. The pack company from Boston, which 
had passed us on the route, and which we found encamped here on 
our arrival, left before our departure. As they had been entirely 


FORT KEARNY. 1 


unaccustomed to the operation of packing, their mules, as was to be 

expected, were in a most horrible condition, with galled backs and 
sides that made one shudder to behold. The proper mode of 
arranging the load of these suffering animals is an art taught only 
by experience. ‘These people, though belonging to a race famous 
for foresight and calculation, had, like others from less thrifty and 
managing portions of the Union, been selling and giving away all 
they could dispense with. While encamped here we have had 
several severe thunder storms, accompanied with heavy rains and 
violent winds. 


CHAPTER II. 
FROM FORT KEARNY TO FORT LARAMIE. 


Thursday, June 21.—Having taken leave of our kind and 
hospitable friends at the fort, we overtook our own train, which 
had been sent ahead in the morning, and found them encamped on 
the bank of the Platte, after a drive of twenty-five miles. Lieu- 
tenant Gunnison, who had gone before in his little wagon, by 
some means missed the camp in the darkness, and did not arrive, 
which gave me no little uneasiness, lest the exposure should prove 
detrimental in his very delicate state of health. We discovered, 
however, in the morning, that he had found good quarters at an 
emigrant encampment on the road. 

The character of the Platte valley for the last forty miles is 
that of a flat prairie, composed of sand and clay, in which, when 
the latter predominates, water is found standing in small pools, 
but when the sand is most abundant, the water passes through it 
like a sieve and is quite drained away. Hence we have passed 
innumerable little wells, dug to a depth of from two to four feet. 
The water is generally clear and cool, but much of the sickness on 
this route has been attributed to its use. The soil thrown out is 
sandy, though not unfrequently having a mixture of clay. The 
water thus obtained is evidently the result of infiltration from 
the higher levels or bluffs, which, in this hidden manner, discharge 
their surplus moisture into the river. The bluffs on the opposite 
side of the river, near Fort Kearny, are apparently formed of 


32 FROM FORT KEARNY TO FORT LARAMIE. 


pure sand, and are much higher, and appear to be more wooded, 
than on the south side of the stream. About twenty miles above 
the fort, the character of the bluffs on the south side seemed 
changed, and presented a more gradual slope to the river; the 
soil contained more clay, and, at a distance of a mile back from 
their escarpment, they were cut up by constant ravines with banks 
precipitous and, in some instances, perpendicular. In one spot, 
such was the tenacity of the soil, that an upright mass of earth in 
the form of a column had been left by the waters. Here were 
found fossils of a character similar to those obtained at the point 
where we first entered the valley: they were, however, in a very 
friable and decomposed state. 

Sunday, June 24.—Bar. 27.56; Ther. 83. Our journey for 
the last two days has been up the valley of the Platte, which, in 
some places, is more than a mile in width. From one spot I 
counted upward of twenty islands, which, being densely covered 
with green willows and cotton-woods, presented, in contrast with 
the naked monotonous country through which we were passing, a 
perfect picture of refreshing beauty. From the fact that the 
islands in the river are, for the most part, covered with trees, the 
almost total absence of this feature in the landscape of the valley | 
must be attributed, in part at least, to the fires which periodically 
sweep over the country in the autumn, destroying every thing be- 
fore them. On our return by this same route, in the fall of 1850, 
the country, for more than three hundred miles, had been com- 
pletely devastated by these conflagrations, insomuch that our ani- 
mals came near perishing for want of herbage. The north side 
of the river does not appear to suffer so much from this cause; 
which may, in part, arise from the direction of the prevailing winds. 
Encamped on the bank of the Platte, fifty-six miles above Fort 
Kearny. The bluffs bounding the valley were of clayey soil, cut 
up by deep ravines, in many instances nearly perpendicular, their 
character becoming bolder as we advanced. ‘The soil is richer and 
contains more clay. ‘The plants seen were Tradescantia, the 
purple mallow, (the root of which resembles the parsnip, and is 
used by the Indians for food,) the small yellow Gnothera, and a 
pretty, small stellate-flowered plant. Over large portions of the 
bottom, no flowers were met with; on the high ground, red mallow, 
Mimosa, Linum, a white Mimulus, and a sort of larkspur. The 
aloe was flowering in abundance on the face of some very steep 
bluffs. 


FOSSIL REMAINS—FORKS OF THE PLATTE. oo 


Monday, June 25.—Ther. at sunrise, 64°; Bar. 27.52. The 
bluffs on our left, which are about two miles distant, are assuming 
a much more broken appearance than heretofore, being cut up into 
peaks and ridges in the most picturesque manner. Upon exa- 
mination they were found to be composed of sandy clay, intersected 
by precipitous ravines, the sections of which presented strata 
slightly differing in colour and hardness. The fossils collected 
were some teeth, apparently of an animal of the lizard tribe, and 
the femur either of a bird or a small lizard; the head of the 
bone and nearly the entire shaft measured three and a-half inches, 
but the latter crumbled on removing it. Both of these were found 
in place. Remains also were seen of what appeared to be bones, 
fully four or five inches in circumference, but in so friable a state 
that it was impossible to remove them from their matrix or accu- 
rately to determine their form. Other fossils were found, but in 
a very imperfect state. The sections showed that the strata were 
perfectly conformable to those already passed, the dip being about 
12° to the south-west, and the north-west sections horizontal. These 
deep and precipitous ravines are doubtless the result, on an ex- 
tended scale, of the action of water, and satisfactorily explain the 
muddy character of the Platte and the Missouri, into which wash- 
ings from these bluffs have been carried for ages. 

Tuesday, June 26.—In the morning we crossed the dry bed of a 
small stream having its banks well covered with trees, the first we 
had met with since entering the valley of the Platte, now a dis- 
tance of one hundred miles. In the afternoon we overtook the 
pack-train of the Boston company, which had left Fort Kearny 
the day before we did. They had seen about a hundred buffaloes 
crossing the river, and having succeeding in killing one, were no 
little elated at their good luck. We had not as yet been so fortu- 
nate as to discover a single one, a circumstance that proved a 
source of great annoyance to our hunters, in whose mind the asso- 
ciation of “the plains’’ with buffalo-meat was fixed and inseparable, 
and who, consequently, by no means relished their almost exclu- 
sive confinement to salt pork. They were now on the gui vive, 
anxiously anticipating the feasts to which they had constantly 
looked forward. No buffalo however, were seen to-day, the herds 
having been frightened from the road. 

Eneamped six miles above the point of junction of the two forks 
of the Platte, on the bank of a small stream of running water 


with a sandy bottom, the first that had blessed our vision since 
3 


34 FROM FOR® KEARNY TO FORT LARAMIE. 


leaving the Blue. This little creek, running parallel with the 
South Fork, winds its very sinuous way, without bank or shrub or 
bush to mark its course, until it discharges its waters into the 
river several miles below. The bluffs on our left continue to pre- 
sent the same wild forms, being also clothed in many places with 
trees, among which the white-cedar appeared to predominate. 
Owing to the sandy nature of the soil, no fossils were found in a 
perfect state, except two varieties of shells: some imperfect re- 
mains of teeth were also seen, but in too frail a condition to be pre- 
served. About six miles below the Forks, the bluffs presented a 
rougher appearance than those passed early in the day. The prin- 
cipal ravines did not appear to extend very far back, but were of 
considerable width, and intersected by others which came into them 
from every direction. Their sides were very steep, rising in some 
cases to the height of two hundred feet; and so entirely was the 
surface of the ground intersected and cut up, that it was difficult 
to find a spot of even a few yards square that did not enter into 
the formation of some one of them. The prodigious quantity of 
earth that has been removed by the action of water cannot be ima- 
gined without witnessing the scene here presented. The soil com- 
posing the hills, although mixed more or less with clay, is sandy, 
and beetle assumes the character of a very friable sandstone. 
Opposite the Forks, however, the formation of the bluffs again 
alters, and begins to assume a more undulatory and less precipi- 
tous appearance, not so much traversed by ravines. This change 
is occasioned by the cropping out of a stratum of a whitish sand- 
stone. The bluffs passed to-day must, at a period long back, have 
extended much closer to the river than where we now find them, 
having, in fact, been gradually washed into it, leaving the valley 
much broader, and, to a certain extent, one of denadsiaaa 
Carduus, ee with a large pieleiginegicien: yellow flower, Amor- 
pha, Tradescantia, a small sunflower, and a species of milk-plant 
were here found. The Amorpha is beginning to bloom. The vetch, 
with its purple clusters, is met with, but seems of a different species 
from that seen heretofore, and has not so much foliage. 
Wednesday, June 27.—To-day the hunters killed their first buf- 
falo; but, in order to obtain it, had to diverge some four or five 
miles from the road, and to pass back of the bluffs, the instinct or 
experience of these sagacious animals having rendered them shy 
of approaching the line of travel. This has not always been the 
case, for it is a well-attested fact, that when the emigration first 


BUFFALO ON THE PLAINS. 35 


commenced, travelling trains were frequently detained for hours 
by immense herds crossing their track, and in such numbers that 
it was impossible to drive through them. In many instances it was 
quite difficult to prevent their own loose cattle from mingling with 
the buffaloes, of which they did not seem to be at all afraid. The 
eyes of our French voyageurs fairly glistened as they rode into camp 
laden with the meat, and their arrival was hailed with a general 
shout of congratulation. The long-desired spoil was soon divided, 
and a busy scene ensued of roasting, boiling, and making boudin, 
which is a sort of sausage, boiled and eaten hot: when skilfully 
prepared, it forms a most excellent dish. Huge marrow-bones 
might now be seen roasting most temptingly by fires madeof bods de 
vache, and a new spirit seemed to be infused into the entire party 
by this return to their favourite diet. Although, in such a com- 
pany, it would have been rank treason to utter the opinion, yet I 
could not help thinking at the time, that the fat of this meat, which 
our men were devouring with so much gusto, had a somewhat rank 
and disagreeable flavour; and I must, in truth, confess that I was 
not a little disappointed by a dish of which I had received such 
glowing accounts. I found afterward, indeed, that such was the 
opinion, not only of us green ones, but even of our mountain con- 
noisseurs themselves, although, at the time, they did all they could 
to persuade us that it was most excellent; for the animal killed 
was a ‘bull,’’ whose flesh is eaten only when no other can be ob- 
tained, whence the males are very seldom killed when the “ cows” 
can be procured. 

The rock, where it cropped out in the bluffs, was composed of 
white or gray sandstone, similar to that previously passed ; in some 
places strata of an ochreous marl were met with, and in one spot, 
twelve miles above the Forks, astratum of gypsum. The fossils 
were in too imperfect a state to be identified. The small sunflower 
was seen in great abundance, and also buffalo-grass. Innumerable 
buffalo-trails were crossed, leading from the river through the ra- 
vines between the bluffs, to the country beyond, some of which 
were well beaten, and pressed nearly a foot deep into the soil. 

- Thursday, June 28.—Morning bright and pleasant. Ther. at 
sunrise, 59°; Bar. 27.18. The day proved fine, notwithstanding 
the threatening appearance of rain last night, with a cool and 
refreshing wind from the north. The Boston train, which had 
kept ahead of us, was seen crossing the South Fork—an operation 
which they effected apparently without difficulty ; but I preferred 


36 FROM FORT KEARNY TO FORT LARAMIRE. 


to follow still further the main road, which soon led us up one of 
those points of heavy sand-bluffs which here put down directly to 
the river, being cut up in the most fantastic manner by precipitous, 
wide, and abrupt ravines of white sand. Keeping back of the 
heads of these, we again turned down to the river, and halted to 
noon on the bank of the Platte. To-day, buffalo were seen from 
the road, for the first time, feeding in large detached herds, seat- 
tered over the prairies like huge droves of cattle. The sight, to 
those of us who had never beheld these animals, was exceedingly 
interesting, and reminded me of the herds I had seen spread for 
miles over the lovely and fertile valley of Mad River, in Ohio. 
There must have been several thousands within our view, grazing in 
peaceful security. The effect upon our hunters, and, in fact, upon 
the whole party, was that of sudden and most intense excitement, 
and a yearning, feverish desire to secure as much as possible of 
this noble game. Archambault, the guide, had started early in the 
morning for the bluffs to hunt, taking with him another man, both 
mounted. About two o’clock they returned, loaded down with the 
choice pieces of three fat cows, which they had killed. After we 
had advanced a mile on the afternoon’s march, a large band of 
these animals was discovered directly ahead and near the road. 
The opportunity was too tempting to be resisted. There was no 
prospect of getting buffalo above, as they were evidently travelling 
south; and I determined therefore, to remain where we were during 
this and the following day, and to send out and secure as much 
meat as possible before crossing the South Fork. <A halt was ac- 
cordingly made at once, the hunters sent forward, and the afternoon 
occupied in drying the meat brought in during the morning. Our 
men, however, failed in approaching the great herd before they 
took the alarm and vanished. The result of to-day’s work, never- 
theless, was five killed, of which four were secured and brought 
into camp. On the succeeding day three more were killed. The 
flesh thus obtained was forthwith cut into thin strips, dried over a 
fire on a willow frame, and packed into sacks for future use. 
Saturday, June 30.—Ther. at sunrise, 61°; Bar. 27.03; with 
heavy fog. The road, all the morning, continued at the foot of 
the gently sloping bluff of the Platte, with a wide level bottom, 
the uniformity of its surface unbroken by a single bush. A large 
island was passed, of about six miles in length, by two and a-half 
in width, level as a floor, with here and there a clump or grove of 
timber on either margin. A small herd of buffalo was descried in 


PRAIRIE-DOG VILLAGE. 3T 


the river, wading about among some small islands in search of 
grass. 

We passed to-day through a large village or settlement of the 
prairie-dog, (Arctomys ludovictana,) extending in length not less 
than half a mile. These little animals are very shy, and, at the 
_ least approach of a stranger, hie themselves with all speed to their 
holes, in which they partly bury their bodies, leaving only their 
heads visible just above the surface of the ground, where, so long 
as the alarm lasts, they keep up a continual barking. The note 
somewhat resembles the bark of a small puppy, but is nevertheless 
so peculiar as to be instantly recognised ever afterward, by any 
one who has once distinctly heard it. They are very hard to get, 
as they are never found far from their holes; and when shot, fall 
immediately into them, where they are generally guarded by a 
rattlesnake—the usual sharer of their subterranean retreat. Several 
were shot by us in this situation, but when the hand was about to 
be thrust into the hole to draw them out, the ominous rattle of 
this dreaded reptile would be instantly heard, warning the intruder 
of the danger he was about to incur. A little, white, burrowing 
owl also (Stryx cunicularia) is frequently found taking up his 
abode in the same domicile; and this strange association of rep- 
tile, bird, and beast seem to live together in perfect harmony 
and peace. I have never personally seen the owl thus housed, but 

have been assured of the fact from so many, so various, and so 
credible sources, that I cannot doubt it. The whirr of the rattle- 
snake I have heard frequently when the attempt was made to in- 
vade these holes, and our men at length became afraid to approach 
them for this purpose. | 

The march, to-day, was prolonged to an unusually late hour, as 
T was in hopes of reaching the ford of the South Fork before night, but 
finding this impracticable, we encamped on the bank of the river 
where the prairie was hard and level, with plenty of excellent grass 
and water. With the exception of a few small willows growing on 
the opposite side, and a large cotton-wood or two a few miles up 
the river, serving to mark the point for crossing, no trees were to 
be seen. We were fortunate enough, however, to find three or 
four old lodge-poles, left by a passing band of Sioux, which, eked 
out by bots de vache for fuel, served to give us a capital roast of 
buffalo-meat, which, wearied as we were by a march of fourteen 
hours in the burning sun, proved a most welcome and acceptable 
refreshment. 


38 , FROM FORT KEARNY TO FORT LARAMIE. 


The formation over which we have passed the last two days has 
been composed of beds of sand and sandy shales. No rock has 
been met with; and wherever a section of the soil was obtained, 
it presented layers of sand containing small proportions of argilla- 
ceous matter. The general profile of the hills is rounded and un- 
dulatory. Rock, apparently in regular strata, is to be seen in the 
opposite bluffs, which are bolder and approach the river more 
nearly than on the south side. 

Sunday, July 1.—Ther. at 9 o’clock, 81°; Bar. 26.74. The day 
being very warm and fine, advantage was taken of it to dry more 
thoroughly the meat we wished to preserve. Our breakfast this 
morning, which was spread out on the ground, with a piece of In- 
dia-rubber cloth for a table, under the shade of one of the wagon- 
covers, consisted, for five persons, of two buffalo-tongues and the 
hump of a fat cow, nearly the whole of which was consumed. The 
flesh of a fat buffalo-cow is perhaps the best beef that can be eaten, 
wholly free from the rank flavour which marks the fat of the male: 
it is at once juicy, tender, nutritious, and very digestible, added to 
which it has a game flavour which renders it far superior to the 
very best beef of the States. It may, in fact, be not improperly 
denominated «game beef.” 

This was the first time that any of my mess had partaken of that 
famous dish, the « hump,” and the quantity disposed of was the best 
proof of the intense relish with which it was enjoyed. ‘This and 
the tongue, tender-loin, bass, and marrow-bones are considered the 
choice parts of the carcass, and, where the animals are plenty, no 
other parts are taken, the residue being left on the ground for the 
wolves. Some idea may be formed of the great digestibility of 
this species of food, as well as of the enormous quantities de- 
voured at a single meal, from the fact that the regular daily allow- 
ance or ration for one employee in the Fur Company’s service is 
eight pounds, the whole of which is often consumed. It is true, 
however, that an old mountaineer seldom eats any thing else. If 
he can get a cup of strong coffee, with plenty of sugar, and as 
much buffalo-meat as he can devour, he is perfectly happy and con- 
tent, never feeling the want either of bread or vegetables. 

A partial examination was made to-day of the north bank of 
the river. The rocks which had attracted attention yesterday 
were distant about four hundred yards from the stream and very 
much weathered, presenting a broken surface, owing to some of 
the strata having disintegrated faster than others. The perpen- 


CROSSING OF THE SOUTH FORK OF THE PLATTE. | 39 


dicular section of the rocks was seen in place. The bluff was in- 
tersected by ravines, many of which also exposed vertical sections. 
The strata were nearly horizontal, with perhaps a slight slip to the 
west, and very fossiliferous, some of them being composed of en- 
crinital remains. The hills were about one hundred and fifty feet 
in height, and consisted of the following strata:—On the surface, a 
yellow shale, containing encrinites and grit; layers of loose detri- 
tus and quartzose conglomerate; brown shale and sand, the former 
containing wood and bones; white sandstone, the exposed surface 
marked by ripples; calcareous sandstone; sandy limestone, com- 
posed principally of remains of stems of encrinites, none of tke 
central portions of which were found, (this layer was some feet 
thick, and was traced fully a mile;) whitish sandstone; botryoidal 
limestone. The fossils found were the remains of encrinites, 
which were abundantly diffused through the different strata, with 
the exception of the detritus. Some bones occurred, but in a very 
imperfect state: a large flat bone was found imbedded in sundy 
shale, and endeavours were made to get it out, which did not suc- 
ceed. The part exposed presented a segment of a circle from 
twelve to sixteen inches in diameter. About half a mile below 
this point was a peak formed by large masses of encrinital lime- 
stone; and at a point still lower down, the same formation was 
Sand. the whole being crowned by a white marl, containing encri- 
nites ome grit. At this pot were discovered the remains of the 
bones of some large animal, only imperfect specimens of which 
could be procured, for want of the means of extracting them. 
One fragment was seen, fully seven inches in circumference; and a 
tooth, exposed for a length of five inches, was broken in the 
attempt to get it out. The condyle of the jaw and what was sup- 
posed to be the foot of some reptile were secured. 

Monday, July 2.—Ther. at sunrise, 68° ; Bar. 26.63. After tra- 
velling up the river for fourteen miles, it was determined to make 
the crossing of the South Fork by fording. In preparation for 
this movement, one of the Wagons, as an experimental pioneer, was 
partially llaaveds by removing all articles liable to injury from 
water, and then deren into the stream; but it stuck fast, and the 
ordinary team of six mules being fina insufficient to haul it 
through the water, four more were quickly attached, and the cross- 
ing was made with perfect safety, and without wetting any thing. 
In the same manner were all the remaining wagons crossed, one 
by one, by doubling the teams, and employing the force of nearly 


40 FROM FORT KEARNY TO FORT LARAMIE. 


the whole party wading alongside to incite and guide the mules, 
lest, from some sudden eccentricity, to which those animals are so 
constantly prone, a wagon might be capsized or precipitated into a 
hole. The water was perfectly opake with thick yellow mud, and 
it required all our care to avoid the quicksands with which the bot- 
tom is covered. The labour was excessive, on both men and ani- 
mals, as the river was nearly half a mile wide, and the current from 
recent rains ran with great rapidity and force. Wading such a 
stream_breast-deep four or five times, with such treacherous foot- 
ing, was very exhausting, and we were glad to encamp, immedi- 
ately after crossing, upon the left bank. Both man and beast suf- 
fered more from this day’s exertion than from any day’s march we 
had yet made. About one and a-half miles above the crossing a 
new Indian lodge was seen standing entirely alone. A fact so 
unusual excited our curiosity: upon going to the place, it was found 
to contain the body of an Indian (probably a chief) raised upon a 
low platform or bier, surrounded by all the implements believed 
' by these simple children of the forest to be necessary for his use 
in the spirit-land. The lodge was carefully and securely fastened 
down at the bottom, to protect its charge from the wolves. It was 
an affecting spectacle. His last battle fought, his last hunt over, 
here he lay in the solitude of death, abandoned by wife and child, 
and all he loved, yet surrounded by the tokens of thew parting 
care, the rude proofs of a love that followed him to an unknown 
hereafter. We are now, by our measurements, four hundred and 
seventy-nine miles from Fort Leavenworth, and one hundred and 
eighty from Fort Kearny. | 

Tuesday, July 8.—Morning cool and delightful; Ther. at sun- 
rise, 71°; Bar. 26.59; Wind S. W., fresh and bracing. To-day 
we crossed the ridge between the North and South Forks of the 
Platte, a distance of eighteen and a-half miles. As we expected to 
find no water for the whole of this distance, the India-rubber bags 
were filled with a small supply. The road struck directly up the bluff, 
rising: quite rapidly at first, then very gradually for twelve miles, 
when we reached the summit, and a most magnificent view saluted — 
the eye. Before and below us was the North Fork of the Nebraska, 
winding its way through broken hills and green meadows; behind 
us the undulating prairie rising gently from the South Fork, over 
which we had just passed; on our right, the gradual convergence 
of the two valleys was distinctly perceptible; while immediately at 
our feet were the heads of Ash Creek, which fell off suddenly into 


ASH HOLLOW. Dia Al 


deep: precipitous chasms on either side, leaving only a high nar- 
row ridge, or backbone, which gradually descended, until, toward 
its western termination, it fell off precipitately into the bottom of 
the creek. Here we were obliged, from the steepness of the road, 
to let the wagons down by ropes, but the labour of a dozen men 
for a few days would make the descent easy and safe. The bot- 
tom of Ash Creek is tolerably well wooded, principally with ash 
and some dwarf cedars. The bed of the stream was entirely dry, 
but toward the mouth several springs of delightfully cold and 
refreshing water were found, altogether the best that has been 
met with since leaving the Missouri. We encamped at the mouth 
of the valley, here called Ash Hollow. The traces of the great 
tide of emigration that had preceded us were plainly visible in 
remains of camp-fires, in blazed trees covered with innumerable 
names carved and written on them; but, more than all, in the total 
absence of all herbage. It was only by driving our animals toa 
ravine some distance from the camp, that a sufficiency for their sub- 
sistence could be obtained. 

The two slopes of the ridge dividing the main forks of the 
Platte, at the point where we crossed it, differ from each other in 
aremarkable manner. On that toward the South Fork, the val- 
leys are wide and long, with gracefully curved lines, gentle slopes, 
and broad hollows. In numerous instances, these hollows are 
without drainage, owing to which large circular or oval basins are 
formed, in the bottoms of which water collects, forming quite ex- 
tensive ponds or lakes: these, however, disappear during the sum- 
mer, leaving their beds clothed with a rich, luxuriant growth of 
herbage. On the opposite side of the’summit the features of the 
country present a striking contrast. Almost immediately after 
crossing the point of “divide,” we strike upon the head waters of 
Ash Creek, whence the descent is abrupt and precipitous. Imme- 
diately at your feet is the principal ravine, with sides four or five 
hundred feet in depth, clothed with cedar: into this numerous 
other ravines run, meeting it at different angles, and so completely 
cutting up the earth, that scarcely a foot of level ground could be 
seen. The whole surface consisted of merely narrow ridges, divid- 
ing the ravines from each other, and running up to so sharp a 
crest that 1t would be difficult for any thing but a mountain-goat 
to traverse their summits with impunity. «Never before had I seen 
the wonderful effects of the action of water on a grand scale 
more strikingly exemplified. 


42 FROM FORT KEARNY TO FORT LARAMIE. 


The soil on the top of the ravine seemed to consist of decom- 


posed sand-rock, among which, however, were found some pieces - 


of silicious limestone, with traces of encrinites. In one of the 
lateral ravines, a complete section of the rock was obtained. It 
was composed of layers of white sandstone, calcareous sandstone, 
sand, sand and clay, and granitic and quartz detritus, cemented 
by calcareous rocks. The only fossils found were the remains of 
encrinites, which existed in almost all the strata; and in some the 
quartzose portion of the rock was composed of their stems. 
Toward the lower part of the gorge was a bed or layer of marl, 
in which were the remains of what very much resembled the seeds 
of a plant. All the strata were horizontal. On examining them 
at the mouth of the hollow, on the following day, a stratum was 


found of botryoidal limestone, consisting of oblong pieces of mag-— 


nesian limestone, cemented by a calcareous rock, and also another 
bed of sandy marl, containing remains of the seeds already men- 
tioned, together with some oblong bodies, which were probably 
those of encrinites. | 

Heavy thunder, wind, and rain, during the night. 

Wednesday, July 4.—At 9 a.M., Bar. 26.76; Ther. 68°. This 
being a national festival, I determined to spend the day here and 
celebrate it as well as our limited means would permit. A salute 
was fired morning and evening, and a moderate allowance of grog 
served out to the men, which, with a whole day’s rest and plenty 
of buffalo-meat, rendered them quite happy. We had observed 
yesterday, on the opposite side of the river, a number of Indian 
lodges, pitched on the bank; but the total absence of any living or 
moving thing about them induced us from curiosity to pay them a 
visit. In order to do this it was necessary to cross the river, here 
nearly a mile in breadth, with astrong, rapid current. I was afraid 
to risk any of the animals, as the bottom was known to be very 
treacherous and full of quicksands; so it was determined we 
should wade it. Having stripped to our drawers, we tied our 
shirts and moccasins around our necks to keep them dry, and, ac- 
companied by five or six of the men, commenced the passage. The 
water was up to our middle, and the strong and constant pressure 
of the current rendered our efforts to bear up against it very 
fatiguing. We struggled on, but very slowly, from the yielding 
nature of the sandy and marly bottom, which was immediately 
washed from beneath the foot every time it was placed on the 
ground. If we stood still in the same spot, even for a short time, 


‘SIOUX DEAD LODGES. 43 


the bottom would be so rapidly excavated from beneath us, that a 
hole of sufficient depth would be formed to render swimming ne- 
cessary. After continuing these tedious and laborious efforts 
until we had nearly reached the opposite shore, on advancing a 
single step we found ourselves in water beyond our depth, (the 
channel of the river running close to the bank,) and the shirts we 
had so carefully endeavoured to keep dry were in a moment 
thoroughly soaked. We made out, however, to scramble ashore. 
I put on my moccasins, and, displaying my wet shirt, like a flag, 
to the wind, we proceeded to the lodges which had attracted our 
curiosity. There were five of them, pitched upon the open prairie, 
and in them we found the bodies of nine Sioux, laid out upon the 
ground, wrapped in their robes of buffalo-skin, with their saddles, 
spears, camp-kettles, and all their accoutrements, piled up around 
them. Some lodges contained three, others only one body, all of 
which were more or less in a state of decomposition. A short dis- 
tance apart from these was one lodge which, though small, seemed 
of rather superior pretensions, and was evidently pitched with 
great care. It contained the body of a young Indian girl of six- 
teen or eighteen years, with a countenance presenting quite an 
agreeable expression: ‘she was richly dressed in leggings of fine 
scarlet cloth, elaborately ornamented; a new pair of moccasins, 
beautifully embroidered with porcupine quills, was on her feet, and 
her body was wrapped in two superb buffalo-robes, worked in like 
manner. She had evidently been dead but a day or two; and to 
our surprise a portion of the upper part of her person was bare, 
exposing the face and a part of the breast, as if the robes in which 
she was wrapped had by some means been disarranged, whereas all 
the other bodies were closely covered up. It was, at the time, the 
opinion of our mountaineers that these Indians must have fallen in 
an encounter with a party of Crows; but I subsequently learned 
that they had all died of the cholera, and that this young girl, 
being considered past recovery, had been arrayed by her friends 
in the habiliments of the dead, enclosed in the lodge alive, and 
abandoned to her fate—so fearfully alarmed were the Indians by 
this, to them, novel and terrible disease. But the melancholy tale 
of this poor forsaken girl, does not end here. Her abandonment 
_ by her people, though with inevitable death before her eyes, may 
perhaps be excused from the extremity of*their terror; but what 
will be thought of the conduct of men enlightened by Christianity, 
and under no such excess of fear, who, by their own confession, ap- 


44 FROM FORT KEARNY TO FORT LARAMIE. 


proached and looked into this lodge while the forsaken being was’ 
yet alive, and able partially to raise herself up and look at them, 
but who, with a heartlessness that disgraces human nature, turned 
away, and, without an effort for her relief, left her alone to die! 
Which company deserved the epithet of savages, the terrified and 
flying red men, or the strong-hearted whites who thus consummated 
their cruel deed? 

Leaving this melancholy scene, we recrossed the river and re- 
turned to our encampment, where preparations had been made for 
a Fourth of July dinner. Although deprived of the vegetable luxu- 
ries upon which our Hastern friends were doubtless feasting, still our 
bill of fare would not have been unacceptable even to an epicure. 
Buffalo-soup, buffalo-ribs, tender-loin, and marrow-bones roasted, 
boiled ham, stewed peaches, and broiled curlew, relished with a couple 
of bottles of cool claret, (which had beeen carefully preserved for 
the occasion,) and crowned by a cup of coffee and a segar, made a 
meal which, notwithstanding the cup was of tin and our table the 
greensward, we thought not entirely unworthy of the day. In the 
evening two men came into camp and requested our hospitality : 
they had been emigrants, but were on their return to the States 
disgusted, having fallen out with their company by the way. 

Thursday, July 5th.—Bar. 26.67. Ther. 56°. We commenced 
our journey to-day up the North Fork of the Platte. The road 
winds along the bottom under the bluffs. The lower stratum con- 
sists of yellow clay, capped by cliffs of sandstone and silicious 
limestone, about two hundred feet in height. This formation was 
traced uninterruptedly for about twenty miles. The limestone 
appeared to contain no fossils—at least, none were discovered. To- 
ward the end of the day’s march the clay was left uncovered by 
the limestone, presenting bald eminences destitute of the least 
vegetation, which, from the action of the weather, had been worn 
into various curious and isolated peaks, of forms extremely pic- 
turesque. Encamped on the bank of the river, after a tedious 
march of twenty-three miles. Just above us, was a village of 
Sioux, consisting of ten lodges. They were accompanied by Mr. 
Badeau, a trader ; and, having been driven from the South Fork by 
the cholera, had fled to the emigrant-road, in the hope of obtaining 
medical aid from the whites. As soon as it was dark, the chief 
and a dozen of the braves of the village came and sat down in a 
semicircle around the front of my tent, and, by means of an in- 
terpreter,, informed me that they would be very glad of a little 


SIOUX VILLAGES—CHOLERA. . 45 


coffee, sugar, or biscuit. I gave them what we could spare. They 
told us there was another and larger band encamped about two 
miles above, many of whom were very sick with the cholera: they 
themselves had been afflicted with it, but had in a great measure 
recovered, although they were in great dread of its return. As 
soon as they were told I had a doctor, or ««medicine-man,’’ with me, 
and received assurances that some medicines should be prepared 
for them, and left with the trader, (who had married among them,) 
they expressed much delight, and returned to their village, where, 
soon after, the sound of the drum and the song, expressive of the 
revival of hope, which had almost departed, resounded from the 
«‘medicine lodge,’ and continued until a late hour of the night. 
In the mean time, I directed a quantity of medicine to be prepared, 
with the necessary directions for using it. The following morning 
we paid a visit in passing, to the upper village, which contained 
about two hundred and fifty souls. They were in the act of break- 
ing up their encampment, being obliged to move farther up the 
river to obtain fresh grass for their animals. A more curious, 
animated, and novel scene I never witnessed. Squaws, papooses, 
dogs, puppies, mules, and ponies, all in busy motion, while the 
lordly, lazy men lounged about with an air of listless indifference, 
too proud to render the slightest aid to their faithful drudges. 
Before the lodge of each brave was erected a tripod of thin slender 
poles about ten feet in length, upon which was suspended his 
round white shield, with some device painted upon it, his spear, 
and a buckskin sack containing his “medicine” bag. It reminded 
me forcibly of the scenes of Ivanhoe and the Crusaders, and im- 
pressed me with the singularity of the coincidence in the customs 
of what were then the most refined nations of the world, with those 
of these wild and untutored savages. The cholera had been quite 
bad among them, and was still raging. I visited nearly every 
lodge, in company with the doctor and Mr. Bissonette the trader, 
and medicine was administered to all who required it. It was 
touching to witness the moral effect produced by the mere presence 
of a “‘medicine-man”’ upon these poor wretches. They swallowed 
the medicine with great avidity, and an absolute faith in its efficacy, 
which, I have little doubt, saved many a life that would otherwise 
have been lost. I shall never forget one poor fellow, a tall, fine- 
looking young man of about twenty-five. He had been sick three 
days, and we found him sitting on the ground, his blanket drawn 
closely around him, and his chin resting upon his knees, the 


46 FROM FORT KEARNY TO FORT LARAMIE. 


image of despair,—very quiet, but the expression of his countenance 
showing that he had made up his mind that he must die. To add 
to his despondence, a young man from the next lodge had just been 
carried out and buried. The doctor examined him closely, and 
then requested the interpreter to tell him that the worst was past, 
and that, with care and attention, he would soon entirely recover. 
Never did I behold any thing like the change which, in an instant, 
came over the expressive countenance of this poor savage. His 
face flushed, the fire came into his eyes, and a radiant smile of 
confidence and hope, which was beautiful to behold, broke through 
the previous gloom. He raised his eyes, till now sternly fixed 
upon the ground, gently smote his hands together, turned his head 
toward his squaw, who was standing behind him, and in a low and 
silvery tone communicated to her the joyful news. It was to him 
a perfect resurrection from the dead; for he seemed now to enter- 
tain no doubt of his recovery, but received the assurance of the 
doctor as if it had been the fiat of fate. It was a moving sight, 
and although we could not understand a single word that passed, 
the whole scene was perfectly intelligible. After administering to 
all who stood in need, a quantity of medicine was left with Mr. 
Bissonette, with the necessary directions. 

The whole village was much revived by this visit from the 
“cmedicine-man,’’ and expressed much thankfulness, and a strong 
desire that he should remain among them. As this was obviously 
impossible, we continued our journey, accompanied for several 
miles by the people of both villages. The whole scene was unique 
in the highest degree. ‘The road was strewn for miles with the 
most motley assemblage I ever beheld, each lodge moving off from 
the village as soon as its inhabitants were ready, without waiting 
for the others. ‘The means of transportation were horses, mules, 
and dogs. Four or five lodge-poles are fastened on each side of 
the animal, the ends of which trail on the ground behind, like the 
shafts of a truck or dray. On these, behind the horse, is fastened a 
light framework, the outside of which consists of a strong hoop bent 
into an oval form, and interlaced with a sort of network of rawhide. 
Most of these are surmounted by a light wicker canopy, very like 
our covers for children’s wagons, except that it extends the whole 
length, and is open only at one side. Over the canopy is spread 
a blanket, shawl, or buffalo-robe, so as to form a protection from 
the sun or rain. Upon this light but strong trellice-work, they 
place the lighter articles, such as clothing, robes, &c., and then 


SIOUX MOVING. Av 


pack away among these their puppies and papooses, (of both which 
they seem to have a goodly number ;) the women, when tired of 
walking, get upon them to rest and take care of their babies. The 
elasticity of the poles makes the motion easy and pleasant. I 
afterward saw an old Indian, that had been crippled in one of the 
skirmishes which so often occur among these tribes, whose only 
mode of locomotion was a contrivance of this kind, from which 
he could not move without assistance. 

The dogs also are made to perform an important part in this 
shifting of quarters. Two short, light lodge-poles are fastened 
together at the small end, and made to rest at the angle upon the 
animal’s neck, the other end of course, trailing upon the ground. 
Over his shoulders is placed a sort of pad, or small saddle, the 
girth of which fastens the poles to his sides, and connects saith a 
little collar or breast-strap. ‘Behind the rie a small platform or 
frame is fastened to the poles, similar to that used for the horses, 
upon which are placed lighter articles, generally puppies, re 
are considered quite valuable, being raised for beasts of burden as 
well as for food and the chase. Iwas informed by Mr. Bissonette, 
that many tribes, having no horses or mules, use dogs altogether 
in moving their villages. We saw a large number of these animals, 
with their sharp noses and wolflike ears, thus harnessed, and 
trotting along as if it were an occupation to which they were per- 
fectly accustomed. The whole duty of taking down and putting 
up the lodges, packing up, loading the horses, arranging the lodge- 
poles, and leading or driving the animals, devolves upon the squaws, 
while the men stalk along at their leisure; even the boys of larger 
growth deeming it beneath their dignity to lighten the toils of their 
own mothers. 

The whole band halted about ten o’clock on the bank of the 
river, but several of the old men and the chief of the village con- 
tinued with us until our noon halt. I invited the latter to lunch 
with us, which he did to his entire satisfaction, devouring as much 
meat as the whole mess beside, and I afterward espied him seated 
at one of the messes of the men, as earnestly engaged in laying in 
an additional supply as if he had not eaten for a week. The In- 
dian, in fact, from his wandering habits and uncertain mode of 
existence, acquires the faculty of laying in, when opportunity offers 
itself, a store of food against the fast that may follow, thus approxi- 
mating the instincts of other wild denizens of the forest. 

After crossing a few running streams, we encamped near a num- 


48 FROM FORT KEARNY TO FORT LARAMIE. 


ber of springs of soft, cold, sandstone water, which proved very 
grateful after the hot ind dusty Journey of the day. — 

The banks on this side of the river have presented little of inte- 
rest, the surface generally consisting of rolling prairie, gradually 
rising to the summit of the hills, which are at a considerable dis- 
tance from the stream. On the opposite side, the cliffs are pre- 
cipitous, displaying sections of horizontal beds of apparently the 
same rock already noticed. About two miles from our noon halt 
the rock was exposed close to the river and but little above its 
level. It consisted of layers of sandstone, with detritus, magne- 
sian limestone, sand, and clay. Some fossils were collected and 
remains of encrinites were observed. On entering the bottom of 
the North Fork, we found the white Znothera, alarge flower, grow- 
ing with but a single flower-stem. Lupines are still found in con- 
siderable numbers, together with phlox and some species of sun- 
flower. The heads of the ravines were clothed with pine and 
cedar. The growth of the latter tree appears to be diminishing ; 
for while numerous dead trees lay strewn along the bottoms, but 
few living ones were found growing on the hills. 

Saturday, July T7.—Ther. at sunrise, 59°; Bar. 26.55. This 
morning we caught a view of the celebrated «Chimney Rock,” and 
also of the «‘ Court-house,’’ which latter consisted of two bald eleva- 
tions, similar in formation to that already passed, to which the 
voyageurs, most of whom are originally from St. Louis, had given 
this name, from a fancied resemblance to a well-known structure in 
their own city. 

In riding out from the road to visit this curious formation, we 
found the main bluff of the river to be about five miles distant, the 
intervening country consisting of rolling hillocks covered with 
grass. In our ride we crossed the dry sandy bed of a stream, 
about two hundred and fifty feet in width, which, in the rainy 
season, must discharge a large quantity of water. It had little or 
no bank, and, from the appearance of drift-wood far out on the 
prairie, must overspread a large surface in the spring. A mile 
and a-half from this creek we came upon another, called on the 
maps « Dry Creek,”’ but known among the mountain-men as « Law- 
rence’s Fork,” from the fact that a man of that name had been 
killed on it by the Sioux. The Court-house was but a few 
hundred yards beyond this stream, which was about thirty feet 
wide and two or three feet deep, flowing with a free, bold, and 
tolerably rapid current: it had cut its bed through the blue clay, 


COURT-HOUSE—LIGNITE. 49 


with a few narrow layers of sand. I attempted to cross it, but the 
bottom consisted of a stiff marly mud, into which the feet of the 
animals sank rapidly, and could with difficulty be withdrawn. 
Fearful of miring them down, I gave up the attempt, and thus lost 
the opportunity of examining this celebrated seat of justice more 
closely. Upon the bank, where we attempted to cross, I found a 
large block of what proved to be lignite, and near by a lump of 
-what was thought to be oolitic clay, from which several species of 
fresh-water shells were taken. The lignite had evidently been 
brought down by the creek, as many large masses of it were seen 
lying in the water and strewn along the banks. Should this sub- 
stance be found in any quantity in the hills to the south, it may 
have a very important bearing upon the future settlement of this 
region; one of the great obstacles to which, even when water 
can be commanded, is the total absence of timber sufficient for 
fuel. Could lignite be found in sufficient quantity for this purpose, 
many spots in this extensive valley might be settled to advantage, 
and would thus furnish stations where the emigrant and mail trains 
might find shelter, protection, and provisions. I should have di- 
rected a more extended examination, but the objects contemplated 
by the expedition had been already too much delayed to justify it. 

After passing the Court-house, a large butte, to the right of 
the Chimney Rock, was before us the whole afternoon, and pre- 
sented so remarkable a resemblance to the capitol at Washington, 
with its dome in the centre and its wings on either side, that it 
was the subject of remark with all of our company who had ever 
seen that building. The resemblance was very strong, and I could 
not but wonder that it had not elicited remark from some previous 
traveller; since, compared with the far-famed Chimney Rock, it 
is a much larger and more imposing object. 

After a warm drive of twenty-five miles, we encamped within 
five miles of the Chimney Rock, upon a point, or rather knoll 
of land, overlooking the prairie toward the Platte, here about a 
mile distant. Directly at the foot of the bluff is a most lovely 
spring, which comes bubbling up from the ground in a clear, pel- 
lucid stream, affording abundance of the coldest and most delicious 
water we had yet found. (Temperature, 524°.) Here we spent the 
following day, which was the Sabbath. No wood was to be ob- 
tained nearer than the bluffs, and I was obliged to despatch a 
wagon thither to procure sufficient for cooking purposes. The 


hunters went out early, and returned about ten o’clock, with as 
: 4 


50 FROM FORT KEARNY TO FORT LARAMIE. 


much meat as two pack-mules could carry—their riding-mules also 
being loaded with the same welcome freight. They had killed three 
elk and an antelope. ‘Ther. in the sun, with fresh breeze, at 124 
o'clock, 984°. | 

Monday, July 9.—Ther. at sunrise, 55°; Bar. 26.26; Wind N. 
W. I determined this morning to examine more particularly the 
curious bluffs or range which extends from the Court-house to 
Chimney Rock, and on beyond nearly to Scott’s Bluffs. Riding 
south from the road for a distance of about five miles over an open 
prairie, cut up in every direction by hollows and little short ridges, 
we arrived at a pass or gorge through what appeared to be the 
main bluff, or southern boundary of the Platte valley. The cliffs 
on either side of the pass were about thirty feet high, and presented 
a section of clay, sandy clay, and calcareous sandstone. In some 
places, projecting from the side of the cliff, were rounded layers of 
the latter rock, disposed in a vertical direction, presenting the ap- 
pearance of the vertebre of some large animal. On reaching the 
summit of the pass, I found that this range, instead of being the 
main bluff bounding the Platte valley, was only a high ridge sepa- 
rating it from that of Lawrence’s Fork. The latter stream here 
runs about north-east, through a broad, level prairie, four or five 
miles wide, bounded by a high bluff on its southern side, and dis- 
charges itself into the Platte many miles below; the stream forks 
just above, and a high, broad ridge, similar in its character to that 
we had just crossed, divided the two branches, the valleys of which 
seemed to extend a considerable distance to the west. A few trees 
were seen on the farther one. In the pass, two handsome varie- 
ties of Digitalis occurred, both of a blue colour, one with glabrous 
leaves, and flowers of a bright blue, the other with pubescent leaves, 
and flowers not so bright; a dwarf white chrysanthemum was also 
found in the same locality. I hoped to find some more specimens of 
lignite on this stream, which is the same that washes the base of the 
Court-house lower down, where that substance was seen yesterday. 
T accordingly rode out to it. It was here a beautiful bubbling brook, 
flowing with a rapid current over sand and rolled stones, brought 
down from the Court-house ridge in immense quantities. No 
lignite was discovered at this spot, although I have little doubt that 
it exists higher up, near the sources of this little stream. 

The south side of the ridge, which we followed until nearly op- 
posite the Chimney Rock, presents the same fantastic appear- 
ance as does that fronting the Platte valley, being worn by the 


CHIMNEY ROCK. 51 


weather into jutting, round abutments and castellated towers. At 
this point the ridge is two miles wide, very much broken, and the 
side so steep that it was impossible to keep the saddle while ascend- 
ing. Descending the north slope, we were guided through a series 
of narrow and extremely intricate ravines by a well-worn buffalo- 
trail into the plain below. Before us was the Chimney Rock, 
@ point on this route so well known and so often described. In the 
strata of clay, sand, sandstones, and siliceous limestones, over which 
we have been travelling for the last three days, the clay is most 
predominant in this vicinity and to the eastward of Scott’s Bluff. 
The partial disintegration of these strata has in some places 
given to the bluffs the most curious shapes, and among others, 
that of the Chimney Rock. This singular conformation has 
been, undoubtedly, at one time, a portion (probably a projecting 
shoulder) of the main chain of bluffs bounding the valley of the 
Platte, and has been separated from it by the action of water. It 
consists of a conical elevation of about one hundred feet high, its 
sides forming an angle of about 45° with the horizon; from the 
apex rises a nearly circular and perpendicular shaft of clay, now 
from thirty-five to forty feet in height. The cone has, I think, 
been formed by the disintegration of the softer portion of the bluff 
arranging itself at its natural angle in a conical form, while the 
remainder of the earth has been carried away by the floods and 
distributed over the plain, leaving the broad valley which is at pre- 
sent found between it and the main bluff. The Chimney, being 
composed of more tenacious materials, has been left standing in a 
vertical position, and has been worn into its present circular form 
by the gradual action of the elements. That the shaft has been 
very much higher than at present, is evident from the correspond- 
ing formation of the bluff, as well as from the testimony of all our 
voyageurs, with whom it was for years a landmark or beacon visi- 
ble for forty or fifty miles, both up and down the river. It is the 
opinion of Mr. Bridger that it was reduced to its present height by 
lightning, or some other sudden catastrophe, as he found it broken 
on his return from one of his trips to St. Louis, though he had 
passed it uninjured on his way down. Its vicinity has long been 
a favourite encamping ground for the emigrants, as there are 
springs of water near and the grass is tolerably good. In crossmg 
over from the valley of Lawrence’s Fork, it was noticed that the 
ridge had been at one time covered with a tolerably dense growth 
of cedar. These trees have nearly all died, and their trunks are 


52 FROM FORT KEARNY TO FORT LARAMIE. 


strewn over the ground. Young pines, however, are rapidly taking 
their place. What could have caused this singular phenomenon ? 
In former years the valley of the Platte was similarly covered with 
these dead cedars, brought down by the freshets, but now there are 
none to be found, they having all been converted into fuel by the 
emigrants. Three miles from the Chimney Rock, the road 
gradually leaves the river for the purpose of passing behind 
Scott’s Bluff, a point where a spur from the main ridge comes 
so close to the river as to leave no room for the passage of teams. 
There was no water between these two points, a distance of more 
than twenty miles, and we were consequently obliged to go on until 
nine o'clock, when we encamped at the bluff, on a small run near 
a delicious spring, after having been in the saddle sixteen hours 
without food, and travelled thirty-one and a-half miles. The march 
was a severe one upon the animals, as they were in harness, after 
the noon halt, for seven successive hours, without water. The 
afternoon was oppressively hot, and the gnats and musquitoes 
almost insufferable. There is atemporary blacksmith’s shop here, 
established for the benefit of the emigrants, but especially for that 
of the owner, who lives in an Indiar lodge, and had erected a log 
shanty by the roadside, in one end of which was the blacksmith’s 
forge, and in the other a grog-shop and sort of grocery. The stock 
of this establishment consisted principally of such articles as the 
owner had purchased from the emigrants at a great sacrifice and 
sold to others at as great a profit. Among other things, an excellent 
double wagon was pointed out to me, which he had purchased for 
seventy-five cents. The blacksmith’s shop was an equally profit- 
able concern; as, when the smith was indisposed to work himself, 
he rented the use of shop and tools for the modest price of seventy- 
five cents an hour, and it was not until after waiting for several 
hours, that I could get the privilege of shoeing two of the horses, 
even at that price, the forge having been in constant use by the 
emigrants. Scott's Bluff, according to our measurement, is five 
hundred and ninety-six miles from Fort Leavenworth; two hun- 
dred and eighty-five from Fort Kearny, and fifty-one from Fort 
Laramie. 

Thursday, July 12.—Bar. 26.18; Ther. at sunrise, 53°. We 
arrived to-day at Fort Laramie, and encamped a short distance 
above, on Laramie’s Fork, a fine, rapid stream, about fifty yards 
wide. Here we remained until the 18th, recruiting our animals, 
getting them shod, repairing our wagons, and making the necessary 


. 
- 
ae 

i 4 


™ AN Aempeoig 6) ¢ YT URULOMOY 


~~ 


FORT LARAMIE. 58 


arrangements for continuing our journey. I here unpacked one 
of the barometers which I had taken charge of for the Smithsonian 
Institution, to be left at this post. It had stood the journey ad- 
‘mirably, was in perfect order, and was gladly received by Lieu- 
tenant Woodbury, of the corps of Engineers. Observations also 
were made for the latitude of the post, which placed it in lat. 
42°.12’ 38." 2, long. 104° 31’ 26”. 

Fort Laramie, formerly known as Fort John, was one of the posts 
established by the American Fur Company for the protection of 
their trade. Its walls are built in the usual style of such structures, 
of adobe or unburnt brick. The company sold it to the United 
States Government; and their people, when we arrived, were tem- 
porarily encamped near the ford of the creek, having recently sur- 
rendered the possession of the post to the troops, whom we found 
engaged in preparing for its extension and in the erection of addi- 
tional quarters, under the superintendence of Lieutenant Woodbury. 
It is garrisoned at present by two companies of Infantry and one 
of Mounted Rifles, under command of Major Sanderson, of the 
latter corps, by whom we were received with the greatest courtesy, 
and promptly furnished with such supplies as were within the 
resources of his command. I procured here fifteen additional mules, 
and our stock now consisted of fifty-six mules, five horses, four 
steers for beef, and two milch-cows, one of which we had found on 
the prairie, abandoned or lost by her owners. 

The country has risen considerably since leaving Scott’s Bluff, 
and the general flora indicates a much drier atmosphere: the 
grasses especially are brown and burned up wherever the earth is 
not directly moistened by proximity to some stream. The soil 
around Fort Laramie appears to be sterile, owing no doubt to the 
extreme dryness of the air and the almost total absence of dews. 
The great quantity of coarse conglomerate, too, which, by its dis- 
integration, leaves the surface covered with gravel, must operate 
as a great impediment to cultivation. ‘The rocks, however, con- 
tain the elements of fertility, being composed of limestone, clay, 
and sand; and I have no doubt that, with the aid of irrigation, the 
bottom lands of Laramie Creek might be made to produce most 
abundant crops. Hay is cut about eight miles up the stream in 
quantity sufficient for the wants of the garrison. 

A short excursion of some seven miles up the Laramie river, 
showed that the sections of the bluffs presented strata of sand- 
stone conglomerate, formed, in some cases, of the detritus of sand- 


54 FROM FORT LARAMIE TO FORT BRIDGER. 


stone and calcareous rocks, cemented in an argillaceous matrix. 
The general direction of the strata was nearly horizontal, but 
there were evident local displacements, caused apparently by 
subterraneous upheavings. In some cases the strata were declined 
as much as 80°, and in opposite directions, within a short space. 
In many places large quantities occurred of the fragments of 
primary rocks, resulting, most probably, from the decomposition 
of conglomerate: the sandstone was often good, although generally 
too scaly for building purposes. 


CHAPTER III. 
FROM FORT LARAMIE TO FORT BRIDGER. 


Wednesday, July 18.—Taking leave of our friends at Fort 
Laramie, we continued our journey this morning. The next place 
where we shall meet with a human habitation will be Fort Bridger, 
_ on Black’s Fork of Green River, distant about four hundred miles. 


While the train followed the travelling track, I took a road 


nearer the river, and examined a quarry which the workmen from 


the fort are here opening. ‘The strata exposed in the bluffs were — 


principally gray sandstone, with some thin calcareous layers, the 
general dip being south and south-west about 17°. On the oppo- 
site bank of the river, which is high and covered with pine, the in- 
clination appeared to be much greater. The fossils were quite 
imperfect. The only ones that could be descried were abundant 
remains of encrinites. The limestone at the quarry is dark, car- 
boniferous, with conchoidal fracture, and slightly foetid. It lies 
in layers of six or eight inches thick, and is immediately overlaid 
by slaty shales and gray sandstone. A considerable number of 
Productus semi-reticulatus was found in it, as well as in the sand 
between the layers: some specimens of a large sort of oyster lay in 
a nearly horizontal position. Some imperfect fossils were also ob- 
tained from the sandstone. The general surface is formed of 
strata of grayish sandstone and clay, the former varying in hard- 


ness, some being very friable, while others are exceeding hard, 


especially some which were slightly coloured by oxide of iron. 


RN a ee ee 


LIMESTONE QUARRY—BITTER CREEK. ou 


Near the quarry was a spring of pure cold water. A vein of trap, 
about six inches wide, passed perpendicularly through the quarry, 
and had evidently affected the rocks on either side. Leaving this 
spot, we struck across to the south, and joined the camp, which 
had been pitched just below a large warm spring that comes bub- 
bling out of the ground and forms immediately a small stream. 
Temperature of the spring, 71°. 

Above the mouth of Warm Spring Creek, the hills become in- | 
creased in height, and a lofty range runs north by west, evidently 
thrown up by internal convulsions, the strata having a considerable 
dip to the south-west. The banks of the Platte where it cuts 
through the range are apparently perpendicular, and from a dis- 
tance appeared to be composed of red sandstone. The general 
dip of the rocks, where not disturbed, seems still to be toward the 
south-west, though very slight. 

Auguste Tesson, one of my very best men, was taken sick to- 
day with something very like the cholera. 

Thursday, July 19.—Bar. 25.68; Ther. 80°. Leaving the 
valley of the Warm Spring Branch, the road crosses over to a 
branch of Bitter Creek, an affluent of the Platte, down the valley 
of which it winds until it reaches the main stream. We followed 
this valley the whole day, crossing the stream several times, and 
encamped on its left bank after a short march of ten and a-half 
miles. We were detained here the following day by the extreme 
illness of Auguste, who was unable to be removed. We passed 
to-day the nearly consumed fragments of about a dozen wagons 
that had been broken up and burned by their owners; and near 
them was piled up, in one heap, from six to eight hundred weight 
of bacon, thrown away for want of means to transport it farther. 
Boxes, bonnets, trunks, wagon-wheels, whole wagon-bodies, cook- 
ing utensils, and, in fact, almost every article of household furni- 
ture, were found from place to place along the prairie, abandoned 
for the same reason. In the evening, Captain Duncan, of the 
Rifles, with a small escort, rode into camp. He had left Fort 
Laramie in the morning, and was in hot pursuit of four deserters, 
who had decamped with an equal number of the best horses be- 
longing to the command. 

Bitter Creek is a fine clear stream, about fifty feet wide, with a 
swift current, and seems, from the great heaps of drift-wood piled 
up on its banks, to discharge a large quantity of water in the 
spring. 


56 FROM FORT LARAMIE TO FORT BRIDGER. 


Upon examining the bluff on the opposite side of the stream, the 
strata were found to be composed of sandstone and clay with sand. 
There was also a layer of sulphate of lime about four inches thick 
and crystalline. In some of the layers of sandstone there were 
ripple-marks of water; others were thickly studded with oval bo- 
dies about the size of pigeons’ eggs. Other strata were formed 
of more compact sandstone, not in layers but in irregular shaped 
masses, as if composed of bones, much resembling what we had 
remarked near Chimney Rock. Some fossils were collected, | 
but in not a very perfect state. In some of the sandstones there 
were evidently a great many, but in the more friable they were 
rotten; and in others the stone, in the endeavour to get them out, 
split in every direction. A crystalline mass of what was thought 
to be sulphate of lime was also found, with dark crystals inter- 
spersed. The top of the hill was covered with masses of primitive 
rock, probably from the decomposition of conglomerate. The 
hunters brought in the choice parts of three fat buffalo-cows to- 
day, which fairly loaded down their pack-mules. The meat was 
estimated to weigh upward of one thousand pounds. 

Saturday, July 21.—We followed up the dry bed of a fork of 
Bitter Creek for three or four miles, when it crosses over a high 
ridge and descends precipitously into a narrow ravine forming the 
heads of a branch of Horse-shoe Creek. Following down this 
ravine, which gradually widens into a broad valley, walled in by 
steep bluffs, much cut by ravines and entirely destitute of timber, 
we reached Horse-shoe Creek, a beautiful stream of running water, 
clear, soft, and very cool. ‘There are two tracks here, one crossing 
below the junction of the two forks, two hundred yards to the 
right, the other crossing both forks. The latter was taken, and 
after crossing the western forks, we followed up its valley for a 
couple of miles, over some very high, rolling country, and 
crossing over to the valleys of two dry sandy beds, came to a 
branch seven miles from Horse-shoe Creek, upon the left. bank of 
which we encamped. 

All the dry beds we have passed to-day give evidence of dis- 
charging large quantities of water, which, at the melting of the 
snows, descend from the Black Hills, a range immediately on our 
left. Their channels are full of rolled primary rock, feldspar, and 
white and pink quartz, brought down by the spring torrents. 
Upon the top of the dividing ridge between Bitter Creek and 


oe eo A ee eel te Pl Soy le) Sle NP GEN aro) 


U 


KN Aempeorg 6/6 Yn] wewtaeyoy 


~ 


a ns ee "a — Te ee es ee ees we ae yy, 


fl i Mas ina on te wi itn ala i titi iia Wares see 


* 


eae a 
Ss 
F. > 


Sahu 


“ 


HORSE-SHOE CREEK. 57 


Horse-shoe, we passed some enormous blocks of granite, lying 
upon the surface, some of which were cubes of twenty feet. 

The road, as usual, was strewn with fragments of broken and 
burnt wagons, trunks, and immense quantities of white beans, 
which seemed to have been thrown away by the sackful, their 
owners having become tired of carrying them farther, or afraid to 
consume them from danger of the cholera. The commanding 
officer at Fort Kearny had forbidden their issue at that post on 
this account. Stoves, gridirons, moulding-planes and carpenters’ 
tools of all sorts, were to be had at every step for the mere trouble 
of picking them up. 

The next day, being Sabbath, was passed in camp, during which 
hourly observations both of the thermometer and barometer were 
made, commencing at six o’clock. We are fifty miles from Fort 
Laramie. 

In descending the ridge into the valley of Horse-shoe Creek, a 
section of a stratum of reddish clay was exposed, some distance 
above the bottom, surmounted by a large and coarse sandstone. 
On the banks of the Horse-shoe, there was a perpendicular section 
of about one hundred feet of a stratum of clay and sandy limestone. 
The rock seemed very fossiliferous, but, owing to its fracturing in 
all directions, few specimens could be obtained. The peaks to the 
left seemed to be of reddish clay, so far as could be judged from 
their appearance and the manner of their disintegration. A con- 
siderable change has taken place in the flora as the country begins 
to ascend. Since leaving Fort Laramie, a variety of geranium has 
been frequent upon the borders of the streams. A small-leaved 
Cnothera, white, and the blue Digitalis, were also found. On the 
north side of the ridge, some plants were seen which we had not 
met with before; Azalea; a small white Gnothera, on a tall stem, 
the flowers not more than a line and a-half in diameter; two spe- 
cies of Potentilla, yellow, and two or three varieties of Campanula. 

Monday, July 23.—Ther. 47°. Ascending from the valley of 
the run where we had encamped, the road winds along a high, un- 
dulating ridge, for several miles, with very deep, precipitous ra- 
vines heading on each side, thus rendering our course very sinuous. 
The road then descends for about a mile and a-half into the broad 
valley of a run which has been on our left for three or four miles, 
and follows its dry bed until it strikes another fork coming in 
from the left, with a fine stream of running water, and a broad 
bottom, covered with willows of a large size. Our course this 


58 FROM FORT LARAMIE TO FORT BRIDGER. 


morning has been about parallel with the range of the Black Hills, 
the base of which could frequently be seen from the more ele- 
vated portions of the road. The valley of the Platte also was to 
be seen far in the distance tothe north. Crossing La Bonté Creek, 
encamped near a fine spring, after a fatiguing march of twenty- 
four miles. 

In passing along the ridge, the only rock exposed consisted of 
coarse sandstone, with a decided dip to the north-east of 15°, and 
beds of clay were interposed between the strata. On descending 
from the ridge into the valley of the fork, some siliceous limestone 
-was found, from which a few fossils were collected. In the same 
locality were masses of sandstone of a reddish colour. On the 
side of a ravine to the left of the road, the same sandstone ap- 
peared to crop out. After crossing this stream, the geological 
character of the country changed materially. The soil assumed 
a red colour, being composed principally of red sandstones and 


shales. To the left rose a high crest or ridge crowned with gray - 


and red sandstone, which was very hard. The general direction 
of this ridge was N.N.W.and 8.8.E., and it was manifestly 
formed by the upheaving of the strata, the dip being at least 45° 
to the north-east, while the south-west face of the rocks was very 
precipitous. At nearly the highest point of the ridge, a small 
ravine occurred, containing a layer of coal shale, apparently over- 
laid by dark shales, and layers of red clay and light slaty shales. 
The portions of these strata which were exposed were in so decom- 
posed a state that no fossils could be obtained except very imper- 
fect ones. 

In the valley of La Bonté Creek, the soil was found to be formed 
by the decomposition of highly ferruginous rock. Before the cross- 
ing of this stream, a ravine exposed a layer of gypsum, which was 
very fine and white, and of considerable thickness. Overlying it 
were layers of red sandstone and shales, conformable with the ap- 
parent dip of the sandstones forming the crest. A mile or two 
beyond the La Bonté, the gray sandstone was seen cropping out, 
overlying the red sandstone, and with a considerable dip to the 
north. Above these were layers of red and light shales and im- 
pure carboniferous limestone, from which a number of fossils 
(Avicula monotis, and Cardinia) were collected. The limestone 
seemed to be composed, to a great extent, of shells: in the more 
sandy and slaty shales the fossils were not so numerous. ‘To the 
left were some eminences composed of the more resisting sand- 


LA BONTE RIVER—LA PRELE. 59 


stones, the same as those forming the crest of the ridge, and which 
seemed to be the result of some force from beneath. The rocks 
were so broken that no indication could be obtained of their strati- 
fication. About one mile from La Bonté, the ridge on the left be- 
came lower, and the dip of the rocks was evidently less. They 
are composed of gray and white sandstone, clayey shales and clay, 
from which some few fossils were collected. 

The formation over which we have been travelling to-day seems 
to have been the result of the upheaving of the underlying strata, 
and the direction of the force would appear to have been from 
N. N. W. to S. 8S. E. It is not improbable that the trap and red 
sandstone seen on the Platte about twelve miles above Fort Lara- 
mie is the result of the same action, as that point would be in 
about the direction in which such a line would strike the river. 
The formation of the high ridges seen on the north side of the 
Platte is undoubtedly owing to the same cause. The strata ob- 
served to the right of the road before leaving the valley of the 
La Bonté were not conformable with those on the ridge, but had 
a slight dip to the south-east. They consisted of clay and coarse 
sandstone, and were analogous to those passed in the early part of 
the day. There can be but little doubt that they have been de- 
posited since the elevation of the ridge. | 

Tuesday, July 24.—Ther. at sunrise, 49°. Temperature of the 
- spring the same. The road to-day passes over a rolling country, 
being spurs of the Black Hills. We crossed the dry beds of several 
small streams, skirted, in some instances, with willows, box-elder, 
wild-cherry bushes, and occasionally with some large cotton-woods, 
until we reached the river La Préle. At this spot the mules were 
watered only, it being impossible to procure a blade of grass for 
them all the day. Consequently, we were obliged to continue on 
some four miles farther, until we reached the Platte, where we en- 
camped in a pretty little grove of large cotton-woods, with but a 
very scanty supply of grass. The artemisia seemed, to-day, to have 
taken complete possession of the country; and what little grass 
once grew along the road has been literally burned out by the 
passing emigrants. 

Upon arriving at the encamping ground, it was found, to our 
great grief, that the barometer was broken. It had been firmly 
fixed in an upright position, in the small spring-carriage, and care- 
fully strapped to one of the stancheons; but in coming down a 
steep and rough ravine, it must have received a jar that snapped 


60 FROM FORT LARAMIE TO FORT BRIDGER. 


the tube in two. The rates of the two chronometers, which were 
in the same wagons, were, however, unchanged. The large spring- 
wagon, containing all the instruments, came near meeting with a 
fatal accident to-day, by the breaking of a breast-chain while de- 
scending a steep hill, at the foot of which was a very precipitous 
ravine, and it was only saved from destruction by the promptitude 
and presence of mind of one of the men. 

The hills over which we passed to-day were composed of coarse . 
sandstone and conglomerate, with a slight dip to the south. From 
the summit of some hills to the left of the road, a view was ob- 
tained of the ridge we had passed over yesterday. It terminated 
in some low hills. To the W. 8S. W. of this ridge, and parallel to 
it, about twelve or fifteen miles distant, was another higher ridge, 
of which Laramie Peak appeared to be the most easterly elevation. 
Toward the west was yet another, running north-west, the soil at the 
base of which was formed from the decomposition of the red sand- 
stone and shales. On the ridge itself were found white and red 
sandstone, very hard, and foetid dark limestone, the same as that ob- 
served at the quarry near Fort Laramie. The dip was here fully 60° 
to north-east. Some fossils, Terebratula and Productus, were collect- 
ed here, principally from the limestone. The strata in the valley ap- 
peared to consist of sandstone and shales, and were not conform- 
able with the rocks forming the ridge. These rocks, nevertheless, 
were found cropping out in some places. On descending into the 
valley of the Platte, we passed a section of some gray sandstone, 
with, perhaps, some dark shales, probably carboniferous. | 

Wednesday, July 25.—Morning bright and cool. Brisk wind 
from north-west. A distance of five miles up the Platte, over asandy 
soil, brought us to Deer Creek, a bright, clear stream, running 
pleasantly through a large grove of timber, principally cotton- 
wood. Judging from appearances, this spot has been a favourite 
camp-ground for the emigrants. Property of every description 
was strewn about in all directions, and in much greater quanti- 
ties than we had yet seen. Just above the mouth of this stream, | 
there was a ferry over the Norh Fork of the Platte, at which I 
determined to cross the train. The means employed for this pur- 
pose were of the rudest and simplest kind. ‘The ferry-boat was 
constructed of seven canoes, dug out from cotton-wood logs, fas- 
tened side by side with poles, a couple of hewn logs being secured 
across their tops, upon which the wheels of the wagons rested. 
This rude raft was drawn back and forth by means of a rope 


ee AC ee ONS ee a Ol NSS Od 


af 


e 


5 


ee ee 


= 


DEER CREEK—COAL BEDS—CROSSING OF NORTH FORK. 61 


stretched across the river, and secured at the ends to either bank. 
Frail and insecure as was the appearance of this very primitive 
ferry-boat, yet all the wagons were passed over in the course of 
two hours, without the slightest accident, although many of them 


were very heavily laden. The animals were driven into the stream ~ 


and obliged to ferry themselves over, which they did without loss, 
although the river was now somewhat swollen by late rains and 
the current extremely rapid and turbid. ‘The ferrymen informed 


me that an emigrant had been drowned here, the day before, in 


essaying to swim his horse across, which he persisted in attempt- 
ing, notwithstanding the earnest entreaties and warnings of his 
friends. They told us that this man made the twenty-eighth vic- 
tim drowned in crossing the Platte this year; but I am inclined 
to believe that this must be an exaggeration. The charge for 
ferriage was two dollars for each wagon. The price, considering 
that the ferrymen had been for months encamped here, in a little 
tent, exposed to the assaults of hordes of wandering savages, for 
the sole purpose of affording this accommodation to travellers, was 
by no means extravagant. 

A short distance above where the road crosses Deer Creek, coal 
was found cropping out of the bluff on the left bank of the stream. 
Ascending the creek, the direction of which was about north by 
west, the strata were inclined at an angle of about three degrees, 
but not at right angles to the dip, which appeared to be north by 
east. The coal was lying on a stratum of white sandstone of con- 
siderable thickness; above it were some dark shales; and above 
these, gray sandstones, in which latter were found fossils of Szgdl- 
laria, and, in those under the coal,stems of Calamites ; but as the 
only examples that could be obtained were from rocks which had 
been exposed to the action of the weather, they were imperfect. 
The stratum of coal was three or four feet thick, and resembled 
the cannel coal very much; but as the only specimens obtained 
were very much weathered, this could not be ascertained with cer- 
tainty. As the strata rose, the coal could be traced ascending 
the hills on the side of the bank, and the deeper underlying rocks 
became more fully exposed. They consisted of sandstones, vary- 
ing in colour from red to gray, and hi ak: many fossils, prin- 
cipally vegetable. 

The road, after crossing the river, runs mostly on the side of 
the bluffs, whieh here approach auek nearer than on the south 
side. They consist of reddish sandstone, containing some curious 


62 FROM FORT LARAMIE TO FORT BRIDGER. 


fossils. Among these were perfectly rounded masses, fully a foot 
in diameter, and others of stone apparently contorted like a rope 
tied into knots; they all appeared to have been attached to a 
stem. The formation, on the left side of the river, consists of 
sandstone with some beds of clayey shales and slaty shales. Salt 
was found efflorescing on the rocks in two or three places, and this 
was the case also on the opposite side, at the coal-beds on Deer 
Creek. Artemisia was almost the only vegetation, and great dif- 
ficulty was experienced in obtaining enough grass to subsist our 
animals. ‘The soil appears very barren, more from the absence 
of moisture than from the character of its constituents; as even 
the alluvial bottoms exhibit the same destitute and naked features. 
The road, since crossing, has been through deep sand, making the 
travel extremely slow and fatiguing. Day’s march, fifteen and 
a-half miles. 

Thursday, July 26.—Harly this morning we passed a small 
island in-the river, promising a welcome supply of grass for our 
wearied animals, which for the last three days have had very little 
to eat, and begin to exhibit the effects of this want of nourish- 
ment. Having afforded them a hearty meal, we continued up the 
valley to a high bluff running to the river: we crossed it and 
encamped in a deep valley beyond, where some pools of standing 
water afforded drink for our stock. The grass was quite sufficient. 
Some of the pools were so highly impregnated with salt as to be 
quite unpleasant to the taste. 

The general character of the formation is the same as that of 
yesterday—sandstone and shales: coal was found in two places, 
the first near the island where we nooned, and again about five 
miles beyond. In both cases the overlying strata were shales and 
clay; and that beneath, sandstone. The dip has been mostly to 
the north; but, where the coal was last seen, the strata were found 
dipping to the south-west, at an angle of about 20°. They then 
became horizontal, and in about half a mile the dip was again to 
the north. Some specimens of shale, with impressions of leaves, 
were collected, and also some crystallized carbonate of soda. 

Friday, July 27.—Morning bright and cool. The road for the 
greater part of the day has been through deep, heavy white sand, 
of which the hills seem to be chiefly composed. Turning off from 
the river, we left the road for the purpose of finding grass, and 
encamped directly behind the Red Buttes, five miles distant from 
the river, on the margin of a small lake or pond, formed by nu- 


ABANDONMENT AND DESTRUCTION OF PROPERTY. 63 


merous little springs of very cold and excellent water. Grass was 
found on a neighbouring hillside sufficient for our animals. 

The road to-day passed over from the Platte, crossing a spur of 
of the mountains. Above this point, a high range of hills, which 
had been observed running to the north-west, inclined rather more 
to the north side of the river, which here forces a passage for 
itself through a gorge of the mountain. The strata there pre- 
sented were of red sandstone and shales, whence the name of 
<«¢ Red Buttes.’’ The rocks were inclined at an angle of about 25°, 
with a dip to the west, as were also the strata on the north side. 
The sections presented were of sandstones, white or red, shales, 
slaty shales, and clay. Considerable quantities of nitrate and car- 
bonate of soda were found on the surface. 

To-day we find additional and melancholy evidence of the diffi- 
culties encountered by those who are ahead of us. Before halting 
to noon, we passed eleven wagons that had been broken up, the 
spokes of the wheels taken to make pack-saddles, and the rest 
burned or otherwise destroyed. The road has been literally 
strewn with articles that have been thrown away. Bayr-iron and 
steel, large blacksmiths’ anvils and bellows, crow-bars, drills, 
augers, gold-washers, chisels, axes, lead, trunks, spades, ploughs, 
large grindstones, baking-ovens, cooking-stoves without number, 
kegs, barrels, harness, clothing, bacon, and beans, were found along 
the road in pretty much the order in which they have been here 
enumerated. The carcasses of eight oxen, lying in one heap by 
the roadside, this morning, explained a part of the trouble. I re- 
cognised the trunks of some of the passengers who had accom- 
panied me from St. Louis to Kansas, on the Missouri, and who 
had here thrown away their wagons and every thing they could not 
pack upon their mules, and proceeded on their journey- At the 
noon halt, an excellent rifle was found in the river, thrown there 
by some desperate emigrant who had been unable to carry it any 
farther. In the course of this one day the relics of seventeen 
wagons and the carcasses of twenty-seven dead oxen have been 
seen. Day’s march, twenty-four miles. 

Saturday, July 28.—Morning bright and pleasant, but at 9 
A. M. the wind rose from the south-west, and blew almost a hurri- 
cane the whole day, tearing up the sand and gravel, and dashing 
it into our faces, as we rode, with such violence as to cause sensi- 
ble pain. It was impossible to look up for a moment, as the eyes 
became immediately filled with sand, so that the teamsters were 


64 FROM FORT LARAMIE TO FORT BRIDGER. 


obliged to fasten their handkerchiefs over their faces to enable 
them to see where they were going. ‘This has been the most dis- 
agreeable day’s travel we have yet experienced; for the wind, in 
addition to its furious violence, was so very hot and dry as to ren- 
der respiration, from the great rarefaction, quite difficult. The 
throat and fauces became dry, the lips clammy and parched, and 
the eyes much inflamed from the drifting dust. <A pair of green 
goggles partially remedied this latter annoyance; and I would ad- 
vise every one who contemplates a journey across these sandy 
plains, to provide himself with several pairs before starting. They 
afford great relief from the incessant glare of a bright sun, to 
which he may make up his mind to be constantly exposed during 
the whole of his weary route. With all our efforts, owing to these 
opposing causes, our day’s march was only eighteen miles, and we 
encamped on the head of a spring, one hundred and sixty-four 
miles from Fort Laramie and forty-four miles from the ferry, 
and remained at this camp over Sunday. 

The country, all the way from the crossing of the Platte, is a 
dry, sterile, and dreary desert. The artemisia constitutes nearly 
the whole growth, and what little grass had come up has been 
completely eaten off by the hundred thousand animals that have 
passed before us. Thirty-one head of dead cattle were passed on 
the roadside to-day, and on the bank of a small drain, where the 
efflorescence of alkaline matter was very abundant and rendered 
the water nauseously offensive, nine oxen lay dead in one heap. 
They had been poisoned, doubtless, by the water. Our accompany- 
ing friends occupied a portion of Sunday in selecting such articles 
as they could best spare, and threw them away_to lighten their 
load, their animals beginning to fail quite sensibly. The day was 
cool, with a fresh breeze from the north. Thermometer at sun- 
down, 52°; and at 10 p. m. 44°. : 

Monday, July 30.—Ther. at sunrise, 29°. Morning very cold. 
Ice, half an inch thick, had formed during the night in the water- 
buckets, and a faint white-frost was visible on the ground. To- 
day we crossed over to the Sweetwater River, descending into its 
valley by the side of a small tributary, whose course was nearly 
south, and encamped on the left bank of this beautiful little stream, 
a mile below Independence Rock. The river is about seventy feet 
wide, from six to eighteen inches in depth, with a uniform and 
tolerably rapid current of clear, transparent water. 

In the valley of the tributary opposite our noon halt, some 


INDEPENDENCE ROCK—DEVIL’S GATE. 65 


masses of igneous rocks, granite and serpentine, protruded con- 
siderably above the soil. The direction of the dike was from north 
to south. The strata through which they protruded did not appear 
to be disturbed, nor were any specimens of metamorphic rocks 
found near them, although, had such existed, their presence would 
have been indicated, as being more resisting than the sandstone of 
which the surface rock is generally composed. Wherever any 
sections of the latter have been obtained, they were found to be 
either coarse white or gray sandstone, and clay, with coarse con- 
glomerate. Some distance below this dike, the igneous rocks 
again became visible, and at this point portions of white sandstone 
were found overlying the lower parts of the rock, and apparently 
in horizontal layers. The surface of the ground in the vicinity, 
where water had apparently stood, was coated with a white saline 
substance, a portion of which was collected. 

The same substance has been observed within the last two days 
on the surface in those localities where water has evaporated. 
Near our encampment this evening, large masses of igneous rock 
protrude in every direction, but the sandstone near them does not 
appear to have been disturbed. 

Few or no flowers have been met with on this portion of the 
route, owing, in part, to the dryness of the atmosphere, (due to 
our increasing elevation,) and in part to the occurrence of frosts 
during the summer months. 

Tuesday, July 31.—Ther. at sunrise, 40°. Leaving camp 
we continued up the valley of the Sweetwater, and passed the 
‘far-famed «Independence Rock,” a large rounded mass of granite, 
which has frequently been described by travellers. It was covered 
with names of the passing emigrants, some of whom seemed deter- 
mined, judging from the size of their inscriptions, that they would 
go down to posterity in all their fair proportions. A short dis- 
tance -beyond was a range of granite hills, stretching entirely 
across the valley, and continuous with a range extending to the 
north. Through this range the Sweetwater passes in a narrow 
cleft or gorge, about two hundred yards in length, called the 
«‘ Devil’s Gate.’’ The space between the cliff, on either side, did 
not in some places exceed forty feet. The height was from three 
to four hundred feet, very nearly perpendicular, and, on the south 
side, overhanging. Through this romantic pass the river brawls 
and frets over broken masses of rock that obstruct its passage, 


affording one of the most lovely, cool, and refreshing retreats from 
5 


66 FROM FORT LARAMIE TO FORT BRIDGER. 


the eternal sunshine without, that the imagination could desire. 
It is difficult to account for the river having forced its passage 
through the rocks at this point, as the hills, a very short distance 
to the south, are much lower, and, according to present appear- 
ance, present by no means such serious obstacles as had been here 
encountered. It is probable, that when the cafion was formed, 
stratified rocks obstructed it in that direction, and that these rocks 
have since disappeared by slow disintegration. The granite rocks 
of the pass were traversed in many places by dikes of trap, which 
were in some instances twenty feet thick, whose direction was 
east and west. South of the pass, at its eastern extremity, stra- 
tified rocks, consisting of conglomerate, were observed, in a nearly 
horizontal position, without exhibiting the least evidence of having 
been disturbed by the igneous rocks around which they were 
placed; indeed, they could be traced in close contact with the 
granite, without any displacement of the strata, proving that their 
formation must have been subsequent to that of the granite, from 
the disintegration of which they were composed. The conglome- 
rate is of the same character as that which was observed before 
coming upon the carboniferous rocks. The rocks were not ob- 
served to have any marked dip. It is highly probable that they 
belong to a period subsequent to that in which the carboniferous 
rocks were formed, and that the eruption of granite took place after 
the latter formation, but before that of the conglomerate. No 
dikes of trap were observed in the granite, except in the immediate 
vicinity of the Devil’s Gate. 

After passing this remarkable cafion, we enter upon a broad 
level valley, bounded on each side by ranges of mountains, their 
summits broken into curious peaks and eminences entirely 
destitute of vegetation. Between these winds the Sweetwater, 
with a current more gentle than heretofore, its banks covered 
with grass. An accident occurring to one of the wagons, the 
remainder of the day was consumed in its repair. Thermometer 
at sunset, 70°. 

Wednesday, August 1.—Ther. at sunrise, 33°. Frost during the 
night; morning clear, calm, and very beautiful. The road passing 
occasionally through deep, heavy sand, continued up the right 
bank of the Sweetwater, which, for the greater part of the morning, 
flowed at the foot of a long, high range of granite bluffs, with here 
and there a stunted cedar growing from the crevices in the rocks. 
The valley is here nearly two miles wide, with rolling hills between 


SALHRATUS LAKE. 67 


the two mountain ranges, which bound it on either side and form 
its limits. The artemisia seems to have taken complete possession 
of the soil, growing five or six feet high, with twisted stems, re- 
sembling somewhat in their texture the ground cedar of the New 
England coast. At the Devil’s Gate I noticed some, quite seven 
feet high and nine inches in diameter. The Wind-River Moun- 
tains were seen on the horizon—their sides, for a third part of the 
distance down from their summits, glistening with snow, in the 
rays of the morning sun, like burnished silver. They presented a 
very beautiful object. 

About a dozen burnt wagons and nineteen dead oxen were 
passed to-day along the road; but the destruction has been by no 
means as great as upon the North Fork of the Platte and the cross- 
ing over to the Sweetwater. 

In the morning we passed what at a distance appeared to be a 
small lake or pond, frozen over and covered with a very light fall 
of drifting snow. The illusion was perfect, and was maintained 
to the last moment, even when riding up to its very margin. It 
was found to bea slight depression, about four hundred yards 
long, by one hundred and fifty in width, covered with an efflo- 
rescence of carbonate of soda, deposited on the ground from the 
evaporation of the water which had held it in solution. This 
substance, indeed, covers a large portion of the country, and is 
quite abundant on the banks of the river. The emigrants use it 
in mixing their bread, and prefer it to the saleratus of the shops 
for that purpose. 

On the south range of mountains, about fifteen miles above the 
Devil’s Gate, a ravine, formed by a small stream, exposed a 
section of the rocks. The strata evidently belonged to the car- 
boniferous system, and had been acted upon by heat. The lime- 
Stone was in some instances converted into marble, and other spe- 
cimens afforded examples of its partial change. It belonged to 
the dark slaty-coloured variety, such as had been found at the 
_ quarry near Fort Laramie, and has since been seen wherever the 
lower beds have been tilted up. The red and white sandstone 
had also been affected by heat, their structure being more crystal- 
line. The dip was to the south, and was very great, varying from 
T0° to 80°, and some of the lower beds of the red sandstone were 
nearly vertical. The sequence of the strata seemed to be exactly 
the same as had been observed near the Red Buttes. On the west 
surface of the hills, which were precipitous, and covered to a great 


68 FROM FORT LARAMIE TO FORT BIDGER. 


extent with pine, dark bands of soil were observed, having the 
appearance of being formed by the decomposition of carbonaceous 
matter, which however must have been altered by heat. It is 
highly probable that these beds have been tilted up by some gra- 
nite rock, although none appeared on the surface, being probably 
covered up by the secondary beds which have been deposited since 
the eruption took place. The valley between the hills and the 
river gradually rose, and where the rocks had been protruded 
must have been at a considerable elevation above the latter. They 
were composed of white and gray sandstones, evidently fossili- 
ferous, but the fossils were in so decomposed a state that no speci- 
mens, with the exception of some imperfect encrinital stems, 
could be obtained. The beds were nearly horizontal, and were ana- 
logous to the same formation over which we had been passing all 
along the Platte. Toward the close of our march to-day, the sum- 
mit of one of these lower hills was found to be composed of clay 
similar to that of the Chimney Rock, and to the formations in 
that vicinity. The elevations, though not so marked as to pecu- 
liarity of form, resembled them in their general features. The 
rocks on the north side of the river are still granitic. 

Encamped, after a march of twenty-one and a-half miles, on the 
right bank of the river, which is here growing smaller and the 
current more gentle. In the course of the day we passed a party 
of eight wagons from Iowa, bound to the land of gold. A number 
of women and children were of the party, and application was 
made for medical attendance upon one of the former, who was 
about being confined. | 

Friday, August 3.—Ther. at sunrise, 31°.. The rocky ridges 
gradually disappeared as we followed up the valley of the Sweet- 
water, occasionally crossing long and lofty spurs which would not 
admit of our passing between them and the river. The soil is 
very barren, producing only the artemisia, and two or three varie- 
ties of grass; a species of Jrzs, which is quite abundant,the Linum 
ceruleum, a yellow Potentilla, a scarlet star-shaped flower, with 
tubular corolla,and Ginothera hispida were seen in the bottoms. 

The character of the valley for the last two days has been ana- 
logous to that of the Platte below Fort Laramie. The last of the 
granite was passed yesterday. The hills on both sides of the 
river have since been formed of sandstones and clay, the latter of 
which forms the irregular outline of the bluffs above Ash Hol- 
low, and in many instances presenting a similar appearance when 


CANON OF THE SWEETWATER. 69 


it is found on the tops of the hills. The strata seemed perfectly 
horizontal. At our nooning point yesterday, the carboniferous 
rocks were found fising up at a considerable angle, but no section 
was obtained. The only rock exposed was the red sandstone, 
which had been rendered partially crystalline by the action of 
heat. The surrounding rocks had not been disturbed. Some | 
beds of very coarse conglomerate were seen cropping out near the 
river. 

In the course of the day, Captain Duncan, of the Rifles, who had 
passed us at Bitter Creek on the 19th of July, in pursuit of de- 
serters, came into camp, having followed his men to within fifty 
miles of Fort Bridger. He had come upon them at daylight, while 
they lay asleep, disarmed them, secured their horses, and was now 
on his return, having taken also another man who had deserted 
previously—so that he had five in all. The pursuit was one of 
_ great hardship, privation, and fatigue, and the energy and perse- 
verance with which it had been continued was the subject of admi- 
ration with all. Encamped on the Sweetwater, at a point where 
the road leaves it to avoid a caiion above, which is impassable for 
wagons for several miles. March, in the last two days, forty 
miles. | 

Saturday, August 4.—Morning clear and cool. Leaving the train 
to follow the beaten track, which makes a short cut over the hills, 
I determined to follow up the cajion of the Sweetwater. The 
stream, as I had anticipated, was shut up between lofty, rocky 
eminences, coming down directly to the water at an angle of from 
45° to 60°, along the sides of which we scrambled, sometimes 
walking and leading our mules over crags where it was impossible 
to ride, crossing and recrossing the stream ever and anon, to 
enable our animals to get along at all. A short distance after 
entering the cafion, the red sandstone was found cropping out at 
an angle of 45°, with a dip to the north; and a little farther 
on the crystalline rocks appeared, forming the sides of the 
eafion. The prevailing rock was gneiss; but sienite and granite 
were found in some places constituting the principal bulk of the 
formation. A narrow bottom occasionally gave room for some 
fine groves of large aspens, the sight of which, after our long 
and dreary ride without a particle of shade, was truly refreshing. 
The bed of the river was filled with large boulders and fragments 
of rock which had fallen from the cliffs above, among which the 
naters foamed and fretted with a gurgling murmur, which, when | 


70 FROM FORT LARAMIE TO FORT BRIDGER. 


contrasted with the flat, silent waters of the Platte, was very plea- 
sant to the ear. It reminded one of the clear, purling streams we 
had left at home. ° 

The river here is truly a mountain-stream, with great fall, 
rapid current, and water as clear as crystal, of the tempera- 
ture of 55°. On emerging from the district of primary rocks, we 
came upon the stratified, which were formed of micaceous, slaty 
shales, and red sandstone, all evidently metamorphic: their beds 
were inclined at an angle of from 40° to 70° to the north. Veins 
of quartz were observed in them in some places. Trap dikes 
were very frequent in the crystalline rocks. The surface rock on 
the hills in the vicinity appeared to be a continuation of the same 
shaly formation. The aspen, beech, willow, and cotton-wood were 
found growing on the bottoms, and on the hills cotton-wood, pine, 
and cedar. Sage hens, (Tetrao urophasianus,) a species of grouse, 
were seen in great numbers, and the men shot as many as we could 
conveniently carry. They are very good eating, and some of the 
older ones were larger than a full-grown barn-door fowl. 

In the afternoon, we met the mail from Great Salt Lake City, 
with upward of six thousand letters, and were glad to avail our- 
selves of the opportunity to write to our friends. Camp on 
Sweetwater. Day’s march, twenty-three miles. 

Monday, August 6.—Leaving the valley of the Sweetwater, we 
crossed this morning through the South Pass over to the head 
branches of Sandy Creek, an affluent of the Colorado, or Green 
River of the West, and nooned at the « Pacific Springs,” at the 
foot of the pass, on the western side. This celebrated depression 
through the Rocky Mountains is now so well known that any fur- 
ther description of it would be superfluous. That of Frémont 
conveys a very accurate idea of the locality, which has nothing 
remarkable in its features. The water at the Pacific Springs 
is not very good, but is quite cold. It is a favourite camping 
ground of the emigrants on account of the grass. Hncamped for 
the night on the banks of Dry Sandy, where we had to dig in the 
bed of the stream for water; but a very scanty supply was ob- 
tained; and the grass moreover was so scarce that our animals 
were allowed to run loose all night under the protection of the 
guard, instead of being picketed as usual. In the afternoon, one 
of our best mules died from the bite of a snake. In the morn- 
ing her jaws and fauces had been observed to be very much 
swollen, and before sundown she became so weak that we were 


SOUTH PASS—EMIGRANT RUSE. TL 


obliged to release her from the wagon, when she lay down by the 
side of the road and in a short time expired. 

Between the Sweetwater and the South Pass, the soil for some four 
or five miles presented the same disintegrated dark shales as had 
been observed on the other side of the river. It then became more 
sandy, and portions of weathered marble were found on the surface. 
On ascending some low hills on the left of the road, and within 
about a mile of the Pass, marble was found in place, contain- | 
ing a considerable incrustation of silex. It evidently cropped 
out on the south side of these hills, on the top of one of which was 
found a stratum of gray sandstone, in which the remains of en- 
crinites were observed. It was quite horizontal, not conformable 
with the marble under it, and was undoubtedly a continuation, of 
the secondary formation which had been observed up the whole 
valley of the Sweetwater. On the left of the road, and a few miles 
distant, were some high hills, which, from their appearance, seemed 
to be capped by the reddish clay which forms the isolated masses 
in the valley of the Platte. Shortly after passing the summit we 
found a stratum of apparently metamorphic clay, horizontal, with 
an east and west direction. Over this were strata of gray sand- 
stone, horizontal, or with a slight dip to the east. Descending 
the western side of the Pass, the soil was composed principally of 
red sand. No rocks were visible. About a mile from Dry Sandy, 
some masses of rock were observed on the right of the road, stand- 
ing up like pillars; they were found to be composed of a coarse 
sandstone, of an ochrey colour. Under them were white and red 
shales, apparently horizontal. The surface of the ground appeared 
to be the result of the decomposition of this ochrey rock. 

I witnessed, at the Pacific Springs, an instance of no little in- 
genuity on the part of some emigrant. Immediately alongside 
of the road was what purported to be a grave, prepared with more 
than usual care, having a headboard on which was painted the 
name and age of the deceased, the time of his death, and the part 
of the country from which he came. I afterward ascertained that 
this was only a ruse to conceal the fact that the grave, instead of con- 
taining the mortal remains of a human being, had been made a safe 
receptacle for divers casks of brandy, which the owner could carry 
no farther. He afterward sold his liquor to some traders farther 
on, who, by his description of its locality, found it without difli- 
culty. | 

Wednesday, August 8.—In our march, yesterday, to the Little 


/ 


12 FROM FORT LARAMIE TO FORT BRIDGER. 


Sandy, where we encamped, nothing of interest was observed. In 
Little Sandy the same strata as had occurred previously were 
found, with a dip of 8° to the south. <A section of the rock in the 
vicinity, exposed on the top the same ochrey-coloured sandstone, 
and then red shales. From the shales were obtained some remains 
of plants, but the rock was in so decomposed a state that they could 
not be identified. On the road, some fragments of limestone were 
found on the surface, containing fossils, but we could not secure 
any specimens. From Little Sandy to Big Sandy, artemisia covers 
the whole face of the country, which has a dreary, barren aspect. 
Near our camp of this morning, a small section presented thin la- 
minated white sandstones and clayey shales; and from the appear- 
ance of the country for several miles, this must have been the 
character of the rocks. Twelve miles from the Little Sandy, on 
descending a ravine, fossiliferous trunks of large trees, some of 
them nearly two feet in diameter, were observed upon the ground: 
the interior of some of these was hollowed out, but concentric rings 
were noticed near the circumference, and, in some specimens, lon- 
gitudinal fibres were found in the interior. The bark appeared to 
be marked in places for the attachment of leavesof Cycadec, but 
they were all much weathered. The rocks on the river-bank were 
white compact sandstone, disposed in thin lamelle, sandy and 
clayey shales, and a gray compact limestone, breaking with a 
conchoidal fracture. Some large portions of trunks of trees were 
protruding from the cliff, imbedded in apparently arenaceous 
shales. Some few specimens of fossils (Mawtilus and corals) were 
collected, but, on account of the weathered state of the rocks, they 
were necessarily imperfect. ‘The limestones. contained but few 
fossils. 

Thursday, August 9.—Our road to-day lay along the right bank 
of Big Sandy, until we reached Green River, which we crossed 
above the junction, and encamped a couple of miles below. The 
increased altitude, and the consequent dryness of the atmosphere, 
had so shrunk the woodwork of many of our wagon-wheels, that 
various expedients had to be resorted to, in order to prevent them 
from falling to pieces. To-day one of the wheels of the instru- 
ment-wagon, that precious and important portion of our train, be- 
came so weak from this cause that I was forced to take out nearly 
all the load, and distribute it among the other teams, to enable us 
to reach camp with it.. We picked up a pair of hagle belonging 
to some emigrant-wagon, but they would not answer; so we were 


BIG SANDY—COLORADO OR GREEN RIVER. 73 


obliged to wedge up the wheel as well as we could, and to sink it 
in the river during the night, to swell the wood. 

The bluffs on Big Sandy presented several sections. The 
strata consisted of thin layers of clayey shales, argillaceous gray 
limestones, and of crystallized sulphate of lime. These strata 
were seldom more than two or three inches in thickness, the 
layers of gypsum being about halfan inch. There were also thin 
laminz of dark slaty shales above the gray limestone. Remains 
of plants were found, and also some imperfect shells. Near the 
junction of Big Sandy with Green River some large nodules of 
ferruginous sandstone were observed, and near them the remains 
of trees of a large size. The dip was about 5°, a little to the west 
of south. Upon Green River we came upon a layer of brownish 
clay, of considerable thickness; and from the appearance of the 
hills on our right, it is probable that they are formed by the 
weathering of this bed of clay. Green River,.or the Colorado of 
the Gulf of California, is here a fine, bold, clear stream, discharg- 
ing a large quantity of water. In its bed are found rounded 
stones, consisting principally of detritus of primary rocks. Day’s 
march, only thirteen miles. 

Friday, August 10.—After travelling about four miles, the road 
leaves Green River and crosses over a ridge dividing it from 
Black’s Fork, one of its tributaries, upon the left bank of which 
we encamped, with abundance of grass, and wood obtained from 
the willows which here fringed its banks. The Uintah mountains 
were distinctly seen far to the south, their sides glittering with 
snow in the rays of the setting sun. 

In a ravine of the western bluff of Green River valley, the rocks 
were partially exposed, particularly some strata of hard white 
sandstone. They consisted of sandstone in lamin, about fifteen 
inches thick, and layers of argillaceous shales, and argillaceous 
limestone, made up principally of shells. There were evidently, 
also, some strata of conglomerate, formed chiefly of serpentine, 
sienite, and trap. Among the rounded pebbles found near the 
top of the ravine was one of dark foetid limestone, partially con- 
verted into marble, and precisely similar to the metamorphic lime- 
stone observed flanking the granite chain in the vicinity of the 
South Pass. A specimen of fossil wood was also found, which ap- 
peared to have been partially converted into lignite. The strata 
were, as far as could be judged, horizontal, and a section on the 
opposite side of the valley showed the same sandstone maintaining 


T4 FROM FORT LARAMIE TO FORT BRIDGER. 


about the same level. If there was any dip, it was to the south, 
although certainly not so great as had been observed in the strata 
since leaving the South Pass. Quite a number of fossils were 
collected here. The character of the soil in the valley of Green 
River would indicate a considerable preponderance of argilla- 
ceous elements in the rocks. The ridge between it and Black’s 
Fork afforded an imperfect section of the rocks. ‘They were, how- 
ever, mostly covered by clay. The less decomposed rock consisted 
of white and slaty sandstone, a coarse-grained grayish sandstone, 
and a thin stratum of limestone. But few fossils could be col- 
lected, the surface of the rocks being almost entirely concealed by 
blue clay—the result of the disintegration of strata which seemed 
to compose the greater portion of the mass. The strata were very 
nearly horizontal, although a slight dip to the north was suspected. 
Common salt was found on some of the stones. The layer of 
limestone was near the top of the rocks. 

Saturday, August 11.—Ther. at 6 o’clock, 40°. <A drive of 
thirty-two miles, during which we crossed Ham’s Fork and Black’s 
Fork three times, brought us to Fort Bridger—an Indian trading- 
post, situated on the latter stream, which here branches into three 
principal channels, forming several extensive islands, upon one of 
which the fort is placed. It is built in the usual form of pickets, 
with the lodging apartments and offices opening into a hollow 
square, protected from attack from without by a strong gate of 
timber. On the north, and continuous with the walls, is a strong 
high picket-fence, enclosing a large yard, into which the animals 
belonging to the establishment are driven for protection from both 
wild beasts and Indians. We were received with great kindness’ 
and lavish hospitality by the proprietor, Major James Bridger, one 
of the oldest mountain-men in this entire region, who has been en- 
gaged in the Indian trade, here, and upon the heads of the Mis- 
sourl and Columbia, for the last thirty years. Several of my 
wagons needing repair, the train was detained five days for the 
purpose, Major Bridger courteously placing his blacksmith-shop at 
my service. 

In a ravine to the right of the ford of Ham’s Fork, on the north 
bank, white sandstone was found cropping out, and here some speci- 
mens of very perfect shells were collected. Under the sandstone 
were argillaceous shales, and above them apparently a thin stratum 
of limestone and rolled pebbles. The general surface of the coun- 
try appeared to be the result of the disintegration of argillaceous 


ee Ala NEED WO AeOud SHAOV la Seaodliaa tao 


ate me 


el 


KN Aempeoig ¢) eC uny ueutexoy : 


ry 2 ms i he 7 
Sa f c 
j 4 
{ 
- 
$ / 
, 
7 
if } 
e 
¥ f 
e - 
P 
r : 5 ; 
¥ + 


a a r 
ep } x © . 4 
7 > oe Pe ee ro 
- a - eet: nla ’ 
‘ ’ " Ts 
. . e Ss ~ Ce r ; 
< : -® ; Gi I 
te ‘ ie = . 


— 


. 


Sy 


ad 
¥ 
» 
. 
: 
~ 
: a 
‘6 
Ry 
ies 
5 
>i 7 
oe 
ae ¢ 
oe me 
‘| 


FORT BRIDGER. 75 


rocks and some sandstones, the former greatly predominating. In 
some places, layers of blue and red clay, of considerable thickness, 
were observed, and also gray argillaceous limestone; but owing to 
the surface of the rocks being covered by the clay detritus, no 
complete sections of them were obtained, nor were any fossils found 
in them. Farther on, upon Black’s Fork, a ridge was crossed in 
which the different layers of the rocks were quite apparent. Here 
two strata of coarse reddish sandstone, which disintegrated but 
slowly, were prominent. ‘They were separated by layers of lime- 
stone, shales, and clay. The shales were so nearly horizontal that 
no dip could be ascertained. Following the valley of Black’s Fork, 
we passed, in the afternoon, strata of green and white sandstone, 
whitish shales, and clay, but too much decomposed to afford us any 
fossils. In the valley were found some blocks of limestone, wholly 
made up of the debris of shells. The valley of Black’s Fork, in the 
vicinity of Bridger’s Fort, is three or four miles broad, and many 
of the numerous little branches into which the stream is divided 
are handsomely fringed with thickets of cotton-wood. The soil is 
composed of the detritus of the surrounding rocks, and huge blocks 
of metamorphic rock, and some trap and serpentine, are found upon 
the surface. At the fort we were shown a piece of rock, evidently 
volcanic, in the cells of which were contained some particles of 
gold. It was stated that it had been found in the bed of Black’s 
Fork in the vicinity. If so, it must have been washed down from 
the Uintah chain of mountains, a lofty range to the south, in which 
the stream heads. These mountains were at this time covered with 
snow for a considerable distance from their summits; but of their 
lithological formation no opinion could be formed, as their great 
distance precluded all opportunity of examining them. The strata 
of the hills around the fort were nearly horizontal, and consisted 
of gray limestone, clayey and slaty shales, and sandstones. Some 
few miles up the stream, sections of these rocks were found, from 
which some fossils were obtained. In the beds of the stream were 
found rounded rocks, composed principally of metamorphic sand- 
stone, and some marble. The same rocks were seen in horizontal 
strata on the hills. 

From Fort Bridger there are now two routes as far as the Hum- 
boldt or Mary’s River, where they again unite. The old road 
strikes Bear River, follows down its valley by the Soda Springs to - 
Fort Hall, whence it pursues a south-westerly course to the Hum- 
boldt. By this route a northing of nearly two degrees is made, 


76 FROM FORT LARAMIE TO FORT BRIDGER. 


and the road, consequently, is much lengthened. The other route 
was laid out by the Mormon community in 1847, and conducts the 
emigrant to their city, in the southern part of the Salt Lake 
valley, causing him to vary from the line of his direction rather 
more than a degree southwardly: this he has to recover by a direct 
north course to the crossing of Bear River near the north end of 
the lake, whence he proceeds in a north-west direction, until he 
intersects the old road from Fort Hall. I was desirous of as- 
certaining whether a shorter route than either of these could 
not be obtained by pursuing a direct course to the head of the 
lake, or to the point where Bear River enters its basin through 
the Wahsatch range from Cache Valley. If practicable, such a 
trace would save the emigration the great detour that has to be 
made by either of the present routes, and would have a direct 
bearing upon the selection of a site for the military post contem- 
plated for this region. I had ascertained that this selection had 
not yet been made, the officer designated for that purpose having 
determined to winter his command in temporary quarters, in the 
vicinity of Fort Hall. A glance at the map will show the import- 
ance of this « cut-off’ to the travel for either Oregon or Califor- 
nia. I therefore determined to make the examination myself, 
accompanied by Major Bridger, and to send forward the train to 
Salt Lake City by the Mormon road, under the command of Lieu- 
tenant Gunnison, whose health had become so far established as 
to enable him to resume his seat in the saddle. ‘The train left, ac- 
cordingly, on the 16th; but as we returned to this point by the 
same route the following year, I defer for the present any descrip- 
tion of it. I was myself detained until the 20th, by the absence 
of the partner of Major Bridger, who was on a trip to Salt Lake 
City, and without whose presence Major B. did not deem it 
prudent to leave the fort. As the examination was intended to be 
a mere reconnoissance, without instruments, a couple of men, with 
as many pack-mules, a little flour and bacon, with some ground 
coffee, and a blanket a piece, comprised all the preparation it was 
thought needful to make, taking care, however, that the little party 
should be well armed. 3 


MEDICINE BUTTE—BEAR RIVER. 17 


CHAPTER IV. 
FROM FORT BRIDGER TO GREAT SALT LAKE CITY. 


Monday, August 20.—We followed the Mormon road for several 
miles, and then took a « cut-off’ leading more to the north, crossing 
the dividing ridge between the waters of Muddy Fork, an affluent 
of- Green River, and those of Bear River, which falls into the 
Great Basin. We crossed the broad valley of Tar-Spring Creek, 
a tributary of Bear River, where the two roads join. The “cut- 
off’ has been abandoned on account of an almost impassable hill 
at the dividing ridge. ‘This, and another almost equally steep, are 
the only objections to this route, the rest of the way being excel- 
lent. Leaving the Mormon road at the crossing of Bear River, 
we followed down its valley six miles, as far as Medicine Butte, an 
elevated knob in the valley. This is a spot well known among the 
Indians, as that to which they were formerly in the habit of re- 
pairing to consult their oracles, or ««medicine-men,” who had lo- 
cated their ‘‘medicine lodge” in the vicinity of this little mountain. 
The route of a road to reach the north end of Salt Lake should 
pursue a nearly west course from Bridger’s Fort to this Butte, a 
distance of about thirty miles; the country, according to the 
representations of our guide, who has passed over it many times, 
being extremely favourable. 

At our encampment on Bear River, near this Butte, abundance 
of speckled trout were caught, resembling in all respects the brook 
trout of the States, except that the speckles are black instead of 
yellow. An ox, which had strayed from some unfortunate emi- 
grant, was found on the bank of the stream, in such capital condi- 
tion that he was shot for food, and such portions as we could not 
carry with us were most generously presented to a small encamp- 
ment of Shoshonee Indians, whose wigwams were erected among 
the bushes on the opposite side of the stream. It was curious to 
see how perfectly every portion of the animal was secured by 
them for food, even the paunch and entrails being thoroughly 
washed for that purpose. The squaws acted as the butchers, and 
displayed familiar acquaintance with the business, while the men 


78 FROM FORT BRIDGER TO GREAT SALT LAKE CITY. 


lounged about, leaning lazily upon their rifles, looking listlessly 
on, as if it were a matter in which they were in no manner inte- 
rested. They had quite a large number of horses and mules, and 
their encampment betokened comparative comfort and wealth. 

The bottom of Bear River is here four or five miles in breadth, 
and is partially overflowed in the spring: the snow lies upon it 
to the depth of four feet in the winter, which prevents the In- 
dians from occupying it during that season of the year, for which 
it would otherwise be well adapted. 

In leaving Fort Bridger, we passed over horizontal lias beds. 
About six miles to the north of the road, the country appeared to 
be much broken up, and not solely by the action of water. The 
Strata seemed dislocated and inclined, presenting much the same 
appearance as those near Laramie. Near this point, Frémont 
states that he found coal, which probably has been thrown up 
here. At Ogden’s Hole, on the eastern slope of the Wahsatch 
Mountains, we found the ranges of hills to be composed of the 
carboniferous strata, thrown up at a very considerable angle; and 
at Bear River, near our encampment of to-day, they were almost 
perpendicular, the later strata being deposited by their side in 
an almost horizontal position, with a very slight dip to the south- 
east. At this latter point, the older sandstones were cropping out 
at an angle of 35°; and on the opposite side of the river, the 
same strata were seen with a dip in the contrary direction, the 
valley being evidently an anticlinal axis. 

Wednesday, August 22.—Crossing the broad valley of Bear 
River diagonally, we forded that stream, and struck over a point of 
bluff into a valley, the course of which being too much to the south 
for our purpose, we passed over to another, and followed it to its 
head, where it opens upon a long ridge, running to the south-west. 
Instead of following the ridge, (which I afterward found should have 
been done,) we crossed over two more ridges into a third valley, 
in which was a small rapid stream running into Bear River. 
Fearful of getting too far south, I ascended the western bluff of 
this stream, in hopes of finding a valley or ridge the course of 
which would give us more westing ; but the country, in that direc- 
tion, was so much broken that we were forced still farther to the 
south, and struck upon the heads of Pumbar’s Creek, a tribu- 
tary of the Weber River, which latter discharges its waters into 
the Great Salt Lake. This valley, our guide insisted, would lead 
us in the right direction, and it was concluded to follow it down, 


PUMBAR’S CREEK—RED CHIMNEY FORK. 79 


which we did for about four miles, and bivouacked for the night. 
We continued down this valley until the middle of the following 
day, when, instead of the broad open appearance which it had at 
first presented, it soon began to contract, until it formed a cafion, 
with sides so steep that it was scarcely passable for mules. A 
blind Indian-trail wound along the hillside, at an elevation of 
several hundred feet above the stream, into which a single false 
step of our mules would instantly have precipitated us. It re- 
quired no small exertion of nerve to look down from this dizzy 
height into the yawning gulf beneath. After following the cafion 
some ten miles, we came to a broad valley coming into it from the 
left, which the guide declared headed in the ridge from which we 
had descended yesterday, and to the eastward of the route we had 
taken. As all prospect of a road by the valley of Pumbar’s 
Creek was now out of the question, I determined to follow up this 
valley and ascertain whether a route could not be obtained in that 
direction. ‘This was accordingly done, and we found it to be as 
the guide had stated. This branch of Pumbar’s Creek, which we 
called Red Chimney Fork, from the remarkable resemblance of 
one of the projections of the cliffs to that object, we found to 
have a very moderate descent from the ridge to its mouth, with 
plenty of room for a road, requiring but little labour to moult 
it a good one. The timber is small and consists of oak, black- 
jack, aspen, wild-cherry, service-berry, and box-elder of large size. 
In many places it is quite abundant. 

On Pumbar’s Creek, the hills were composed of strata of mar- 
ble and metamorphic sandstone, inclined at an angle of 80° to the 
north-east. Lower down, the horizontal strata were found lying 
by the side of these inclined rocks. On Red Chimney Fork, the 
strata were nearly horizontal, consisting principally of layers of 
red sandstone conglomerate, formed from metamorphic rocks with 
calcareous cement, and white sandstone with layers of conglo- 
merate interposed. Near its junction with Pumbar’s Creek, strata 
of slaty shales occurred, cropping out at an angle of 70°. 

Below the Red Chimney Fork, the valley of Pumbar’s Creek 
opens sufficiently to allow the passage of a road through the bot- 
tom; but, as its course was leading us from our intended direc- 
tion, we availed ourselves of a ravine, which, a mile below, comes 
into it from the north-west, and followed this up to its head, thus 
attaining the height of the general level of the country. The 
ascent is quite regular, but the road would have to be made all 


80 FROM FORT BRIDGER TO GREAT SALT LAKE CITY. 


the way up, and a considerable quantity of small cotton-wood tim- 
ber cut out. ‘The upper strata on this branch appeared to be 
nearly analogous to those met with on Red Chimney Fork. We 
followed this ridge or table in a north-west direction for several 
miles, when we became involved among numerous ravines which 
ran to the south, and were too deep and abrupt to be available. 
In order to avoid them, the trace must be thrown so much to the 
north, that even were a road practicable up to this point, it would 
be entirely too crooked; and great difficulty, moreover, would have © 
to be encountered in crossing the immense ravines which lay at 
the eastern base of the ranges bordering the Salt Lake. Some of 
these ravines run down into Ogden’s Creek, and others into Bear 
River below the point at which we crossed it. Time would not 
admit of my pursuing the examination farther in this direction. 
My train had left Fort Bridger several days before me, and would 
be awaiting my arrival at Great Salt Lake City to commence the 
survey which was the more immediate object of the expedition. 
I, therefore, although with the greatest reluctance, concluded to 
make the best of my way to the lake, passing through Ogden’s 
Hole, and thence crossing the high range dividing it from Salt 
Lake Valley, by a pass which the guide informed me existed there. 
We accordingly changed our course, and turning down a steep, nar- 
row ravine for wood and water, encamped. The night was very cold, 
and ice formed in the buckets nearly an inch thick. We constructed 
a semicircular barricade of brush to keep off the wind, and, by the 
aid of a large fire of pine-logs, passed the night very comfortably. 
The soil on the ridge passed over to-day, seemed formed princi- 
pally from red sandstone, and the boulders are primitive. The 
country is much better wooded, the timber being willow, aspen, 
and, in the ravines, tall firs and pines. The geranium was abund- 
ant: two or three yellow composite and asters were observed. 
Sunday, August 26.—Morning very cold. Ther. at sunrise, 
16°. Our provisions being nearly exhausted, I determined to go 
on for at least a part of the day, although contrary to my usual 
practice, this being the first Sabbath on which any travelling has 
been done since the party left the Missouri. After following some 
miles down the ravine upon which we had encamped, we struck 
~ upon an Indian lodge-trail, leading either to Cache Valley or to 
Ogden’s Hole. This we followed in nearly a southerly direction, 
crossing many deep hollows and very steep ridges, up which we 
had to scramble, leading our mules, (it being impossible to ride,) 


INDIAN SIGNAL-FIRES—OGDEN’S HOLE. 81 


until we struck upon the head of a broad, green, beautiful valley, 
with an even, gentle descent, which led us, in about three miles, 
down to Ogden’s Creek, just before it makes a caiion, previous to 
entering Ogden’s Hole. There we encamped for the remainder of 
the day, with abundance of excellent grass, wood, and water. The 
same alternations of red and white namdion appeared here as were ~ 
seen on the Red Chimney Fork. | : 

Just before descending into this valley, we had observed from 
the high ground, the smokes of numerous Indian signal fires, rising 
in several directions—an intimation that strangers had been dis- 
covered in their country. A strict watch was therefore maintained 
during the night, lest our animals should be stolen. Wild cherries 
were found in tolerable abundance, and the trail was strewn over 
with their smaller branches, thrown away by the Indians, who had. 
evidently passed only a day or two before, in considerable numbers. 

Monday, August 27.—We followed down Ogden’s Creek about 
a mile, when we found that the broad valley was shut up between 
two ranges of hills; or rather mountains, leaving a flat, low, level 
bottom, densely covered in places by willows, through which the 
stream meanders from side to side, for three miles, washing alter- 
nately the base of either range. After passing through this cafion, 
the ridge separated, and before us lay a most lovely, broad, open 
valley, somewhat in the shape of a crescent, about fifteen miles 
long, and from five to seven miles in width, hemmed in on all sides, 
especially on the south and west, by lofty hills and rocky moun- 
tains, upon the tops and sides of which the snow glistened in the 
rays of the morning sun. ‘The scene was cheering in the highest 
degree. The valley, rich and level, was covered with grass; 
springs broke out from the mountains in every direction, and the 
facilities for irrigation appeared to be very great. Ogden’s Creek, 
breaking through its barriers, flows in a crystal stream at the base 
of the mountains on the south, for rather more than half the 
length of the valley, when it forces a passage through the huge 
range which divides this « gem of the desert’’ from the Salt Lake 
Valley, by a caiion wild and almost impassable. On the north, a 
beautiful little brook, taking its rise in the elevated ground sepa- 
rating this from Cache Valley, washes the base of the western 
hills, and joins Ogden’s Creek just before it enters the cafion, after 
passing through which the latter discharges its waters into the 
Weber River, a tributary of the Great Salt Lake. Numerous 
bright little streams of pure running water were met with in abun- 


82 FROM FORT BRIDGER TO GREAT SALT LAKE CITY. 


dance, rendering this the most interesting and delightful spot we 
had seen during our long and monotonous journey. 

Rather more than half-way between the cafion of Ogden’s Creek 
and the north end of the valley, a pass is found by which a cross- 
ing of the mountain into the Salt Lake Valley can be effected. 
The ascent of the western side is, for the first four or five hundred 
yards, very abrupt and rocky, and would require a good deal of 
grading to render a road practicable; but after this, little or no 
labour would be necessary, except to cut away the brush, which, in 
places, is quite thick. The length of the pass is about three miles, 
and the height of the range through which it makes the cut, from 
eight hundred to a thousand feet above the valleys on each side. 
The valley of Ogden’s Creek, or Ogden’s Hole, (as places of this kind, 
in the nomenclature of this country, are called,) has long been the 
rendezvous of the North-west Company, on account of its fine range 
for stock in the winter, and has been the scene of many a merry 
réunion of the hardy trappers and traders of the mountains. Its 
streams were formerly full of beaver, but these have, I believe, entirely 
disappeared. Some few antelope were bounding over the green, but 
the appearance of fresh «Indian sign’’ accounted for their scarcity. 

During our ride through the valley we came suddenly on a 
party of eight or ten Indian women and girls, each with a 
basket on her back, gathering grass-seeds for their winter’s pro- 
vision. They were of the class of ‘root-diggers,” or, as the 
guide called them, «‘snake-diggers.”” The instant they discovered 
us, an immediate and precipitate flight took place, nor could all the 
remonstrances of the guide, who called loudly after them in their 
own language, induce them to halt for a single moment. Those 
who were too close to escape by running, hid themselves in the 
bushes and grass so effectually, that in less time than it has taken 
to narrate the circumstance, only two of them were to be seen. 
These were a couple of girls of twelve or thirteen years of age, 
who, with their baskets dangling at their backs, set off at their 
utmost speed for the mountains, and continued to run as long as 
we could see them, without stopping, or so much as turning their 
heads to look behind them. The whole party was entirely naked. 
After they had disappeared, we came near riding over two girls of 
sixteen or seventeen, who had “cached” behind a large fallen tree. 
They started up, gazed upon us for a moment, waved to us to con- 
tinue our journey, and then fled with a rapidity that soon carried 
them beyond our sight. 


WAHSATCH MOUNTAINS—SALT LAKE VALLEY. 83 


In the pass through which we entered Ogden’s Hole, the carboni- 
ferous rocks were again found, thrown up at an angle of 70° or 
80°, with a dip to the north-east. On the western side of the 
high rangé of hills which extended to the north-west and formed 
the eastern boundary of Ogden’s Hole, the edges of the strata 
cropped out as if a great fault had been formed at the point of 
elevation. No debris of primitive rock were discovered, nor was 
any observed in place during the whole journey from Bridger’s 
Fort. In the pass leading to Salt Lake, through the Wahsatch 
range, the rock were metamorphic. Some beautiful specimens of 
marble were observed, and also some white crystalline sandstones. 
The strata again appeared on the western side of the range, and 
were inclined to the north-east about 70°. The chain evidently 
was not formed on a central axis. No fossils were collected dur- 
ing this part of the journey, as we travelled rapidly, and the 
means of transporting them were necessarily limited. 

Descending the pass through dense thickets of small oak-trees, 
we caught the first glimpse of the Great Satt Laxg, the long- 
desired object of our search, and which it had cost us so many 
weary steps to reach. A gleam of sunlight, reflected by the water, 
and a few floating, misty clouds, were all, however, that we could 
see of this famous spot, and we had to repress our enthusiasm for 
some more favourable moment. I felt, nevertheless, no little grati- 
fication in having at length attained the point where our labours 
were to commence in earnest, and an impatient longing to enter 
upon that exploration to which our toils hitherto had been but 
preliminary. 

Emerging from the pass, we entered the valley of the Salt Lake, 
and descending some moderately high table-land, struck the road 
from the Mormon settlements to the lower ford of Bear River, 
whence, in two or three miles, we came to what was called Brown’s 
Settlement, and rode up to quite an extensive assemblage of log 
buildings, picketed, stockaded, and surrounded by out-buildings 
and cattle-yards, the whole affording evidence of comfort and 


abundance far greater than I had expected to see in so new a set- 


tlement. Upon requesting food and lodging for the night, we 
were told to our great surprise that we could not be accommodated, 
nor would the occupants sell us so much as an egg or a cup of milk, 
so that we were obliged to remount our horses; and we actually 
bivouacked under some willows, within a hundred yards of this in- 
hospitable dwelling, turning our animals loose, and guarding them 


84. ARRIVAL AT GREAT SALT LAKE CITY. 


all night, lest, in search of food, they should damage the crops of 
this surly Nabal. From a neighbouring plantation we procured 
what we needed ; otherwise we should have been obliged to go sup- 
perless to bed. I afterward learned that the proprietor had been 
a sort of commissary or quartermaster in Colonel Cook’s Mormon 
Battalion, in California, and had some reason to expect and to 
dread a visit from the civil officers of the United States, on ac- 
count of certain unsettled public accounts; and that he had 
actually mistaken us for some such functionaries. Subsequent 
acts of a similar nature, however, fully evinced the ungracious 
character of the man, strongly contrasted as it was with the frank 
and generous hospitality we ever received at the hands of the 
whole Mormon community. 

The following day we reached the City of the Great Salt Lake, 
and found that the train had arrived safely on the 23d, and was 
now encamped near the Warm Springs on the outskirts of the city, — 
awaiting my coming. 

The result of the reconnoissance we had thus completed was 
such as to satisfy me that a good road can be obtained from Fort 
Bridger to the head of the Salt Lake; although I incline to the 
opinion that it should pass farther north than the route taken by 
me, entering the southern end of Cache Valley, probably by Black- 
smith’s Fork, and leaving it by the cafion formed by Bear River 
in making its way from that valley into the lake basin. A more 
minute examination than the pressure of my other duties allowed 
me time to make will, I think, result in the confirmation of this 
view and the ultimate establishment of this road. Should such 
prove to be the case, it will, in addition to shortening the distance, 
open to the emigration, at the season they would reach it, the inex- 
haustible resources of Cache Valley, where wood, water, abundance 
of fish, and the finest range imaginable for any number of cattle, 
offer advantages for recruiting and rest possessed by no other point 
that I have seen on either side of the mountains. 

Before reaching Great Salt Lake City, I had heard from various 
sources that much uneasiness was felt by the Mormon community 
at my anticipated coming among them. I was told that they would 
never permit any survey of their country to be made; while it was 
darkly hinted that if I persevered in attempting to carry it on, 
my life would scarce be safe. Utterly disregarding, indeed giving 
not the least credence to these insinuations, I at once called 
upon BrigHam Young, the president of the Mormon church and 


J Broadway NY 


i 


h 


erman jit 


Ack 


Pi! ViiW OPseGRhAT SAG LAKE VALLEY FROM AMOUNTAIN FASS 


RUMOURED HOSTILITY OF THE MORMON AUTHORITIES. 85 


the governor of the commonwealth, stated to him what I had 
heard, explained to him the views of the Government in directing 
an exploration and survey of the lake, assuring him that these 
were the sole objects of the expedition. He replied, that he did 
not hesitate to say that both he and the people over whom he 
presided had been very much disturbed and surprised that the 
Government should send out a party into their country so soon 
after they had made their settlement; that he had heard of the 
expedition from time to time, since its outset from Fort Leaven- 
worth; and that the whole community were extremely anxious as 
to what could be the design of the Government in such a move- 
ment. It appeared, too, that their alarm had been increased by 
the indiscreet and totally unauthorized boasting of an attaché of 
General Wilson, the newly-appointed Indian Agent for California, 
whose train on its way thither had reached the city a few days be- 
fore I myself arrived. This person, as I understood, had declared 
openly that General Wilson had come clothed with authority from 
the President of the United States to expel the Mormons from the 
lands which they occupied, and that he would do so if he thought 
proper. The Mormons very naturally supposed from such a de- 
claration that there must be some understanding or connection 
between General Wilson and myself; and that the arrival of the 
two parties so nearly together was the result of a concerted and 
combined movement for the ulterior purpose of breaking up and 
destroying their colony. The impression was that a survey was to 
be made of their country in the same manner that other public 
lands are surveyed, for the purpose of dividing it into townships 
and sections, and of thus establishing and recording the claims of 
the Government to it, and thereby anticipating any claim the Mor- 
mons might set up from their previous occupation. However un- 
reasonable such a suspicion may be considered, yet it must be 
remembered that these people are exasperated and rendered 
almost desperate by the wrongs and persecutions they had pre- 
viously suffered in Illinois and Missouri; that they had left the 
confines of civilization and fled to these far distant wilds, that they 
might enjoy undisturbed the religious liberty which had been prac- 
tically denied them; and that now they supposed themselves to 
be followed up by the General Government with the view of 
driving them out from even this solitary spot, where they had 
hoped they should at length be permitted to set up their habitation 
- In peace. 


_86 SALT LAKE CITY—BRIGHAM YOUNG. 


Upon all these points I undeceived Governor Young to his entire 
satisfaction. I was induced to pursue this conciliatory course, not 
only in justice to the Government, but also because I knew, from 
the peculiar organization of, this singular community, that, unless 
the «« President” was fully satisfied that no evil was intended to his 
people, it would be useless for me to attempt to carry out my in- 
structions. He was not only civil governor, but the president of 
the whole Church of Latter-Day Saints upon the earth, their pro- 
phet and their priest, receiving, as they all firmly believed, direct 
revelations of the Divine will, which, according to their creed, form 
the law of the church. He is, consequently, profoundly revered 
by all, and possesses unbounded influence and almost unlimited 
power. I did not anticipate open resistance; but I was fully 
aware that if the president continued to view the expedition with 
distrust, nothing could be more natural than that every possible 
obstruction should be thrown in our way by a “ masterly inactivity.” 
Provisions would not be furnished; information would not be af- 
forded; labour could not be procured; and no means would be 
left untried, short of open opposition, to prevent the success of a 
measure by them deemed fatal to their interests and safety. So 
soon, however, as the true object of the expedition was fully under- 
stood, the president laid the subject-matter before the council 
called for the purpose, and I was informed, as the result of their 
deliberations, that the authorities were much pleased that the ex- 
ploration was to be made; that they had themselves contemplated 
something of the kind, but did not yet feel able to incur the ex- 
pense; but that any assistance they could render to facilitate our 
operations would be most cheerfully furnished to the extent of their 
ability. This pledge, thus heartily given, was as faithfully re- 
deemed ; and it gives me pleasure here to acknowledge the warm 
interest manifested and efficient aid rendered, as well by the 
president as by all the leading men of the community, both in our 
personal welfare and in the successful prosecution of the work. 


LOWER FORD OF BEAR RIVER. i 8T 


CHAPTER V. 


EXPLORATION OF A ROUTE FROM GREAT SALT LAKE CITY TO FORT 
HALL, AND RECONNOISSANCE OF CACHE VALLEY. 


MattERS being thus satisfactorily adjusted, as the provisions 
which had been laid in at the beginning of the journey were nearly 
exhausted, I left the city on the 12th of September, with teams 
and pack-mules, for Fort Hall, to procure the supplies for the party 
which had been forwarded to that post by the supply-train at- 
tached to Colonel Loring’s command; and at the same time to 
carry out that portion of my instructions which directed me to ex- 
plore a route for a road from the head of Salt Lake to Fort Hall. 
The main party was left under the command of Lieutenant Gun- 
nison, with instructions to commence the survey upon a basis already 
laid down. I was accompanied on this trip by Mr. John Owen, 
the sutler of the regiment of Mounted Rifles, and Mr. T. Pomeroy, 
a merchant from St. Louis, on his way to California. Our route, 
as far as the crossing of Bear River, near the head of the lake, was 
that usually pursued by emigrants passing through Salt Lake City 
to California. It skirts the eastern shore of the lake throughout 
its whole length, from north to south, as far as the ford, where 
the road turns off to the west. As the country passed over in this 
part of the journey is embraced within the limits of the survey, it 
requires, at present, no farther notice. 


From the crossing, the emigrant road pursues a W.N. W. course, 


until it intersects that from Fort Hall. 'The ford of Bear River 
at this point is not very good. The banks are high and steep on 
both sides, and the stream, which is about two hundred and fifty 
feet wide, is quite rapid. The bottom is a hard, firm gravel. In 
the spring and early part of summer, the waters are too high to 
admit of fording, and temporary ferries become necessary. Leavy- 
ing the emigrant road at this point, our route may be described, 
generally, as following up the Malade (called by Frémont the 
Roseaux) to its head; thence crossing a high dividing ridge, we 
fall upon the heads of the Pannack, a tributary of the Port Neuf, 
(which latter is an affluent of Lewis’s Fork of the Columbia,) and 


88 FROM GREAT SALT LAKE CITY TO FORT HALL. 


following down its valley to within five miles of Fort Hall, we cross 
the Port Neuf, and passing over a wide level plain, reach that cele- 
brated trading-post. But this line is deserving of notice rather 
more in detail. About two miles above the ford, Bear River, in 
emerging from Cache Valley, breaks through the chain forming 
the eastern boundary of the valley of Salt Lake. The range, 
which here sinks quite suddenly, for a short distance to the 
south of the cafion or gate through which the river has forced its 
passage, consists of low, rounded hills, which present no trace of 
rock on the surface. ‘The river indeed appears to cut through 
rock, but an opportunity did not occur to ascertain this by actual 
observation. After crossing and following up its right bank for 
two and a-half miles, we left the river, and struck into a broad and 
beautiful valley, formed by the Roseaux, or Malade, which, flowing 
from the north, discharges itself into Bear River some miles below 
the ford. The valley is five or six miles wide, and its western 
boundary is formed by a chain of high, rounded hills, being the 
continuation of a lofty rocky promontory, projecting into the north 
end of the Lake. The eastern boundary of this valley is formed 
by the continuation, in a northern direction, of the Wahsatch range, 
which divides it from the Bear River and Cache Valley. Ascend- 
ing the valley, these mountains rise to a considerable height, the 
strata dipping to the north-east, and the direction of the chain 
inclining to the west. The valley of the Malade is extremely 
level, free from underbrush, with very little artemisia, and affords 
ground for an excellent wagon-road. Water to-day was found in 
quantities sufficient for the animals, at points conveniently distri- 
buted, and grass was abundant. Several fine springs were passed, 
in which the water was cold and clear. Continuing up the valley 
until four o’clock, we came to a superb little stream, coming out 
of the eastern mountain, running with great swiftness over a bed 
_ of breccia, and discharging a large quantity of clear, cold water. 
“The fall was great and the quantity of water ample for the irri- 
gation of a very large farm, for which the lay of the land offers 
great facilities. Here we encamped, with plenty of fine grass. 
Distance from the city, one hundred and three miles; and from 
Bear River ford, twenty-four and a-half. 

Thursday, September 20.—Our march to-day was only eleven 
miles, owing to the necessity of making a road across a small 
stream with steep banks, which comes through a depression in the 
eastern hills, through which a road from Sheep Rock, near the 


VALLEY OF THE MALADE. 89 


Soda Springs, had been partially explored by Mr. Owen, whose 
‘wagons had come through it some two weeks since, on their way to 
Salt Lake City. He describes the country as rough and rolling, 
with several high and steep ridges to be crossed. The road to-day 
has been level, with wood and water abundant. Encamped on the 
left bank of the Malade, here six feet wide and two feet deep. 

Friday, September 21.—Following up the left bank of the Ma- 
lade for four miles, we crossed a small swift fork coming in from 
the north-east, affording abundance of water for irrigating a con- 
siderable extent of its valley on each side. The valley of the Ma- 
lade is becoming gradually narrower and the hills lower. Crossing 
another fork from the east, we strike upon «« Hedspeth’s Cut-off,”’ 
which leads from Sheep Rock, near the Soda Springs, to the Mor- 
mon road at Goose Creek. ssepicinue one hundred and twenty-five 
and a-half miles. 

The valley of the Malade seems to be formed principally of 
whitish clay, in which, however, no good section was found, so that 
it is uncertain whether it presents any stratification. Occasionally 
ridges of limestone and conglomerate push out from the side of the 
mountains; and in one instance the river was found flowing over a 
bed of breccia. The rock on the west side of the valley consisted 
of dark compact limestone, with a dip of 20° to the south-west. 
Shortly after reaching the Cut-off, a belt of high hills extended 
across the valley from east to west, composed of dark limestone 
containing a considerable number of fossils. These hills we 
ascended by one of the handsomest passes I had seen in the 
country. The inclination in no instance exceeds 5°; the soil is 
hard and porous; the natural road perfectly drained. The length 
of the pass is four miles, from the summit of which we descended 
to the east fork of the Malade, upon which we encamped, with in- 
tensely cold, pure water, willows for firewood, and good grass. 
In the pass some specimens of obsidian and volcanic debris were 
collected, evidently of secondary formation, and not conformable 
with the limestone ridges. Trachytic rock was also found on the 
side of the stream, forming a considerable hill, and overlaid by 
dark limestone. | 

Saturday, September 22.—Directly after starting, crossed the 
east fork of the Malade, and still following the Cut-off, the track 
of which is hard and well beaten, we ascended another pass, in a 
north direction, very similar in its character to that we came up 
yesterday. From the top of this pass, which is the dividing ridge 


90 FROM GREAT SALT LAKE CITY TO FORT HALL. 


between the Malade and the waters of the Port Neuf, the road 
descends by a gentle slope to the dry bed of a small stream, which 
forms a narrow gorge; emerging from which, and proceeding north, 
we descended to a small stream forming one of the heads of a 
branch of the Port Neuf. It flows at the foot of a spur of the 
range of hills which constitute the dividing ridge between it and 
the Pannack, (another affluent of the Port Neuf,) and rises in a 
broad valley lying to the westward of the road. It is bounded on 
the west by a high range of hills extending to the southward, and 
in that direction forming the « divide’ between the waters of the 
Malade and those of the Pannack. 

The secondary or lower hills in this valley seem to be composed 
principally of white clay containing volcanic debris. Crossing the 
stream, we left the cut-off altogether, and turning to the left, 
crossed over this ridge, which, where we crossed it, is very high 
and steep, and a mile and a-half in width. Descending its western 
slope, we struck upon the heads of one of the main forks of the 
Pannack, down which an excellent road can be obtained without 
difficulty, the descent being moderate and the ground generally 
level. 

Descending the valley of this stream, we encamped on its right 
bank with plenty of grass, fine cool water, and a profusion of 
willows for fuel. Day’s march, fifteen miles. 

At the dividing ridge between the waters of the Port Neuf and 
the Malade, the direction of the stratification has evidently changed. 
Near the south end of the pass, an escarpment of dark limestone 
is seen on the eastern side, lying on and conformable with layers 
of feldspathic rock. A short distance farther on, the same rock is 
again seen, overlaying the dark limestone, and with a dip of about 
50° to the north-east. From this point the centre of elevation, 
consisting evidently of this hypogene rock, appears to take a di- 
rection to the north-west, striking the chain of hills contmued 
from the west side of the valley of the Malade. It is plainly to 
be seen that this has been a region of great disturbance, which 
did not cease until a period subsequent to the deposition of the 
secondary rocks that repose on the limestones, although not con- 
formable with them. Passing this ridge, several high conical hills 
were observed on the right, which seemed to be formed of second- 
ary rocks, the stratification of which was apparently much more 
horizontal than that of the limestones. The dividing ridge be- 
tween the Port Neuf and the Pannack is composed of dark lime- 


VALLEY OF THE PANNACK. 91 


stones, altered shales, and veins of the same feldspathic rock noticed 
in the pass. The strata were inclined east by north, at an angle 
of 70°. The ridge seems to run a little west of north, until it 
disappears in the valley of the Snake River. Upon the summit 
of this «+ divide” was found what was at first thought to be altered 
coal, but upon farther examination it appeared to be an aluminous 
rock, containing but a small trace of carbonate of lime. Its 
colour was black, hardness greater than that of feldspar, and the 
form a rhombic prism. The limestone was crystalline, and con- 
tained numerous specimens of shells and corals, but in so altered 
a state that it was impossible to determine them. 

The length of the fork of the Pannack which we descended is 
sixteen miles. It pursues a westerly direction, until it joins the 
main stream, which latter flows from the southward, through what 
appeared to be a well-defined valley. The ground for a road is ex- 
cellent, with only one or two exceptions, which are not of a serious 
character. 

On descending the dividing ridge in which it heads, the rocks 
were hidden by a black, rich soil; occasional boulders of granite 
were seen on the surface, but no section could be obtained until 
we came to a gorge about five miles down the valley. Here the 
river cuts through a much lower ridge of hills, composed of lime- 
stone, dipping to the east, at an angle of about 63°: below this the 
stream has cut its bed through secondary hills formed of argilla- 
ceous sandstone and clay, both of which are white, and mixed with 
pieces of obsidian and occasional boulders of serpentine: still 
lower down the valley, a section in a ravine to the right of the 
road, discovered some rocks which might almost be considered 
cretaceous ; alternating with white argillaceous sandstone, they con- 
tamed a considerable quantity of organic remains, principally 
coral, but so much altered by heat that it was impossible to dcter- 
mine them with precision. The dip of these strata was about 40° 
north-east. The beds were covered by the remains of disaggre- 
gated conglomerate, composed principally of porphyry and granite. 
Proceeding down the stream, metamorphic sandstones, crystallized 
almost to the whiteness of white quartz, were found, forming es- 
carpments of the lower hills; a short distance below iki, point, a 
ridge of hills, composed of limestone, shales, and red sandstone, 
extended across the valley; they were all much inclined, with a 
dip to the east. At this point, where the river cuts a passage 
through this chain, a mass of feldspathic rock was seen. The dip 


92 FROM GREAT SALT LAKE CITY TO FORT HALL. 


of these strata and also of the crystalline sandstone was about 


E.8.E., at an angle of from 60° to 70°. 
carn the junction of the two forks, the valley changes its direc- 


tion to the N. N. W., which it maintains until it merges in that of | 


the Port Neuf, a sraieines of eighteen miles; it becomes broader, the 
bottoms are high, hard, very level, and entirely covered with arte- 
misia. Coarse red metamorphic sandstone was found on the side 
of the valley at this point, with a considerable dip to the north-east. 
Clayey shales also occurred; and, from the appearance of the soil, 
a great deal of argillaceous rock, must-exist in the vicinity. Five 
miles below the forks, a remarkable isolated hill stands on the 
western side of the valley, called by the traders the «“ Windmill 
Rock.” Here a dike of trap was met with, running north-east and 
south-west, forming the axis of a chain extending across the valley, 
and of which the isolated hill seemed to form a part. The dike 
constituted the summit of a high hill on the east side of the river: 
on the west side, the same rock was found, but not sohigh. Meta- 
morphic sandstone (red) was found overlaying the trap, and what 
appeared to be porous basalt was found in considerable abundance: 
no section of the stratification of the sandstone could be obtained. 

Beyond this point, the valley of the Pannack gradually sinks 
down into that of the Snake River. The hills that enclose it are 
not high, and seem formed almost wholly of white clay; at least, 
this was the only soil exposed, even in some very deep ravines. 
The same character of soil is found on the whole country this sons 
of Snake River. 

Twelve miles from the forks, we leave the Pannack, which there 
makes a curve to the westward, around the point of a ridge which 
is quite low, and the ascent gentle and regular. Upon reaching 
the level of the table-land, nothing was to be seen, as far as the 
eye could reach, but the eternal artemisia, which had taken com- 


plete possession of this barren, dreary waste, and extended quite 


to the Port Neuf. Upon reaching this stream, we struck upon 
the emigrant road by Fort Hall to California; and descending a 
bluff, or rather a cliff, two hundred feet in height, and composed 
entirely of argillaceous soil, we crossed the Port Neuf and en- 
tered the valley of the Columbia. From the top of the bluff, an 
extensive level plain, clothed with grass, is spread out before us, 
like a beautiful picture; while the fringe of heavy timber, stretch- 
ing far away to the north and west, indicates the position of 
Lewis’s Fork of the great river of the West. Five miles to the 


VALLEY OF LEWIS FORK OF THE COLUMBIA. 93 


north, Fort Hall, with its whitewashed walls, is plainly in view. 
The «Three Buttes” rise in the distance, while the Port Neuf, 
with its bright, sparkling waters, flows at our feet. The scene was 
one of surpassing beauty, and richly repaid us for our dreary ride 
across the desert plain of sage. 

The Port Neuf, where we forded it, is a fine, clear, bold stream, 
one hundred yards wide and three feet deep, with a moderately 
rapid current and pebbly bottom. ‘The plain between it and Snake 
River presents a level bottom, formed principally of decomposed 
vegetable mould, reposing on sandy loam and gravel. Numerous 
springs of cold, pellucid water, abounding in speckled trout of de- 
licious flavour, break out in every direction, giving rise to many 
little streams, which rapidly increase in size and afford great 
facilities for irrigation as well as for the construction of mills. 
Passing over this delightful plain, we left Fort Hall on our left, 
and five miles beyond it terminated our journey, at Cantonment 
Loring, our point of destination. 

I was most courteously received by Lieutenant-Colonel An- 
drew Porter and the officers of his command, which consisted 
of two companies of the regiment of Mounted Rifles, left here by 
Colonel Loring on his way to Oregon, with the view of selecting 
a permanent post for the protection of the vast emigration across 
the continent. The troops were quartered in tents, but were 
busily engaged in the erection of quarters, of a more substantial 
character, for the winter. 

The result of this exploration has been to demonstrate the entire 
practicability of obtaining an excellent wagon-road from Fort 
fall to the Mormon settlement upon the Great Salt Lake. With 
the exception of the ridge dividing the waters of the Pannack 
from those of another affluent of the Port Neuf, the line traced 
is unexceptionable, and offers facilities for the best natural road I 
ever saw. Although when we passed there had not been even a track 
broken, so favourable is the surface of the country that I trans- 
ported my provisions over it without the slightest difficulty, load- 
ing my wagons with not less than thirty-five hundred pounds each. 
The ridge referred to can, by a little labour, be rendered easy to 
cross; and even as it is, offers but little obstruction. In seasons 
of high water, Bear River and the Port Neuf would have to be 
crossed by ferries; or, should the travel ever demand it, timber for 
the construction of bridges could be obtained in the vicinity of 
both localities. 


94 RECONNOISSANCE OF CACHE VALLEY. 


The supply-train from Fort Leavenworth, with my provisions, 
had not arrived at the post, as I expected, and I was consequently 
detained until the 6th of October, when, having obtained them, I set 
out onmyreturn. ‘The frank and generous hospitality we received 
during our stay at the post demands a grateful acknowledgment. 

Returning, I was accompanied by Colonel Porter, with a small 
escort, as far as the crossing of Bear River. He was desirous 
that we should make conjointly a reconnoissance of Cache Valley, 
to ascertain its fitness for the location there of a permanent mili- 
tary post. | 

Following the same route which I had taken when coming up, 
we arrived at Bear River on the evening of the 11th, and en- 
camped. The examination of Cache Valley occupied several days. 
Crossing over the range of low, rounded hills through which Bear 
River has cut a passage, we entered this beautiful and picturesque 
valley, which was then covered with a profusion of rich green 
grass, and adorned and diversified by numerous clumps of willows. 
Our attempt to cross it directly was frustrated by meeting with a 
deep, quiet stream, called the Muddy, which rises in the hills 
dividing the southern end of the valley from Ogden’s Hole, and 
winds through the tall grass without banks, until it discharges its 
waters into Bear River, just before that stream enters the valley 
of the Salt Lake. We were in consequence driven some eight 
miles to the south, and effected our crossing where the valley is 
full of swampy springs, affording abundance of good sweet water, 
and excellent grass. Speckled trout of large size abounded in the 
streams. After crossing the Muddy, we skirted the eastern side 
of the valley for thirty-five miles in a northerly direction, crossing 
successively Blacksmith’s Fork, Logan’s Fork, High Fork, Gros 
Bois, and Rush Creek, all tributaries of Bear River, which latter 
stream traverses the valley from the north, until it breaks through 
the range forming its western boundary and enters that of the 
lake. The streams on the east side take their rise m a heavy 
range running to the north and constituting the eastern limit of 
the valley, which has an average width of about ten miles. The 
cafions which they form before leaving the mountains abound in 
timber, consisting principally of cotton-wood, with some maple. 
They afford desirable facilities for irrigation, presenting at the same 
time advantageous sites for the erection of mills. These ravines 
abound in fine timber in quantities sufficient for fuel and building 
purposes. 


CACHE VALLEY. 95 


As the object of the reconnoissance was principally to ascertain 
what were the capabilities of this valley to afford sustenance for a 
military post, if established in its vicinity, the examination was but 
a general one, and was not directed to the selection of any particu- 
lar portion of it for sucha purpose. At the time the reconnoissance 
was made, all the information that could be obtained from the oldest 
mountain-men, induced both Colonel Porter and myself to believe 
that it was one of the most eligible spots in the whole country for 
wintering stock. It had been a rendezvous for the American Fur 
Company for many years, and stock had been wintered there by 
them with great advantage. The snow was seldom deep, and the 
cattle not only retained their flesh, but grew fat during the winter. 
So flattering were the appearances, and so great the advantages 
offered by this lovely valley, that nearly the whole number of cattle 
and mules belonging to the cantonment were, upon the return of 
Colonel Porter to that post, driven down here under the care of a 
proper guard, to be wintered. The season, however, proved un- 
usually severe; the snow fell in the valley to a depth unprecedented ; 
and more than one-half of the herd, in which were included some 
of my own animals, perished in consequence. The fact of the lia- 
bility of the valley to a similar occurrence in future will doubtless 
have its due influence in finally deciding upon its eligibility as the 
best site for a post in the vicinity of Salt Lake. 

The soil of the valley is very rich, being principally alluvial, with 
a great deal of vegetable mould. Facilities for irrigation are very 
great, and water could be commanded to a large extent for farming 
purposes. Any amount of hay might be cut without in the least in- 
terfering with the range for cattle. The only objection to this, as a 
most desirable spot for settlement, is the danger from snow; and 
even this might be in a great degree obviated, by the erection of 
suitable sheds for protection of the stock during the more severe 
portions of the seasons. These seldom last beyond a few weeks. 

Should the road to which I have already adverted be established 
from Fort Bridger, through the valley of Blacksmith’s Fork, it 
would at once attract to it the travel to Oregon and California; a 
fact which would have its due weight in the selection of a site for a 
military post for the protection of this part of the country. 

The advance of the season precluded the making of much geo- 
logical examination beyond the immediate vicinity of the route 
travelled, which led through the valley at the base of the ranges. 
The only rocks met with were those composing the lower hills, 


96 CACHE VALLEY. 


which consisted principally of conglomerates overlaying some argil- 
laceous sandstones and beds of white and red clay. The conglome- 
rates on the lower hills were formed principally of dark limestone, 
much worn. On the higher benches, large boulders: of feldspar 
were found. Albite and serpentine also occurred, and metamor- 
phic sandstones, some of which were very beautifully veined, as if 
the strata had been disturbed before they had hardened. 

Returning to the southern end of the valley, we again struck the 
Muddy, and followed it up to where it forks, amid the hills forming 
the « divide” from Ogden’s Hole. The eastern fork makes an im- 
passable cafion, but we followed up the west fork about four miles, 
whence we crossed the Wahsatch range, and descended into a beau- 
tiful, level, circular valley, about a mile in diameter, hemmed in by 
an amphitheatre of lofty and steep mountains. Several fine springs 
head in this singular little hollow, which uniting and emerging on 
the south-west side, form the head of Box-elder Creek, a tributary 
of the Salt Lake. The pass or gorge through which this little 
stream rushes down the mountain to the plains below is steep, 
rugged, and very narrow, being in places scarcely passable for 
mules. I had hoped it would afford a passage over the range for 
wagons, but this I soon found to be impracticable. Descending 
this wild pass for about two miles, we reached the lake valley, and 
repaired to our camp on Bear River. 

In crossing the Wahsatch range at this point, the lower hills on 
the eastern side were composed of broken conglomerate. Large 
boulders of serpentine were met with on the surface, and also al- 
tered sandstones and limestones. Ascending from Cache Valley, 
the dark limestones were found cropping out, but the surface was 
so completely covered with vegetable soil that no section could be 
obtained. The limestones seemed to form the summits of the highest 
elevation of the range, but as we passed through the deep gorge 
of Box-elder Creek, this could not be positively ascertained. No 
trap was observed, but large boulders of granite were seen in the 


sides of the pass. ‘The rocks had been so much worn, and the. 


surface was so covered by fallen masses, that no section of the 
stratification was visible. 


PREPARATION FOR EXPLORATION. 97 


CHAPTER VI. 


RECONNOISSANCE OF THE DESERTS AROUND THE WESTERN SHORES OF 
THE GREAT SALT LAKE. 


Tux two following days were busily occupied in making prepa- 
_ ration for an exploration around the western shore of the lake, 
which I desired to complete previous to entering upon a more 
minute survey of its waters. The expedition was deemed neces- 
sary, to enable me as well to ascertain its general features as to 
gain some knowledge of the means and appliances necessary to 
carry on the survey with safety and expedition. By the old moun- 
tain-men such a reconnoissance was considered not only hazardous 
in the highest degree, but absolutely impracticable, especially at 
so late a season of the year. In this opinion they were confirmed 
by the representations of the Indians, who represented water to 
be extremely scarce and the country destitute of game. It was 
affirmed that the contemplated circuit had been repeatedly at- 
tempted by old and experienced trappers, in search of beaver, but 
always without success; the adventurers being invariably obliged 
to return with the loss of most of their animals. This was dis- 
couraging ; but im addition to these objections, it was known that 
mortal offence had been taken by the Shoshonee or Snake Indians, 
{through whose country we would be obliged to pass,) arising from 
a gross and wanton outrage which had been a short time before 
inflicted upon them by a company of unprincipled emigrants, by 
whom their women had been most brutally treated, and their 
friends murdered while attempting to defend them. Fears were 
entertaimed lest, in the wilds of this inhospitable region, where 
foot of white man had never trod, we should fall a sacrifice to the 
just vengeance of those infuriated savages. 

Having determined, however, that the examination was necessary 
to enable me to carry out the instructions of the department, I re- 
solved to proceed, or at least to make the attempt. My prepara- 
tions were simply to kill a beef and dry as much of the flesh as we 
could carry upon our pack-mules; since it would have been unsafe to 
risk the existence of the party upon the chances of killing game by 

7 


f 


98 RECONNOISSANCE AROUND GREAT SALT LAKE. 


the way. We also provided ourselves with three India-rubber bags, 
of the capacity of five gallons each, and a small keg, for transport- 
ing water across the desert; some sacks of flour, a small tent with- 
out poles, a tent-fly, and a blanket to each man. In addition, each 
person carried a few pounds of fresh beef attached to his saddle, 
which might be used before resorting to our store of dried meat. 
Thus equipped and well armed, we set out on the afternoon of the 
19th of October, the little party numbering five men and sixteen 
mules. The provision-train for the surveying party was sent forward 
by the emigrant road, on the east side of the lake, under Lieutenant 
Howland, with orders to report to Lieutenant Gunnison at Salt Lake 
City. Colonel Porter had left us the day previous, on his return to 
Cantonment Loving. 

From the ford of Bear River we followed the emigrant road 
westward for about four miles, which brought us to the Malade 
River. The crossing here was very difficult, and we found it impos- 
sible to get our animals over with their packs on, because of the depth 
of water; they were accordingly unloaded and dragged or driven 
over, one at a time, and some of them came near being swamped 
in the soft, sticky mud cemposing the bottom. The men were 
obliged to strip, and carry the packs over on their heads, the lighter 
articles being thrown across. Wood was very scarce: we had but 
artemisia-bushes and a few charred sticks found amid the ashes 
of the extinguished fires left by the emigrants. These were, how- 
ever, sufficient for cooking purposes. Grass there was none; and 
we began already to have some foretaste of the hardships to which 
our poor animals were about to be expesed. The night was cold; 
thermometer 22°. As wood could not be obtained even for tent- 
poles, we contented ourselves with stretching our weary bodies 
upon the ground, and, wrapped in our blankets, slept soundly till 
the morning. 

The bottoms of Bear River and the Malade are composed of 
white clay, in which no trace of organic remains was discovered. 
The current of the Malade is here slow, and the water brackish 
and nauseous. 

Saturday, October 20.—Ther. at daylight, 26°. Contimued on 
the emigrant road about four miles, when we left it and turned 
more to the southward, with the intention of doubling a lofty 
promontory that puts into the lake from the north, and forms the 
western boundary of the Malade valley. In about a mile we came 
upon three or four beautiful springs of clear, bright water: they 


_ MUD PLAINS—SHOSHONEE INDIANS. 99 


were gushing out from a rocky point, (of dark limestone and coarse 
argillaceous sandstone, with a dip of about 20° to the east,) and 

~ unite to form a branch which runs southward some miles, and then 
sinks in the sand, before reaching the lake. The water was, how- 
ever, warm, brackish, and entirely unfit for drinking. Following 
down this stream for several miles, we struck on a succession of 
bare, level plains, composed of white clay and mud, with occasion- 
ally pieces of limestone and obsidian scattered on the surface. 
These dreary plains were occasionally separated from each other, 
by patches of salt grass and scattered clumps of artemisia. They 
had apparently formed, at some remote period, a part of the lake, 
and it is probable were partially covered during the freshet months. 
Some portions of the ground were still moist, and too soft to admit 
the passage of our mules without danger of miring. Where dry, 
the surface was hard and smooth. 

In the afternoon, as I felt apprehensive of being overtaken by 
night without water for our animals, we turned more to the west- 
ward, and directed our steps toward the promontory range pre- 
viously mentioned. Before reaching it, however, we came upon 
a small stream, fifteen feet wide and a foot deep, but it was quite 
salt, and almost unfit to drink; yet, as we had no prospect of find- 
ing better, we were fain:to bivouac on its bank for the night. 
Artemisia was abundant, furnishing plenty of excellent fuel, 
although it reminded me somewhat of the scriptural phrase, 
‘crackling of thorns under a pot,” so constantly did the fire 
require replenishing. Day’s travel, twenty-two miles. 

Sunday, October 21.—Ther. at daybreak, 27°. There being 

“neither grass nor water at this point, we left it early, and made in 
a south-west direction for the foot of the mountain, travelling over 
a hard, even surface of dry mud, as level as a floor and without a 
particle of vegetation of any kind. Before reaching the base of 
the hills, we descried some Indians at a distance, who, as soon as 
they discovered us, commenced a most rapid and precipitate flight. 
As they were on foot, I despatched the guide after them at full 
gallop to bring them to a parley, being desirous of obtaining from 
them some information, and if possible, to prevail upon some of 
them to act as guides through the unknown regions before us. 
The man overtook them at the foot of the mountain, when several 
of them, finding their retreat about to be cut off, halted, and ad- 
vanced upon him with their guns presented, but were restrained 
from firing by an old Indian, who seemed to act as their chief. 


100 RECONNOISSANCE AROUND GREAT SALT LAKE. 


As soon as they perceived the rest of our party moving toward 
them from the plain, the whole band, consisting of some six or 
eight men and half a dozen squaws, retreated incontinently up the 
mountain, and in a few minutes totally disappeared, nor did we 
see them again. As we continued to advance, we passed through 
their encampment, which they had abandoned in such haste that 
they left every thing as it was at the moment of their flight— 
the kettle was boiling over the fire, and a good gun rested against 
a bush. We left all untouched, and did not even dismount, as we 
knew they were watching us from behind the rocks, and I was de- 
sirous of convincing them of our peaceable disposition. 

Following down the eastern base of the promontory for about two 
miles, we éncamped on a small spring-branch, coming down from the 
mountains, furnishing very tolerable water and plenty of grass— 
refreshment most welcome to our jaded and famished animals, 
which had not had a full meal for nearly two days. At the In- 
dian camp there was a spring, but the water, although abundant, 
was salt and unfit for use. ‘Temperature of the spring, 84°. The 
mountain or main promontory seemed to be composed of limestone, 
altered shales, and sandstones: it rises from fifteen hundred to 
two thousand feet. 

Monday, October 22.—Ther. at sunrise 25°. Morning clear 
and calm. The Salt Lake, which lay about half a mile to the 
eastward, was covered by immense flocks of wild geese and ducks, 
among which many swans were seen, being distinguishable by 
their size and the whiteness of their plumage. I had seen large 
flocks of these birds before, in various parts of our country, and 
especially upon the Potomac, but never did I behold any thing © 
like the immense numbers here congregated together. Thousands 
of acres, as far as the eye could reach, seemed literally covered 
with them, presenting a scene of busy, animated cheerfulness, in 
most graceful contrast with the dreary, silent solitude by which 
we were immediately surrounded. 

Our course until noon was south, along the base of the high pro- 
montory which puts into the lake from the north. On our left, for 
about three miles from our encampment, was an isolated knob or hill, 
separated from the main range by a grassy plain. It consisted of 
limestone and slaty shales, in the former of which were some small 
caves. The rocks were thrown up at a very high angle, and in some © 

places were perpendicular, and rested, as far as could be ascertained, 
on a primitive formation below. Toward the southern end of the 


i 


’ 
) 


PANORAMIC 


‘Ackerman, Lith 379BrvadwayN” 


AMIC 


PROMONTORY RANGE—VIEW OF THE LAKE. 101 © 


promontory, the limestones.disappeared, and the surface rock was 
formed of conglomerate composed chiefly of the older sedimentary 
rocks, and some boulders of serpentine and porphyry. Upon ex- 
amining several isolated masses of this, it was found that each stone 
(principally rounded pebbles of quartz) was surrounded by a crystal- 
line layer of satin spar, as if it had formed a nucleus around which 
the lime had crystallized. In about ten miles we reached the south- 
ern extremity of this high rocky range, where it juts into the lake. 
Within this distance we passed five or six springs, some of them 
with very good water, bursting from the foot of the mountain. 
Innumerable salt and sulphur springs break out of the bank all 
along, but are soon lost in the broad sand and mud flat which 
lies between the banks and the water. This flat is about two 
miles broad, entirely without vegetation, and has, I think, been 
slightly covered by the lake in the spring and summer. Both 
yesterday and to-day, considerable quantities of small drift-wood 
was seen lying on the sands—a fact which favours this opinion. 

The mirage along the lake shore, and above the moist, oozy 
plains, has been, for the last two days, very great, giving rise to 
optical illusions the most grotesque and fantastic, and rendering all 
estimate of the distance or form of objects vague and uncertain. 
Two miles farther we reached a small rill of brackish, indifferent 
water, upon which we bivouacked, fearing to go on, lest we should 
be left without any. 

The evening was mild and bland, and the scene around us 
one of exciting interest. At our feet and on each side lay the 
waters of the Great Salt Lake, which we had so long and so ar- 
dently desired to see. They were clear and calm, and stretched 
far to the south and west. Directly before us, and distant only 
a few miles, an island rose from eight hundred to one thousand 
feet in height, while in the distance other and larger ones shot up 
from the bosom of the waters, their summits appearing to reach 
the clouds. On the west appeared several dark spots, resembling. 
other islands, but the dreamy haze hovering over this still and 
solitary sea threw its dim, uncertain veil over the more distant fea- 
tures of the landscape, preventing the eye from discerning any one 
object with distinctness, while it half revealed the whole, leaving 
ample scope for the imagination of the beholder. The stillness of 
the grave seemed to pervade both air and water; and, excepting 
here and there a solitary wild-duck floating motionless on the 
bosom of the lake, not a living thing was to be seen. The night 


* 102 RECONNOISSANCE AROUND GREAT SALT LAKE. 


proved perfectly serene, and a young moon shed its tremulous light 
upon a sea of profound, unbroken silence. I was surprised to find, 
although so near a body of the saltest water, none of that feeling 
of invigorating freshness which is always experienced when in the 
vicinity of the ocean. The bleak and naked shores, without a sin- 
gle tree to relieve the eye, presented a scene so different from what 
I had pictured in my imagination of the beauties of this mais. 
spot, that my disappointment was extreme. 

Tuesday, October 23.—Ther. at daylight, 87°. Morning eal 
and calm; the lake and mountains to the eastward yet wrapped in 
mist. The west side of the extremity of the promontory is composed 
of porphyry, interspersed with seams of white quartz, which 
veined it in the most beautiful manner. The quartz veins in some 
instances were several feet thick. These rocks, evidently in place, 
rose boldly, forming escarpments looking to the south-west, with a 
dip, apparently to the north, of about 50°. Decomposed lime- 
stone, containing organic remains, and also trap rock, were here 
observed. The ground near our encampment was covered with a 
species of Astragalus, the seed-pods of which were covered by a 
- substance resembling cotton, and presented the appearance of oval 
white balls, about the size of a robin’s egg. I afterward found 
this plant upon most of the islands of the lake. 

Rounding the point of the promontory, the shore of the lake 
trends off to the northward, forming several picturesque little bays 
with bold rocky headlands. After travelling about nine miles, we 
came to several springs of good and most welcome water, and we 
stopped to refresh our animals and to noon. The finding of this 
water was entirely unexpected, as, from the representations of an 
old Shoshonee Indian, made to us before leaving Bear River, I did 
not look for any for two days, and had in consequence dismounted 
one of the men to enable us to carry the more vessels, all of which 
had been filled before leaving our camp in the morning. I went 
down to the shore of the lake to taste of the water: it was as salt as 
very strong brine, and clear and transparent as diamond. A large 
flock of gulls was swimming about near the shore. After feasting our 
animals upon the grass that grew among the tall rushes and canes 
around the spring, we continued along the shore of the lake for 
about nine miles farther, and succeeded in discovering three springs 
within that distance, at the last of which we halted for the night. © 

After doubling the southern end of the promontory, the broad 
flais, which had characterized the shore at its eastern base, en- 


AVE NNTVY WOdd GNVISI SHdOTSHLINV 40 LYVd GNV GNVISI S.LNOWAYa 30 GNA LSamM 
AN Aempeorg ¢/¢ Wr] BEMLaPApYy Z 


WESTERN BASE OF PROMONTORY RANGE. 108 


_ tirely disappeared, and the water, although apparently shallow, 


came nearly up to the base of the hills. Near the margin of the 
lake it is not safe in all places for animals to pass, as the almost 
constant exudations of salt water from the edge of the grass, un- 
dermine the surface, rendering the narrow intervening beach 
treacherous and miry. The water to the westward appears bold 
and deep; and enough has been seen to convince me that a large 
sail-boat will be absolutely indispensable in the contemplated sur- 
vey, for the supply of the different parties with provisions and 
water. Wood there is none. Fuel for cooking, can, however, be 


generally obtained from the artemisia which abounds almost every- 


where; but timber for the construction of the triangulation sta- 
tions, will, in most instances, have to be transported by water, or 
hauled down from the cafions of the mountains. 

The rocks observed were porphyry, gneiss, dark slaty shales, and 
metamorphic sandstone, dipping to the north-east. After proceed- 
ing some miles to the north, dark limestones with white marble 
veins occurred, alternating with clayey shales. The rocks on this 
side of the promontory are much more rugged than on the other, 
or eastern slope, presenting numerous lofty escarpments where 
they crop out, the dip being to the east. A cactus, with very long 
prickles, was observed near our morning camp; and at the spring 
where we nooned, a small jointed cane trailed on the ground, in 
some instances to a distance of more than thirty feet. The men 
made excellent pipe-stems of this material. The spring where we 
encamped for the night was an oval hole-or pit, with perpendicular 
sides, about fifteen feet long, six broad, and four deep. The water 
was tolerably good: a small spring, rising at the base of the 
hill, ran into the lake close by. These springs afterward afforded 
us nearly all the water used upon the survey of the west shore of 
the lake; but a voyage of fifty miles was frequently necessary to 
obtain a supply even for a few days. 

Wednesday, October 24.—Clear and calm. Ther. at daylight, 
19°; sunrise, 24°. Continuing our journey up the lake-shore, we 
shortly came to a brackish spring, where there had been a camp of 
Indians the night before. We had thought last night that we saw 
their fires, but they had fled, alarmed probably by the report of 
some guns that had been discharged in ourcamp. A quantity of 
some species of seeds they had been beating out lay in small heaps 
around, and I found an old water-bottle they had left in their haste. 
It was ingeniously woven of a sort of sedge-grass, coated inside 


104 RECONNOISSANCE AROUND GREAT SALT LAKE. 


with the gum of the mountain pine, by which it was rendered per- 
fectly water-tight. I afterward saw some similarly shaped vessels, 
and made of the same material, that would hold nearly two gallons. 
As nothing was to be gained by rigidly following the lake-shore, 
I determined to cut across the projecting points, keeping the gene- 
ral features of the lake inview. At this point we came upon a low 
range of basaltic hills, extending some miles west of the mountains 
which continued to the northward, and presenting a steep escarp- 
ment on the lake, where we again struck it. This lower series of 
hills extended also to the north, and we followed along their base 
for many miles, the range gradually falling off to the east as we 
advanced. The general soil was white clay, formed from the de- 
composition of the rocks. At three o’clock, having travelled 
eighteen miles without water, we halted, removed the packs from 
the backs of our weary beasts, and served out from our scanty store 
a pint of water to each mule, which the famished creatures eagerly 
drank from a tin pan. We remained here a couple of hours, to 
allow them to graze on some tolerably good bunch-grass, when we 
again saddled up at sundown, and continued our journey, deter- 
mined to go on till water should be found, or at least as long as 
the animals could travel. At ten o’clock we reached a small slug- 
gish stream, containing some water entirely too salt for our use, 
but which the poor animals drank with great avidity, having been 
without for more than twelve hours. Here we lay down for the 
night, both man and beast much fatigued with the day’s march. 
The country passed over to-day has been barren, desolate, and 
forlorn tothe last degree. Artemisia has prevailed to the exclusion 
of all other vegetation. Not the note of a bird nor the chirp of 
an insect was to be heard. A solitary crow and one grasshopper 
were the only living things seen during the whole day’s march. 
Thursday, October 25.—Ther. at sunrise, 24°. We had an op- 
portunity this morning of seeing fully the ground over which we 
had passed the night previous. It consisted of an oval flat of clay 
and sand, some four or five miles broad from east to west, and ex- 
tending double that distance toward the north; bounded on both 
sides by lofty hills, with high mountains in the background. 
North of the flat the ridge was much lower, and it appeared as if 
there were a pass or depression through it, leading to another valley 
or plain beyond. Three streams came down from this low ridge, 
and, flowing to the southward, either sank into the sand or discharged 
themselves into the lake, which we now judged to be some six or 


HAV LVS LVAH9 LNIOd AYOLNONOYd WOOF LSIM HLYON DNIMOOT MFIA 


2 
f ta 
itt 
i , 


NORTH END OF THE LAKE—INiAND SEA. 105 


eight miles to the southward, the flat extending in that direction 
to the water’s edge. Two of these streams (all of which were salt) 
we crossed without much difficulty ; but the third, on the western 
side of the flat, was impassable, and we had to ascend it for three 
miles before we could obtain a crossing. On the west side of this 
latter branch comes in a small tributary, in the bed of which, near 
its source, a beautiful spring, ten feet wide, bubbles up from the 
bottom, with a column of water rising in its centre six inches in 
diameter. The water was clear as crystal, but salt and sulphurous, 
which latter quality might account for the numerous tracks of the 
antelope around its margin, as that animal is known to delight in 
‘waters of this character. ; 

This extensive flat appears to have formed, at one time, the 
northern portion of the lake, for it is now but slightly above its 
present level.. Upon the slope of a ridge connected with this 
plain, thirteen distinct successive benches, or water-marks, were 
counted, which had evidently, at one time, been washed by the 
lake, and must have been the result of its action continued for 
some time at each level. The highest of these is now about two hun- 
dred feet above the valley, which has itself been left by the lake, 
owing probably to gradual elevation occasioned by subterraneous 
causes. If this supposition be correct, and all appearances conspire 
to support it, there must have been here at some former period 
a vast znland sea, extending for hundreds of miles; and the isolated 
mountains which now tower from the flats, forming its western 
and south-western shores, were doubtless huge islands, similar to 
those which now rise from the diminished waters of the lake. 

In passing over this mud-plain, the glare from the oozy sub- 
stance of which it is composed was extremely painful to the eyes. 
Leaving it behind us, we ascended a ridge to the west of it, two 
or three miles broad, passing over some remains of shales and 
altered limestone with conglomerate, the crest being composed of 
porous trap, underlying the sedementary rocks, and cropping out 
to the west. It may be remarked here, that the general direction 
of all the ridges noticed in this region is north and south, and 
they terminate most frequently in sharp, bold promontories, to the 
south. A herd of antelope was seen on this ridge, numbering 
about a hundred, but too wild to be approached. 

Descending its western slope, we came into another plain, some- 
what similar to the last in form, but much more extensive in all di- 
rections, and densely covered with artemisia. Over this desolate, 


. 106 NIGHT MARCH—BIVOUAC. 


barren waste, we travelled until nearly dark, when we reached @ 
rocky promontory, constituting the southern point of a lew ridge of 
hills jutting into the plain from the north. The rock was porous trap, 
in which no stratification could be made out. The mules having been 
without water or grass the whole day, and our stock of the former 
being insufficient to give them even their stinted allowance of one 
poor pint, we halted for a couple of hours, and drove them upon the 
side of the mountain to pick what they could get from the scanty 
supply of dry bunch-grass that grew im tufts upon its side. The 
prospect of water now began to be rather gloomy; and I was 
obliged to put the party upon allowance, lest we should be left 


entirely destitute. At eight o’clock we replaced the packs upon — 


our mules, all of which began to show the effects of their unusual 
abstinence, and rode on till near midnight by the light of the 
moon, in a south-westerly direction, over a country similar to that 
we had traversed during the day ; when, finding the indications of 
water growing less and less promising, and that our animals were 
nearly worn out, we halted, and, covered with our blankets, we 
lay down on the ground till morning, regardless of a heavy shower 
that fell during the night. | 

Friday, October 26.—The poor animals presented this morning 
a forlorn appearance, having been now without a drop of water 
for more than twenty-four hours, during eighteen of which they 
had been under the saddle, with scarcely any thing toeat. 1 now 
began to feel somewhat anxious. Should our mules give out before 
we could reach the mountains west of us, to which I had deter- 
mined to direct our course as speedily as possible, we must all 
perish in the wilderness. Sweeping the horizon with a telescope, 
I thought I discovered something that looked like willows to the 
north-west, distant about four or five miles. Reanimated by this 
gleam of hope, we saddled up quickly and turned our steps in 
that direction. We soon had the lively satisfaction of finding our 
expectations confirmed; for, arriving at the spot, we found, after 
some search, a small spring welling out from the bottom of a little 
ravine, which having with some labour been cleaned out, we soon 
enjoyed a plentiful, most needed, and most welcome supply of 
excellent water for all. | 

The whole party being much exhausted from their long absti- 
nence and unceasing exertions, we halted here for the day, to 
afford opportunity for our animals to recruit their wasted strength 
upon the plentiful supply of grass which grew all around us. 


IRON ORE—ANCIENT LAKE SHORE. 107 


Old decayed wigwams, constructed of willows, indicated that this 
spot had long been a favourite place of resort for the Indians, for the 
same reason, doubtless, which rendered it so welcome to ourselves. 

On the summit of a ridge south-west of our halting-place, large 
masses of magnetic iron ore were discovered, some of which were 
partially encased in basaltic rock. In the ravine whence the spring 
broke out, were found pebbles of alabaster, obsidian, and other 
_ rocks, apparently the result of the disintegration of beds of con- 
glomerate, none of which, however, was seen in place. 

Saturday, October 27.—Ther. at sunrise, 85°. Resuming our 
journey, we took a course south by east, which led us past the 
ridge upon which we had halted two nights before. The forma- 
tion was porous trap, and the direction of the ridge north by west 
and south by east. We then passed along the base of a range of 
low hills, composed apparently of trap and basalt. After travel- 
ling ten miles, we came to a’ range of higher hills extending north- 
west and south-east. Here the dark limestone was again observed, 
but the stratification could not be ascertained. We then passed, 
in a southerly direction, through deep sand, along what at one 
time had been the beach of the lake, as. drift-wood was frequently 
seen lying on the sands that stretch out to the eastward for many 
miles. In one instance a drifted cotton-wood log was seen, lying 
near what had evidently been the water-line of the lake, as thick 
as the body of a man. On-our right was a high ridge or promon- 
tory, with a narrow bottom sloping down to the edge of the flat. 

The soil here was not so clayey as heretofore, beg composed in 
many places of calcareous: sand and decomposed conglomerate. 
Some masses of the latter were seen, resembling exactly that met 
with on the eastern side of the promontory range putting out into 
the lake. The country to-day has been similar to that passed over 
previously—dry, barren, and entirely destitute of water. We dug 
a well some five feet deep on the edge of the flat, which soon filled 
with water. The mules crowded around the hole, and seemed to 
watch the process of our labour, as if sensible of the object of our 
exertions, but upon tasting the water, refused to drink, although 
they had been travelling the whole day without a drop. Day’s 
march, about sixteen miles. 

Sunday, October 28.—Our little stock of water had become so 
reduced that we were compelled to forego our coffee this morning, 
and the most rigid economy in the use of the former was strictly 
enforced. We were on the road very early, and followed for seve- 


108 SUFFERING FOR WATER—DESERT PLAINS. 


ral miles, down the edge of the sand at the foot of the range of hills 
on our right, when we ascended it, taking a course south-west by 
west, and passing over beds of conglomerate, which presented a 
stratification almost horizontal. The ridge was about five miles 
wide, stretching off to the southward, and about five hundred feet 
above the level of the beach. ‘The soil consisted of decomposed 
conglomerate, and was much cut up by deep ravines. On the west 
side, volcanic rock was again met with. 

Leaving the ridge, we entered upon a plain or sort of bay, 
partly covered with artemisia, and partly (to the westward) with 
mud and salt. It appeared to be bounded on the west, about 
thirty miles distant, by a high mountain-range, extending far to the 
northward, upon an eastern spur of which I judged we had encamped 
on Friday. The plain contained several island mountains, rising 
from it as from the water. To one of these, distant about twelve 


miles south-west by west, we directed *our course and reached it — 


about an hour before sunset. Here we stopped for a short time to 
prepare our scanty supper, and to give the mules a chance to pick 
a little grass, which was scarce and dried up. Nota drop of water 
had we met with the whole day ; but at noon [ had ordered a pint to 
be served out to each animal. Before arriving at this spot, one of 
the poor creatures “gave out,” and we thought we should have to 
leave him to the wolves, but he afterward partially recovered, and 
another pint of water being given him, he went on. The rocky 
island, at the north end of which we halted, extended many miles 
to the southward, and was apparently surrounded on all sides by 
the mud-plain. One of the party ascended it, but could see nothing 
of the lake, nor any appearance of water in any direction. The 
rocks were formed of altered clayey and sandy shales, and strata 
of conglomerate, all of which had been much contorted, but evi- 
dently at different periods, as they were not conformable. 

It now became a matter of serious importance to find water for the 
mules, as they had been without for nearly forty hours, most of the 
time under the saddle, and almost without food. Nothing, therefore, 
remained but to go on as far as possible during the night, so as to 
reach the western ridge bounding this basin as early the following day 
as practicable. We accordingly saddled up about dark and proceed- 
ed on the same course, directing our steps toward another island in 


the plain, which appeared to be about fifteen miles distant. Thenight © ° 


was quite cold,and the moon shone as bright as day. Our course 
lay over a flat of damp clay and salt mud, in many places soft and 


Te q . 
Te ee es 


PERILOUS POSITION—ANCIENT CRATER. 109 


‘deep, which made the travelling slow and laborious. All trace of 
vegetation had vanished, and even the unfailing artemisia had dis- 
appeared. ‘The animals were so tired and weak that the whole 
party was on foot, driving our herd before us. The mule which 
had given out in the afternoon was now unable to proceed, and 
had to be abandoned in the midst of the plain, where it no doubt 
perished. Many others showed symptoms of extreme exhaustion, 
so that their packs had to be shifted and lightened repeatedly. I 
began to entertain serious fears that I should not be able to reach 
the mountain with them; nor was I certain that when we did reach 
‘it we should be able to find water in time to save their lives. The 
night was consequently passed in a state of great anxiety. We 
continued on until after midnight, crossing occasionally some little 
drains of salt water coming from the north, when we reached a 
small isolated butte, which was only a pile of barren rocks, with 
‘scarce a blade of grass upon it. Wood or water there was none; so, 
although the night was quite cold, we laid ourselves down, fireless and 
supperless, upon the sand, wearied to exhaustion by a continuous 
march of eighteen hours. The only sign of vegetable life to be 
seen here was a small chenopodeaceous plant, without leaves, but 
having long prickles. The artemisia had entirely disappeared. 
On each side of us, to the north and the souta, was a rocky island 
or butte, similar in character to the one near which we had halted, 
but much larger. 

Monday, October 29.—On abil early, we found the mules 
gathered around us, looking very dejected and miserable. They 
had searched in vain for food, and were now in nearly a starving 
condition. Before us, indeed, lay the mountain where we hoped 
to find both food and water for them, but between lay a mud- 
plain fifteen or twenty miles in extent, which must be crossed be- 
fore we could reach it. I was much afraid the animals were too 
weak to succeed in the attempt, but it was our only hope. We 
set out, the whole party on foot, pursuing the same general course 
of south-west by west that we had followed yesterday. 

_ The island, at the foot of which we had slept last night, presented 
sections of sandstones and shales, which appeared to be of com- 
paratively recent origin. They iad evidently been somewhat 
altered by heat, but not to any great extent. At the north-east 
point of the island on our left, the strata were inclined at an angle 
of 70° to north-east. No fossils were found in them. Near the 
western side of this rocky protrusion, I observed what appeared 


110 FIELD OF SALT—WELCOME RELIEF. 


to have been an ancient crater, forming three-fourths of an inverted 
cone, open to the north-west, around which were sections of shales 
and sandstones, very much contorted, and dipping in opposite 
directions on opposite sides. ‘The lower part of the cone was 
filled with claystone. No volcanic rocks were found at the point 
where we crossed these islands, but decomposed conglomerate and 
alabaster occurred in considerable quantities. 

The first part of the plain consisted simply of dried mud, with 
small crystals of salt scattered thickly over the surface. Crossing 
this, we came upon another portion of it, three miles in width, where 
the ground was entirely covered with a thin layer of salt ina state of 
deliquesencce, and of so soft a consistence that the feet of our mules 
sank at every step into the mud beneath. But we soon came upon a 
portion of the plain where the salt lay in a solid state, in one un- 
broken sheet, extending apparently to its western border. So 
firm and strong was this unique and snowy floor, that it sustamed 
the weight of our entire train, without in the least giving way or 
cracking beneath the pressure. Our mules walked upon it as upon 
a sheet of solid ice. The whole field was crossed by a network 
of little ridges, projecting about half an inch, as if the salt had 
expanded in the process of crystallization. I estimated this field 
to be at least seven miles wide and ten miles in length. How much 
farther it extended northward I could not tell; but if it covered the 
plain in that direction as it did where we crossed, its extent must 
have been very much‘greater. The salt, which was very pure and 
white, averaged from one-half to three-fourths of an inch in thick- 
ness, and was equal in all respects to our finest specimens for 
table use. Assuming these data, the quantity that here lay upon 
the ground in one body, exclusive of that in a deliquescent state, 
amounted to over four and a-half millions of cubic yards, or about 
one hundred millions of bushels. ‘ 

At two o’clock in the afternoon we reached the western edge 
of the plain, when to our infinite joy we beheld a small prairie 
or meadow, covered with a profusion of good green grass, through 
which meandered a small stream of pure fresh running water, 
among clumps of willows and wild roses, artemisia and rushes. 
It was a most timely and welcome relief to our poor famished 
animals, who had now been deprived of almost all sustenance 
for more than sixty hours, during the greater part of which 
time they had been in constant motion. It was, indeed, nearly 
as great a relief to me as to them, for I had been doubtful 


PILOT PEAK—INDIAN CEDAR LODGES. 111 


whether even the best mule we had could have gone more than 
half a dozen miles farther. Several of them had given out in 
crossing the last plain, and we had to leave them and the baggage 
behind, and to return for it afterward. Another day without 
water and the whole train must have inevitably perished. Both 
man and beast being completely exhausted, I remained here three 
days for refreshment and rest. Moreover, we were now to prepare 
for crossing another desert of seventy miles, which, as my guide in- 
formed me, still lay between us and the southern end of the lake, 
He had passed over it im 1845, with Frémont, who had lost ten | 
mules and several horses in effecting the passage, having afterward 
encamped on the same ground now occupied by our little party. 

During our stay here, it rammed almost every day and night. 
_ The salt plain, which before had glistened in the sunlight like a 
sheet of molten silver, now became black and sombre; the salt, 
over which we had passed with so much ease, dissolved, and the 
flat, in places, became almost impassable. We had encamped at 
the eastern base of a range of high mountains, stretching a great 
distance to the north, and terminated, three miles below, in an ab- 
rupt escarpment, called Pilot Peak: upon the lofty summit of which 
rested a dark cloud during the whole of our stay. For three miles 
from the base the ascent is gradual, the surface being covered with 
gravel and boulders of granite, feldspathic rock, and metamorphic 
sandstones, all evidently waterworn. Higher up the mountain, 
the only stratified rocks seen were micaceous schists and slaty 
shales, intersected in various directions by veins of quartz, and 
very much displaced. The general dip was north by east from 
70° to 80°. Proceeding south a few miles along the mountain, 
the same stratified rocks were again noticed, evidently much al- 
tered by heat, being interspersed with veins of granite and quartz. 
Dwarf cedar was growing here, and, higher up the mountain, dwarf 
pine; in the bottom, white and red willow, and Hquisetwm. 

In a nook of the mountain, some Indian lodges wereseen, which had 
apparently been finished but a short time. They were constructed 
in the usual conical form, of cedar poles and logs of a considerable 
size, thatched with bark and branches, and were quite warm and 
comfortable. The odour of the cedar was sweet and refreshing. 
These lodges had been put up, no doubt, by the Shoshonee Indians 
for their permanent winter-quarters, but had not yet been occu- 
pied. The savages had been in the neighbourhood to collect the 
nuts of the pine-tree, called here pifion, for food, but what they 


-_ 


112 STOLEN MULE—HASTINGS’S CUT-OFF. | 


left had been destroyed by insects. While at this camp, one of 
our best mules, was stolen. A couple of men, whom I had sent 
back across the plain to search for a revolver that had been lost 
in our last night march, reported, on their return to the camp, that 
they had discovered the tracks of two Indians on our trail, and 
had seen their fires in the mountains. These stealthy depredators 
must have followed us at a distance and watched their opportunity 
to plunder. The only wonder is that they did not steal more than 
a single mule; for the country was so utterly desolate, that we 
never once thought that any human being would ever be found 
where we had passed, except from absolute necessity, and conse- 
quently the vigilance of our night-guard was relaxed. Snow fell 
the night before we left this camp, and covered the ridge about 
halfway down from its summit. 

Friday, November 2.—Ther. at sunrise, 19°. As we were aware 
that immediately before us lay another desert plain, without wood, 
water, or grass, for seventy miles, some little preparation was ne- 
cessary before undertaking to cross it. This consisted simply in 
baking bread and cooking meat enough to last us through, and in 
packing upon our mules as much grass as they could carry, which 
we had cut, a handful at a time, with our hunting-knives. We had 
only vessels sufficient to carry twenty gallons of water—a small 
supply for so many men and animals. The mules, however, were 
now much recruited by their rest, and we started in good spirits. 
Following the western edge of the mud-plain at the foot of the 
range for three miles, we came to the southern point of the moun- 
tain, where there had been an encampment of emigrants, who had 
taken this route from Salt Lake City in 1848. There were here 
several large springs of excellent water, and the encampment had 
apparently been quite a large one. The usual destruction of pro- 
perty had taken place. Clothes, books, cases of medicine, wagon- 
wheels, tools, &., lay strewn about, abandoned by their owners, 
who had laboriously brought them two thousand miles only to 
throw them away. 


The route from the Salt Lake to this point was first taken by’ 


Colonel Frémont, in 1845. A year afterward, it was followed by 
a party of emigrants under a Mr. Hastings, whence its present 
name of « Hastings’s Cut-off.” A portion of his company, which 
had followed at some distance behind him, becoming belated in 
crossing the Sierra Nevada Mountains, a number of them pe- 
rished, and the remainder were reduced to the revolting necessity 


Pit - ‘ : 
a a 


THE SEVENTY-MILE DESERT. 113 


of living upon the bodies of their dead comrades, until they were 
rescued by relief from Sutter’s Fort. 

The road to California from this point follows around the south- 
ern end of the ridge, passes to the north of another high moun- 
tain, and thence on to the head of Humboldt’s or Mary’s River. 

Leaving the springs, we crossed, once more, though in an oppo- 
site direction, the same mud-plain over which we had been obliged 
to pass in order to reach the mountain. It was twelve miles in 
width; and now, in consequence of the recent rains, was soft and 
slippery—all the salt having disappeared, except a few crystals 
left in some old wagon-tracks. The travelling was in consequence 
heavy and laborious. After crossing, we passed, by a gentle as- 
cent, over a neck of land which connected the high ridge on our 
left, at the north end of which we had bivouacked on the 29th, 
with another and broader one to the south, and which latter turned 
off considerably to the south-west. Here we halted for a short 
time, to give our mules their last chance to pick a little bunch- 
grass which grew in thin scattered tufts on the mountain-side. 

The strata, at this point, were very much contorted, as at the 
northern end of the same protruded ridge, inclining in all direc- 
tions. The higher hills were composed of dark limestone, traversed 
in various directions by veins of white marble, some of which were 
of considerable thickness. The dip was to the north-west, 65°. 
Over the limestone were beds of conglomerate, not conformable ; 
the lower layers of which, or those in immediate contact with the 
limestone, consisted of portions of the rock that had not been 
waterworn. Lower down, near the base of the hill, was found a 
coarse, imperfect oolitic limestone, dipping about 50° to north-west, 
and under these some sandstone, not conformable, and imperfect. 

After halting an hour, we pursued our journey along the east- 
ern base of this isolated mountain or butte, where the dark lime- 
stone was again seen, with gypsum, conformable and at right 
angles with the strata. Some six miles farther on, we passed 
another isolated butte, upheaved through the level mud-plain, con- 
taining what appeared to be another crater, analogous to that seen 
on the northern end of the ridge, open to the eastward, with the 
strata dipping in every direction. The main butte appeared to be, 
at this end, about ten miles wide from east to west, and had mani- 
festly been very much disturbed. | 

From this point we travelled on until past midnight, over a 
level mud-plain, lighted by the rays of the moon, which struggled 

8 


114 SEVENTY-MILE DESERT—EMIGRANT CACHE. 


through a mass of dark and threatening clouds. The wind was 
fresh and cold, and the mud soft and tenacious, making the tra- 
velling very slow and fatiguing. During the night, we passed five 
wagons and one cart, which had stuck fast in the mud, and been 
necessarily left by their owners, who, from appearances, had aban- 
doned every thing, fearful of perishing themselves in this inhos- 
pitable desert. Great quantities of excellent clothing, tool-chests, 
trunks, scientific books, and, in fact, almost every thing, both use- 
less and necessary on a journey of this kind, had been here left 
strewn over the plain. Many articles had not even been removed 
from the wagons. ‘The carcasses of several oxen lying about on 
the ground satisfactorily explained the whole matter. In attempt- 
ing to cross the plain, the animals had died from exhaustion and 
want of water, and the wagons and their contents had of course 
to be abandoned. 

About one o’clock in the morning, we halted in the midst of 
the plain, enticed by the sight of a broken ox-yoke, the remains 
of a barrel, and part of an old wagon-bed, which served for 
fuel sufficient to boil a little coffee, of which all hands stood 
very much in need. The mud was ankle-deep; and the only 
place upon which we could spread down a blanket to sleep was 
around some scattering bushes of artemisia, where the wind had 
collected a little sand, presenting a spot rather higher and not so 
wet as the mud-flat around. ‘The whole scene was as barren, 
dreary, and desolate as could be well imagined. We gave the 
mules a portion of the grass that had been packed upon them in 
the morning, and two pint-cups of water each—the only liquid 
they had tasted during the day. We then fastened them up as 
well as we could to the artemisia-bushes, and, wrapping ourselves 
in our blankets, lay down to wait for the morning. The night was 
windy and quite cold, and the poor mules kept up such a pitiful and 
mournful cry, that we were but little recruited by our night’s rest. 

It may well be supposed that there were few attractions to de- 
tain us long on this spot. We had exhausted our fuel last night, 
and there was nothing with which to cook breakfast ; so we started 
quite early without any, pursuing the same general course through 
the heavy mud. The wind, uninterrupted by any obstacle, blew hard 
over the level plain; and although the thermometer stood at only 
47°, yet it was very cold, and brought into requisition all appliances 
for preventing the escape of animal heat. In the course of the 
morning, we passed a spot where some emigrants had made a large 


SEVENTY-MILE DESERT. 115 


“cache” of such things as they could not carry. But it had been 
constructed in such a bungling manner, that it had easily been dis- 
covered and robbed: twelve ox-yokes remained in a heap on the 


_ ground. After travelling until noon, we came to a low ridge of . 


hills running nearly north and south. We sheltered ourselves be- 
hind it, and finding plenty of artemisia, kindled a fire, and boiled 
our coffee, which, with a piece of bread and cold bacon, constituted 
our first and only meal for the day. 

Our poor animals looked wretchedly, and two of them giving out 
before reaching the ridge, were with great difficulty driven up. As 
they had been without water for twenty-four hours, except the 
cupful which had been served out to them last night, after filling a 
few canteens for our own use, the remainder of our little stock was 
divided among them. 

The ridge was composed of porous trap. The hills were higher 
toward the north, where they were connected with a range which 
seemed to form a spur from the mountains east of us. They gradu- 
ally diminished to the south, not extending more than a mile or 
two in that direction. 

Before us, distant about twelve miles, was a high mountain-range, 
on the eastern side of which, the guide informed me, there was a 
spring with plenty of water. I had hoped to be able to cross it to- 
day, but the state of our animals was such that it proved impractica- 
ble, since it was dark before we reached its western base. I the less 
regretted this, as in the course of the afternoon we had found 
several little pools of rain-water, from which the mules drank with 
great avidity and to repletion. The ascent to the range was gen- 
tle, and we encamped at the mouth of a narrow, winding pass 
through it, amid plenty of large cedar-trees and very large arte- 
misia—a welcome sight, as the day had been cold and blustering, 
and there was every prospect ofa heavy storm. Large fires were soon 
blazing, and every one was tired enough at once to seek his blanket, 
without going to the trouble of preparing the evening meal. Indeed, 
there was little or nothing to prepare; our bread was all gone, 
and there was not water enough either to make bread or coffee, and 
none could be sought for in the dark. The two mules that had 
failed in the morning, again gave out before reaching the moun- 
tain, and had finally to be abandoned. . 

_ Sunday, November 4.—Ther. 33°. Upon rising we found it 
snowing hard, and the ground covered to the depth of two inches. 
Tt soon ceased, however, and before night had melted in the plain, 


116 if SPRING VALLEY RANGE. 


although the neighbouring mountains continued all whitened by it. 
After much search, water was found in a deep ravine near by; and 
grass was tolerably abundant, though dry and hard. As the mules 
were nearly exhausted and much stiffened by their journey across 
the deep mud-plains, I determined to remain here for the day, to 
recruit them. A couple of men were sent back afoot, to try and 
recover those left yesterday: they returned, after dark, bringing 
with them one only; the other had strayed from the road, and all 
efforts to recover it were vain. It was the third lost on this trip. 

A deep ravine at the foot of the mountain presented sections 
of the strata. The lowest exposed was dark limestone with white 
veins, inclined to the south-east, at an angle of 85°; in fact, almost 
vertical. Ascending the ravine, the limestone was found to be 
overlaid by red sandstone, and this again by clayey shales. All 
these rocks had been altered by heat. No organic remains were 
found in the sandstones ‘or shales, but some corals were seen in the 
limestone. The rocks were all veined with white marble. Large 
crystalline nodules of this substance were found, which assumed 
the form of arragonite. Some specimens of iron ore were also 
found, apparently a carbonate, but not in place. Ther. at noon, 
37°; sunset, 31°. 

Monday, November 5.—Ther. 23°. Morning clear and quite 
cold. Crossed the mountain through the pass. The snow was about 
two inches deep and the ground frozen hard. Fcllowed down the 
eastern slope for about two miles, when we came to a spring-branch 
issuing from a gorge of the mountains where there was plenty of 
green grass—the first full supply our animals had enjoyed for 
several days. 

The only rocks observed in crossing the mountain were lime- 
stones, containing remains of encrinites and corals. A wide dike 
of trap formed the crest; and, on the eastern side, another dike 
was seen running north by east, and south by west, forming the 
summit of a lower ridge. The limestones were tilted up almost 
vertically, but as the surface of the ground was covered with snow, 
the nature of the strata and their direction could not be very ac- 
curately ascertained. A piece of altered coal was found at the 
eastern base of the mountain, but not in place. The lower hills 
were covered with conglomerate not conformable. 

Leaving the spring, our true course lay about east, to strike the 
southern point of another range ten miles distant, Ge forming the 
eastern boundary of a broad, green, intervening valley, Ga ex- 


SPRING VALLEY. 117 


tended northward to the southern shore of the Great Salt Lake, 
and was covered with grass, the first we had seen since leaving Pilot 
Peak. It was shut in toward the south by a range of comparatively 
low hills, connecting the two mountain ranges that formed its 
eastern and western boundaries. A direct course could not be 
taken for this point, owing to numerous springs, which rendered 
the valley in that direction marshy and wholly impassable. We 
were consequently forced many miles to the southward, and obliged 
to make a circuit of more than a semicircle to gain the opposite 
side. We followed down the western base of the mountain for two 
or three miles, passing a fine spring, with good grass, near which we 
encamped for the night, among some dwarf cedars, that both fur- 
nished us with fuel and afforded a protection against the wind, 
which blew fresh and cold from the north-west. Ther. at sunset, 43°. 

Tuesday, November 6.—Ther. at sunrise, 30°. Continued our 
journey in a northerly direction, along the western base of the 
mountain, for twelve miles, when we reached its northern ex- 
tremity, which was about a mile and a-half wide, and terminated 
in bold escarpments five or six hundred feet high. One of these 
resembled, in a remarkable manner, a huge castle, the vertical walls 
of which were not less than three hundred feet in height. 

Before reaching this point of the mountains, I remarked, on our 
left, in the middle of the valley, a curious, isolated mass of rocks, 
resembling a small fortification or redoubt: it was surrounded by ~ 
marshy meadow-land, and could, in case of need, be defended by a 
small force against almost any number of Indians. Numerous 
springs broke out from the mountain and at the edge of the prairie; 
but they were all saline, with a temperature of 74°, and totally 
unfit to drink. To this place we gave the name of «Spring Val- 
ley.” Near the point of the mountain was a very large spring, 
which discharged its waters northward into the lake. The water 
was very salt, nauseous, and bitter, with a temperature of 70°; 
notwithstanding which it swarmed with innumerable small fish, and 
seemed to be a favourite resort for pelicans and gulls. 

In a shallow ravine near our morning camp, limestone was found 
cropping out, with a dip of 80° to the north-west. This rock was 
seen as we followed the range, appearing at the spurs; and dikes of 
trap were observed, forming peaks farther back up the mountain- 
side. The ridge gradually became less elevated as we proceeded 
toward the point, where the stratified rocks (limestone and shales) 
were found in a horizontal position. Along the northern termina- 


118 TUILLA VALLEY. : 


tion of the range, the strata were again found to be much dis- 
placed and almost vertical. They were composed of limestone 
and shales, overlaid in some places by conglomerate. Salt springs 
were very numerous in this locality. 

After doubling the point, we came upon another valley, similar 
to the one through which we had just passed, and from which it was 
divided by the ridge or mountain just described. Our true course 
here, also, was to cross this valley in an easterly direction, and 
strike the northern point of another range where it terminates im- 
mediately on the southern shore of the Salt Lake, now plainly 
visible; but the numerous salt springs, as in the case of that 
passed yesterday, rendered a straight course impracticable. Con- 
sequently, after following the eastern base of the ridge about 
six miles to the south, we began gradually to diverge from it to the 
eastward, and at dark encamped in the prairie, near a noble spring 
of fresh, cold water, with abundance of excellent grass, and an ex- 
tensive grove of large willows for fuel. A fierce gale sprang up 
from the south-east, which kept us in a constant state of alarm 
during the night, lest we should be burned in our beds from the 
tall dry grass taking fire. It had in fact kindled several times, 
and the flame was extinguished with some difficulty, rendering a 
strict watch necessary until morning. ‘This valley is called « Tuilla 
Valley’ by the Mormons, and forms an excellent pasturage for 
numerous herds of cattle, wintered here by them under the charge 
of keepers. The grass is very abundant, and numerous springs are 
found on both sides of it. | 

On the eastern side of the mountain, which divides it from 
Spring Valley, the same geological appearances occur as were seen 
yesterday at the point of the range and on its western side. The 
limestones were thrown up at avery great angle, and in some 
places the strata were perpendicular. 

Another mule gave out to-day, and was necessarily abandoned. 
Ther. at sunset, 43°. 


Wednesday, November T.—Ther. at sunrise, 47°. Starting 


early in the morning, we crossed to the eastern side of the valley, 
followed the base of the mountain to its northern extremity, and 


reached the shores of the Great Salt Lake near Black Rock, 


whence we crossed the valley of the Jordan, over sterile artemisia 
plains, and reached the city in the afternoon—being the first party 
of white men that ever succeeded in making the entire circuit of 


the lake by land. Attempts had, in early times, been made to 


Caen: Ae “t PT i es 


Ce ee - 


re i> ee 


ihe 


_ 


aS 
ae 


RESULTS OF THE RECONNOISSANCE. 119 


circumnavigate it in canoes, by some trappers in search of beaver ; 
but they all proved unsuccessful, from want of fresh water. 

The examination just completed proves that the whole western 
shore of the lake is bounded by an immense level plain, consist- 
ing of soft mud, frequently traversed by small, meandering rills 
of salt and sulphurous water, with occasional springs of fresh, all 
of which sink before reaching the lake. These streams seem to 
imbue and saturate the whole soil, so as to render it throughout 
miry and treacherous. For a few months, in midsummer, the sun 
has sufficient influence to render some portions of the plain, for a 
short time, dry and hard: in these intervals the travelling over it 
is excellent ; but one heavy shower is sufficient to reconvert the 
hardened clay into soft, tenacious mud, rendering the passage of 
teams over it toilsome, and frequently quite hazardous. 

These plains are but little elevated above the present level of the 


_ lake, and have, beyond question, at one time formed a part of it. It 


is manifest to every observer, that an elevation of but a few feet 
above the present level of the lake would flood this entire flat to 
a great distance north and south, and wash the base of the Pilot 
Peak range of mountains, which constitute its western boundary ; 
thus converting what is now a comparatively small and insignificant 
lake into a vast inland sea. This extensive area is, for the most 
part, entirely denuded of vegetation, excepting occasional patches 
of artemisia and greasewood. The minute crystals of salt which 
cover the surface of the moist, oozy mud, glisten brilliantly in the 
sunlight, and present the appearance of a large sheet of water so 
perfectly, that it is difficult, at times, for one to persuade himself 
that he is not standing on the shore of the lake itself. High rocky 
ridges protrude above the level plain, and resemble great islands 
rising above the bosom of this desert sea. 

The mirage, which frequently occurs, is greater here than I ever 


witnessed elsewhere, distorting objects in the most grotesque man- 


ner, defying all calculation as to their size, shape, or distances, and 


giving rise to optical illusions almost beyond belief. With the ex- 


ception of the two valleys lying at the south end of the lake, the 


country is, as a place of human habitation, entirely worthless. 


There is, however, one valuable use to which it may and perhaps 
will be applied: its extent, and perfectly level surface, would furnish 
a desirable space on which to measure a degree of the meridian. 


120 SALT LAKE CITY—INDIAN TRADING PARTY. 


CHAPTER VII. 


TERMINATION OF THE FIELD-woRK OF 1849—A WINTER AMONG THE 
MORMONS. 


Upon my arrival at Salt Lake City, I found that the camp, under 
Lieutenant Gunnison, wasthen about sixty miles to the southward, 
upon Utah Lake. I accordingly joined him as soon as possible. 
The work, during my absence, had been carried forward by that 
officer with energy, industry, and judgment. 

I had hoped, from the representations which had been made to me 
of the mildness of the two previous winters, that we should be able 
to keep the field the greater part, if not the whole of theseason; but, 
in the latter part of November, the winter set in with great and un- 
usual severity, accompanied by deep snows, which rendered any 
farther prosecution of the work impracticable. I was therefore com- 
pelled to break up my camp, and to seek for winter quarters in the 
city. These were not obtained without some difficulty, as the tide of 
emigration had been so great that houses were very scarce, and not 
a small portion of the inhabitants, among whom was the president 
himself, were forced to lodge portions of their families in wagons. 

Upon terminating the field-work for the season, I despatched 
three men, one of whom was my guide and interpreter, with a small — 
invoice of goods, to trade for horses among the Uintah Utahs, with 
directions to await my orders at Fort Bridger. Reports after- 
ward reached us that a bloody fight had taken place between the 
Sioux and the Yampah Utahs, which latter tribe reside in the 
vicinity of the Uintahs, and great fears were entertained that the 
little party had been cut off by one or the other of the contending 
tribes. Such a calamity, aside from the loss of life, would have 
been of serious consequence to the expedition, as the horses I ex- 
pected to obtain were almost indispensable to the return of the 
party to the States, the number of our animals having been much 
diminished by death and robbery. . 

It may as well be mentioned here, that ‘the party thus des- 
patched subsequently joined me in the spring, as soon as the melt- 
ing of the snows rendered communication with Fort Bridger prac- 


re —————————————————————— EEE 


=a “WNL ASV AHL AO GNA LS id ees 


wr 


area. © on TBs apt ege —s 


a 


AN Aempeoig gue UIT meuLayoy = - Ee rion : oe bs aitine : eres 


PROGRESS OF THE SURVEY IN 1849. 121 


ticable, bringing with them a drove of twenty-five horses. They 
had met with very rough usage from the Indians, having been 
robbed of a number of their horses, beside the whole of what 
remained of their goods, and narrowly escaped with their lives. 

From the report by Lieutenant Gunnison of his operations 
during my absence, I make the following synopsis. 

A thorough exploration was made, with the view of ascertaining 
_ the points for such a base line as would best develop a system of 
triangles embracing both the Salt Lake and Utah valleys. 

A line was selected, and carefully measured by rods constructed 
for the purpose, and tripod stations erected over the termini, 
which were marked by metal points set in wooden posts sunk 
flush with the surface of the ground. The length of the base is 
thirty-one thousand six hundred and eighty feet. 

Fourteen principal triangulation stations were erected, consisting 
of large pyramidal timber tripods, strongly framed, to be covered, 
when required for use, by cotton cloth of different colours, accord- 
ing tothe background. The triangles extended to the south shore 
of Utah Lake, and embraced an area of about eighty by twenty-five 
miles. 

A survey and sounding had been made of the Utah Lake, and 
also of the river connecting it with Salt Lake: this operation re- 
quiring a line to be run of one hundred and twenty-six miles, 
principally by the back angle, with the theodolite. 

Although such a result, from less than two months’ labour, 
would be entirely satisfactory under ordinary circumstances any- 
where, and would reflect credit on the energy and capacity of the 
officer in charge of the work, yet it may be remarked that it would 
be very unfair to judge of it by a comparison with similar results 
obtained in the Eastern States. There, all the accessories to such 
a work, especially water and timber, are abundant, and generally 

at a convenient distance: here, on the contrary, both are very 
scarce and hard to be obtained. All the water, for instance, used 
both for cooking and drinking, that was consumed on the base 
_ Tine, (requiring seven days of incessant labour in its measurement,) 
had to be transported upon mules from the river, which lay a mile 
east of its eastern terminus; and the force employed in the erection 
of most of the triangulation stations had to be supplied in a like 
manner. But the principal difficulty was the scarcity of timber. . 
Wood grows nowhere on the plains ; all the wood used for cooking 
in camp, and all the timber, both for posts on the base line and 


122 WINTER IN SALT LAKE CITY. 


for the construction of the stations, had to be hauled from the 
mountains, in many cases fifteen or twenty miles distant, over a 
rough country without roads. Almost every stick used for this 
purpose cost from twenty to thirty miles’ travel of a six-mule 
team. ‘This, together with the delays of getting into the caiions, 
where alone the timber can be procured, cutting down the trees, 
and hauling them down the gorges by hand to the nearest spots 
accessible to the teams, involved an amount of time and labour 
which must be experienced before it can be appreciated. All this 
had to be done, however, or the prosecution of the work would 
have been impracticable. 

Before leaving the Salt Lake City for Fort Hall, I had engaged 
the services of Albert Carrington, Esq., a member of the Mormon 
community, who was to act as an assistant on the survey. He 
was without experience in the use of instruments; but, being a 
gentleman of liberal education, he soon acquired, under instruction, 
the requisite skill, and, by his zeal, industry, and practical good 
sense, materially aided us in our subsequent operations. He con- 
tinued with the party until the termination of the survey, accom- 
panied it to this city, and has since returned to his mountain home, 
carrying with him the respect and kind wishes of all with whom 
he was associated. 

The winter season in the valley was long and severe. The 
vicinity of so many high mountains rendered the weather ex- 
tremely variable; snows fell constantly upon them, and fre- 
quently to the depth of ten inches in the plains. In many of the 
cafions it accumulated to the depth of fifty feet, filling up the 
passes so rapidly that, in more than one instance, emigrants who 
had been belated in starting from the States, were overtaken by 
the storms in the mountain gorges, and forced to abandon every 
thing, and escape on foot, leaving even their animals to perish in 
the snows. All communication with the world beyond was thus 
effectually cut off; and, as the winter advanced, the gorges became 
more and more impassable, owing to the drifting of the snow into 
them from the projecting peaks. 

We remained thus shut up until the third of April. Our quar- 
ters consisted of a small unfurnished house of unburnt brick or | 
adobe, unplastered, and roofed with boards loosely nailed on, 
which, every time it stormed, admitted so much water as called 
into requisition all the pans and buckets in the establishment to 
receive the numerous little streams which came trickling down 


\ 


MORMON BED-ROOMS. 123 


from every crack and knot-hole. During this season of compara- 
tive inaction, we received from the authorities and citizens of the 
community every kindness that the most warmhearted hospitality 
could dictate; and no effort was spared to render us as comfort- 
able as their own limited means would admit. Indeed, we were 
much better lodged than many of our neighbours; for, as has been 
previously observed, very many families were obliged still to lodge 
wholly or in part in their wagons, which, being covered, served, 
when taken off from the wheels and set upon the ground, to make 
bedrooms, of limited dimensions it is true, but yet exceedingly 
comfortable. Many of these were comparatively large and commo- 


‘dious, and, when carpeted and furnished with a little stove, formed 


an additional apartment or back building to the small cabin, with 
which they frequently communicated by a door. It certainly argued 
a high tone of morals and an habitual observance of good order and 
decorum, to find women and children thus securely slumbering in 
the midst of a large city, with no protection from midnight moles- 
tation other than a wagon-cover of linen and the gis of the law. 
In the very next enclosure to that occupied by our party, a whole 
family of children had no other shelter than one of these wagons, 
where they slept all the winter, literally out of doors, there being 
no communication whatever with the inside of their parents’ house. 

The founding, within the space of three years, of a large and 
flourishing community, upon a spot so remote from the abodes of 
man, so completely shut out by natural barriers from the rest of 
the world, so entirely unconnected by watercourses with either of 
the oceans that wash the shores of this continent—a country offer- 
ing no advantages of inland navigation or of foreign commerce, but, 
on the contrary, isolated by vast uninhabitable deserts, and only 
to be reached by long, painful, and often hazardous journeys by 
land—presents an anomaly so very peculiar, that it deserves more 
than a passing notice. In this young and progressive country of 
ours, where cities grow up in a day, and states spring into exist- 
ence in a year, the successful planting of a colony, where the 
natural advantages have been such as to hold out the promise of 
adequate reward to the projectors, would have excited no surprise ; 
but the success of an enterprise under circumstances so at variance 


with all our preconceived ideas of its probability, may well be con- 


sidered as one of the most remarkable incidents of the present 
age. 
A brief reference to the early history of this people, and to the 


, 224 EARLY HISTORY OF THE MORMONS. 


events and motives which led to their planting such a settlement 
in the midst of a barren wilderness, may not be without interest. 

The City of the Great Salt Lake, the capital of the settlement, — 
was founded in 1847, by a religious community of people known 
among us by the name of Mormons, but who style themselves the 
« Latter-day Saints of the Church of Jesus Christ.’’ It is situated 
in lat. 40° 46’ north, and long. 112° 6’ west, at the foot of the 
western slope of the Wahsatch Mountains, an extensive chain of 
lofty hills, forming a portion of the eastern boundary of what is 
known in our geography as the « Great Basin.” 

The origin of this new religious sect in our country is well 
known, and therefore it will only be necessary to advert to it very 
briefly. It was first organized in 1830, under the auspices of 
Joseph Smith, the founder; and, after a temporary residence in 
Kirtland, Ohio, was removed to Jackson county, Missouri, where 
by divine revelation “the saints’ were directed to build a magnifi- 
cent temple, the pattern of which was to be revealed from on high. 
The corner-stone of this edifice was laid, but the builders were 
eventually driven from the State by an armed mob. They next 
removed to Illinois, where, upon the bank of the Mississippi, 
they built a flourishing city, which they called Nauvoo. They 
lived here until 1844, when they became obnoxious to the inha- 
bitants of that State also, and were finally attacked by an en- 
raged multitude, and their prophet, Joseph Smith, and his brother 
Hyrum, murdered in the jail of Carthage. During the year 1845, 
these persecutions continued; and threats of greater outrages 
being held out, the Mormons found their situation no longer tole- 
rable within the boundaries of that State, and at length, in a solemn 
council, determined to abandon their homes in their city of 
Nauvoo, and to seek, in the wilds of the Western wilderness, a spot 
remote from the habitations of men, where, secure from lawless 
violence, they might worship according to the rites of the new reli- 
gion they had introduced. 

Into the particular causes which led to the expulsion of the Mor- 
mons from Missouri and Illinois it is not the province of this report 
to inquire. The facts have long been before the country, and its 
judgment has been passed upon them; but the results of the per- 
secutions to which they were subjected have been as curious as they 
were wholly unlooked-for. 

The Mormons having resolved to emigrate, preparations for the 
journey were immediately commenced, by hastily and at much 


“HIQON SHI WOT. ALIO YMMV LIVS IVHHD 


me 7 ; 


se ieialieente sctatatine teste tee pe nine errata ee ne Pee hale 


a EL Na it a at ir GN ty SO ata Gi th A NES ith cc, iil 5 Malia alata la ss 


EXODUS FROM ILLINOIS. 125 


sacrifice exchanging such property as they could dispose of for 
animals, wagons, and breadstuffs; and in the beginning of Fe- 
bruary, 1846, a large proportion of the community crossed the 
Mississippi from Nauvoo, and formed a rendezvous near Montrose, 
in Iowa. Here they remained, exposed to intense cold and deep 
snows, until March, when, being joined by several hundred wagons 
and a large number of women and children, they organized their 
company under the guidance of Brigham Young, president of the 
church, and successor of Joseph Smith their founder and seer. 

In their progress westward, through the northern part of Mis- 
souri, they were again driven from that State, by violent threats, 
into the southern borders of Iowa, whence, after much hardship and 
suffering, they reached, in the course of the summer, the banks of 


the Missouri, beyond the limits of the States. Here they enclosed 


land and planted crops, leaving some of their number to reap the 
fruits, which were to be applied to the sustenance of other compa- 
nies, that were to follow as soon as they should be able to provide 
the means. They were about crossing the river to pursue their 
journey westward, when an officer of the United States Government 
presented himself, with a requisition for five hundred men to serve in 
the war with Mexico. This demand, though sudden and unexpected, 
was promptly and patriotically complied with ; but in consequence, 
the expedition was broken up for the season. Those that remained, 
being principally old men, women, and children, prepared to pass 
the winter in the wilds of an Indian country, by cutting hay and 
erecting log and sod huts, and digging as many caves as time 
allowed and their strength enabled them. 

During this winter, owing to the great privations incident to 
such a life, and to the want, in many instances, of the most com- 
mon necessaries, great numbers sickened and died: their cattle, 
too, were stolen by the Indians, or perished by starvation. : 

In the succeeding spring of 1847, the people were again organized 
for their journey ; and on the 8th of April, a pioneer company, con- 
sisting of one hundred and forty-three men, seventy-two wagons, and 
one hundred and seventy-five head of horses, mules, and oxen, with 
rations for six months, agricultural implements and seed-grain, 
manfully set out in search of a home beyond the Rocky Mountains. 

Pursuing their route up the left bank of the Platte, crossing at 
Fort Laramie, and passing over the mountains at the South Pass, 
the advanced guard at length reached the valley of the Great Salt 
Lake, on the 21st of July. On the 24th, the presidency and the 


126 SETTLEMENT OF SALT LAKE VALLEY. 


main body arrived. A piece of ground was selected, consecrated 
by prayer, broken up, and planted; and thus, in 1847, was formed 
the nucleus of what, in 1850, was admitted as a Territory of the 
Union, and which bids fair ere long to present itself at the door of. 
the national legislature for admission as one of the States of the 
confederacy. 

In a short time after the doctieah of the pioneer company, ground 
was surveyed and laid out into streets and squares for a large 
city; a fort or enclosure was erected, of houses made of logs and 
sun-dried brick, opening into a large square, the entrance to which 
was defended by gates, and formed a tolerably secure fortification 
against Indian attacks. In October following, an addition of be- 
tween three and four thousand was made to their number, by the 
emigration of such as had been left behind, and the fort was neces- 
sarily enlarged for their accommodation. Agricultural labours 
were now resumed with renewed spirit; ploughing and planting 
continued throughout the whole winter and until the July follow- 
ing, by which time a line of fence had been constructed, enclosing 
upward of six thousand acres of land, laid down in crops, besides 
a large tract of pasture land. During the winter, and spring, 
the inhabitants were much straitened for food; and game being 
very scarce in the country, they were reduced to the neces- 
sity of digging roots from the ground, and living upon the hides 
of animals which they had previously made use of for roofing their 
cabins, but which were now torn off for food. But this distress 
only continued until the harvest, since which time provisions of all 
kinds have been abundant. 

This year, (1848,) a small grist-mill was erected, and two saw- 
mills nearly completed. The following winter and spring, a settle- 
ment was commenced on the banks of the Weber River, a bold, clear 
stream which breaks through the Wahsatch Mountains, forty miles 
north of the city, and discharges its waters into the Salt Lake. 

Upon Ogden Creek, an affluent of the Weber, a city has since 
(1850) been laid out, and called Ogden City, and is already sur- 
rounded by a flourishing agricultural population. 

In the autumn, another large immigration arrived under the 
president, Brigham Young, which materially added to the strength 
of the colony. Building and agriculture were prosecuted with 
renewed vigour. Numerous settlements continued to be made 
wherever water could be found for irrigation. A handsome coun- 
cil-house was commenced, to be built of red sandstone procured 


ASV WN ACO T= ALT) SS eel ON a a Bia i 
KNAempr org eye wy uewaaxay : 


" 
mie meet 
re 


aan 


ON eee eS 
i : 


CIVIL GOVERNMENT—-STATE OF DESERET. 127 


from the neighbouring mountain, and two grist-mills and three 
saw-mills, added to those already in operation. 'The winter of this 
year was much more severe than the preceding one, and snow fell 
on the plain to the depth of ten inches. 

In the following spring (1849) a settlement was commenced, and 
a small fort built near the mouth of the Timpanogas or Provaux, 
an affluent of Lake Utah, about fifty miles south of the city. 
During this summer, large crops of grain, melons, potatoes, and 
corn were raised, and two more saw-mills erected. 

The colony had now become firmly established, and all fear of 
its ability to sustain itself were, from the overflowing abundance 
of the harvest, set at rest. Nothing could be more natural than 
that the people should turn their attention to the formation 
of a system of civil government. Hitherto they had been under 
the guidance of their ecclesiastical leaders only, and justice had 
- been administered upon principles of equity simply, enforced by 
the government of the church alone. This would answer very 
well while the community remained small, and consisted only of 
those who acknowledged the binding force of spiritual rule in 
matters purely temporal also. But, as the colony increased, it was 
not to be expected that it would continue to consist solely of mem- 
bers of the church, willing to submit to such a jurisdiction, without 
the sanctions of an organized civil government. 

A convention was therefore called «of all the citizens of that 
portion of Upper California lying east of the Sierra Nevada 
mountains, to take mto consideration the propriety of organizing 
a Territorial or State government.”’ 

The convention met at Great Salt Lake City, on the Sth of 
March, 1849, and on the 10th adopted a constitution, which was to 
remain in force until the Congress of the United States should 
otherwise provide for the government of the territory. 

It «‘ordained and established a free and independent govern- 
ment, by the name of the State oF DesERET;”’ fixed the bounda- 
ries of the new State; provided for the election of governor, sena- 
tors, representatives, and judges: all of whom, as well as the 
other officers created by it, were required to take an oath to sup- 
port the constitution of the United States. On the 2d of July, 
the legislature, created by the organic law, met, elected a delegate 
to Congress, and adopted a memorial to that body, in which, among 
other things, they state that «the inhabitants of the State of 
Deseret, in view of their own security, and for the preservation of 


128 CITY OF THE GREAT SALT LAKE. 


the constitutional right of the United States to hold jurisdiction 
there, have organized a provisional State government, under which 
the civil policy of the nation is duly maintained.” «That there 
is now a sufficient number of individuals residing within the State 
of Deseret to support a State government.’’ They therefore asked 
‘«‘that, if consistent with the constitution and usages of the Federal 
Governnient, the constitution accompanying the memorial be rati- 
fied, and that the State of Deseret be admitted into the Union on 
an equal footing with other States’ —«or such other form of civil 
government established, as Congress in its biciizne and magna- 
nimity might award.” 

A constitution and petition for a Territorial organization had 
been previously forwarded to Congress; but in consequence of in- 
formation received afterward, a memorial for a State government 
was substituted in its room. Such is a brief sketch of the origin 
and progress of this colony, and the condition in which we found 
it upon our arrival in August, 1849. 

A city had been laid out upon a magnificent scale, being nearly 
four miles in length and three in breadth; the streets at right 
angles with each other, eight rods or one hundred and thirty-two 
feet wide, with sidewalks of twenty feet; the blocks forty rods 
square, divided into eight lots, each of which contains an acre and 
a-quarter of ground. By an ordinance of the city, each house is 
to be placed twenty feet back from the front line of the lot, the 
intervening space being designed for shrubbery and trees. The 
site for the city is most beautiful: it lies at the western base of 
the Wahsatch Mountains, in a curve formed by the projection 
westward from the main range, of a lofty spur which forms its 
southern boundary. On the west it is washed by the waters of 
the Jordan, while to the southward for twenty-five miles extends 
a broad level plain, watered by several little streams, which, flow- 
ing down from the eastern hills, form the great element of fertility 
and wealth to the community. Through the city itself flows am 
unfailing stream of pure, sweet water, which, by an ingenious mode: 
of irrigation, is made to traverse each side of every street, whence 
it is led into every garden-spot, spreading life, verdure, and beauty 
over what was heretofore a barren waste. On the east and north 
the mountain descends to the plain by steps, which form broad and 
elevated terraces, commanding an extended view of the whole val- 
ley of the Jordan, which is bounded on the west by a range of 


‘“ALIO AMV! LIvs IveHdD WOM SATW AAYHL DNIvas 10H 


ee 


ee 


te 
peed 


CITY OF THE GREAT SALT LAKE. 129 


rugged mountains, stretching far to the southward, and enclosing 
within their embrace the lovely little Lake of Utah. 

On the northern confines of the city, a warm spring issues from 
the base of the mountain, the water of which has been conducted 
by pipes into a commodious bathing-house; while, at the western 
point of the same spur, about three miles distant, another spring 
flows in a bold stream from beneath a perpendicular rock, with a 
temperature too high to admit the insertion of the hand, (128° 
Fahr.) At the base of the hill it forms a little lake, which in the 
autumn and winter is covered with large flocks of waterfowl, at- 
tracted by the genial temperature of the water. 

Beyond the Jordan, on the west, the dry and otherwise barren 
plains support a hardy grass, (called bunch-grass,) which is pecu- 
liar to these regions, requiring but little moisture, very nutritious, 
_ and in sufficient quantities to afford excellent pasturage to nume- 
rous herds of cattle. To the northward, in the low grounds border- 
ing the river, hay in abundance can be procured, although it is 
rather coarse and of an inferior quality. 

The facilities for beautifying this admirable site are manifold. 
The irrigating canals, which flow before every door, furnish abun- 
dance of water for the nourishment of shade-trees, and the open 
space between each building, and the pavement before it, when 
planted with shrubbery and adorned with flowers, will make this 
one of the most lovely spots between the Mississippi and the Pa- 
cific. One of the most unpleasant characteristics of the whole 
country, after leaving the Blue River, is the entire absence of trees 
from the landscape. The weary traveller plods along, exposed to 
the full blaze of one eternal sunshine, day after day, and week after 
week, his eye resting upon naught but interminable plains, bald 
and naked hills, or bold and rugged mountains: the shady grove, 
the babbling brook, the dense and solemn forest, are things un- 
known here; and should he by chance light upon some solitary 
cotton-wood, or pitch his tent amid some stunted willows, the op- 
portunity is hailed with joy, as one of unusual good fortune. The 
studding, therefore, of this beautiful city with noble trees, will 
render it, by contrast with the surrounding regions, a second “ Dia- 
mond of the Desert,” in whose welcome shade, like the solitary 
Sir Kenneth and the princely Ilderim, the pilgrim, wayworn and 
faint, may repose his jaded limbs and dream of the purling brooks 
and waving woodlands he has left a thousand miles behind him. 


The city was estimated to contain about eight thousand inhabit: 
9 


130 PROVISIONAL STATE GOVERNMENT. 


ants, and was divided into numerous wards, each, at the time of 
our visit, enclosed by a substantial fence, for the protection of the 
young crops : as time and leisure will permit, these will be removed, 
and each lot enclosed by itself, as with us. The houses are built, 
principally, of adobe or sun-dried brick, which, when well covered 
with a tight projecting roof, make a warm, comfortable dwelling, 
presenting a very neat appearance. Buildings of a better descrip- 
tion are being introduced, although slowly, owing to the difficulty of 
procuring the requisite lumber, which must always be scarce and 
dear in a country so destitute of timber. 

Upon a square appropriated to the public buildings, an immense 
shed had been erected upon posts, which was capable of containing 
three thousand persons. It was called « The Bowery,” and served 
as a temporary place of worship, until the construction of the Great 
Temple. This latter is to surpass in grandeur of design and gor- 
geousness of decoration all edifices the world has yet seen; and is to 
be eclipsed only by that contemplated in Jackson county, Missouri, 
—to be erected when “the fulness of time shall come,”’ and which 
will constitute the head-quarters or central point, whence light, 
truth, and the only true religion shall radiate to the uttermost 
parts of the earth. A mint was already in operation, from which 
were issued gold coins of the Federal denominations, stamped, 
without assay, from the dust brought from California. | 

The provisional State government, with all the machimery of 
executive, legislative, and judicial functionaries, was in regular 
and harmonious action, under the constitution recently adopted. 
The jurisdiction of the «State of Deseret’? had been extended over 
and was vigorously enforced upon all who came within its borders, 
and justice was equitably administered alike to “‘saint’’ and “ gen- 
tile’’—as they term all who are not of their persuasion. Of the 
truth of this, as far at least as the gentiles were concerned, I soon 
had convincing proof, by finding, one fine morning, some twenty of 
our mules safely secured in the public pound, for trespass upon 
the cornfield of some pious saint; possession was recovered only 
by paying the fine imposed by the magistrate and amply remu- 
nerating the owner for the damage done to his crops. Their 
courts were constantly appealed to by companies of passing emi- 
grants, who, having fallen out by the way, could not agree upon 
the division of their property. ‘The decisions were remarkable. 
for fairness and impartiality, and if not submitted to, were sternly 
enforced by the whole power of the community. Appeals for 


Adsl tol ol io deves. = ab ec) 


KN ACMPROLA 648 WL] WweuLayOy 


UNION OF CHURCH AND STATE. 131 


protection from oppression, by those passing through their midst, 
were not made in vain; and I know of at least one instance in 
which the marshal of the State was despatched, with an adequate — 
force, nearly two hundred miles into the western desert, in pur- 
suit of some miscreants who had stolen off with nearly the whole 
outfit of a party of emigrants. He pursued and brought them 
back to the city, and the plundered property was restored to its 
rightful owner. 

While, however, there are all the exterior evidences of a govern- 
ment strictly temporal, it cannot be concealed that it is so intimately 
blended with the spiritual administration of the church, that it 
would be impossible to separate the one from the other. The 
first civil governor under the constitution of the new State, elected 
by the people, was the president of the church, Brigham Young; 
the lieutenant-governor was his first ecclesiastical counsellor, and 
the secretary of state his second counsellor: these three indi- 
viduals forming together the “ presidency” of the church. The 
bishops of the several wards, who, by virtue of their office in the 
church, had exercised not only a spiritual but a temporal authority 
over the several districts assigned to their charge, were appointed, 
under the civil organization, to be justices of the peace, and were 
supported in the discharge of their duties, not only by the civil 
power, but by the whole spiritual authority of the church also. 
This intimate connection of church and state seems to pervade 
every thing that is done. The supreme power in both being lodged 
in the hands of the same individuals, it is difficult to separate their 
two official characters, and to determine whether in any one in- 
stance they act as spiritual or merely temporal officers. 3 

The establishment of a civil government at all, seems to me to 
have been altogether the result of a foreseen necessity, which it 
was impossible to avoid. As the community grew in numbers and 
importance, it was not to be expected, as has been before remarked, 
that the whole population would always consist solely of members 
of the church, looking up to the presidency, not only as its spiritual 
head, but as the divinely commissioned and inspired source of law 
in temporal matters and policy also. It became necessary, there- 
fore, to provide for the government of the whole, by establishing 
some authority which could not be disputed by any, and would 
exercise a control over them as citizens, whether they were mem- 
bers of the church or not; and which, being acknowledged and 
recognised by the Government of the United States, would be sup- 


ee 


132 CIVIL AND ECCLESIASTICAL REVENUE. 


ported by its laws and upheld by its authority. The civil govern- 
ment, therefore, was wholly precautionary, and only for such gen- 
tiles as might settle among them, the power and authority of the 
church over its members being amply sufficient where they alone | 
were concerned. In the organization of the civil government, 
nothing could be more natural than that, the whole people being 
of one faith, they should choose for functionaries to carry it into 
execution, those to whom they had been in the habit of deferring 
as their inspired guides, and by whom they had been led from a 
land of persecution into this far-off wilderness, which, under their 
lead, was already beginning to blossom like the rose. Hence 
came the insensible blending of the two authorities, the principal 
functionaries of the one holding the same relative positions under 
the other. Thus the bishop, in case of a dispute between two 
members of the church, would interpose his spiritual authority as 
bishop for its adjustment, while in differences between those not sub- 
ject to the spiritual jurisdiction, and who could not be made ame- 
nable to church discipline, he would act in the magisterial capacity 
conferred upon him by the constitution and civil laws of the State. 
Thus the control of the affairs of the colony remained in the 
same hands, whether under church or state organization; and these 
hands were, in a double capacity, those into which the constituents 
had, whether as citizens or as church-members, themselves chosen 
to confide it. 

The revenue of the new State seemed to partake of the same 
double character ; the treasures of the church being freely devoted, 
when necessary, to the promotion of the temporal prosperity of the 
body politic. These are derived from a system of tithing, similar 
to that of the ancient Israelites. Hach person, upon profession of 
his faith, and consequent reception into the bosom of the church, 
is required to pay into “the treasury of the Lord” one-tenth of 
all that he possesses; after which, he pays a tenth of the yearly 
increase of his goods; and in addition contributes one-tenth of his 
time, which is devoted to labour on the public works, such as roads, 
bridges, irrigating canals, or such other objects as the authorities 
may direct. The whole amount thus collected goes into the coffers 
of the church, and is exacted only from its members. A tax is 
also laid upon property as with us, which is levied upon all, both 
‘saint’ and gentile, and which constitutes the revenue of the civil 
government. All goods brought into the city, pay as the price of 
« license, a duty of one per cent., except spirituous liquors, for 


TAXES—PROSPRRITY OF THE COLONY. 1338 


which one-half of the price at which they are sold is demanded: 
the object of this last impost being avowedly to discourage the in- 
troduction of that article among them. It has, indeed, operated 
to a great extent as a prohibition, the importer, to save himself 
from loss, having to double the price at which he could otherwise 
have afforded to sell. The result of this policy was, when we were 
there, to bring up the price of brandy to twelve dollars per gallon, 
of which the authorities took six; and of whisky to eight dollars, _ 
of which they collected four dollars. The circulating medium is 
principally gold of their own coinage, and such foreign gold as is 
brought in by converts from Europe. 

Notwithstanding this heavy, and as it would be to us, insupport- 
able burden upon industry and enterprise, nothing can exceed the 
appearance of prosperity, peaceful harmony, and cheerful content- 
ment that pervaded the whole community. Ever since the first 
year of privation, provisions have been abundant, and want of the 
necessaries and even comforts of life is a thing unknown. A de- 
sign was at one time entertained (more, I believe, as a prospective 
measure than any thing else) to set apart a fund for the purpose 
of erecting a poorhouse; but after strict inquiry, it was found that 
there were in the whole population but two persons who could be 
considered as objects of public charity, and the plan was conse- 
quently abandoned. 

This happy external state, of universally diffused prosperity, is 
commented on by themselves, as an evidence of the smiles of Hea- 
ven and of the special favour of the Deity: but I think it may be 
most clearly accounted for in the admirable discipline and ready 
obedience of a large body of industrious and intelligent men, and 
in the wise councils of prudent and sagacious leaders, producing a 
oneness and concentration of action, the result of which has asto- 
nished even those by whom it has been effected. The happy 
consequences of this system of united and well-directed action, 
under one leading and controlling mind, is most prominently ap- 
parent in the erection of public buildings, opening of roads, the 
construction of bridges, and the preparation of the country for the 
speedy occupation of a large and rapidly growmg population, 
shortly to be still further augmented by an immigration even now 
on their way, from almost every country in Kurope. 

Upon the personal character of the leader of this singular people, 
it may not, perhaps, be proper for me to comment in a communica- 
tion like the present. I may nevertheless be pardoned for saying, 


134 BRIGHAM YOUNG——-TREATMENT OF EMIGRANTS. 


that to me, President Young appeared to be a man of clear, sound 
sense, fully alive to the responsibilities of the station he occupies, 
sincerely devoted to the good name and interests of the people over 
which he presides, sensitively jealous of the least attempt to under- 
value or misrepresent them, and indefatigable in devising ways and 
means for their moral, mental, and physical elevation. He ap- 
peared to possess the unlimited personal and official confidence of 
his people ; while both he and his two counsellors, forming the pre- 
sidency of the church, seemed to have but one object in view, 
the prosperity and peace of the society over which they pre- 
sided. | 

In their dealings with the crowds of emigrants that passed 
through their city, the Mormons were ever fair and upright, taking 
no advantage of the necessitous condition of many, if not most of 
them. They sold them such provisions as they could spare, at 
moderate prices, and such as they themselves paid in their dealings 
with each other. In_the whole of our intercourse with them, 
which lasted rather more than:a year, I cannot refer to a single 
instance of fraud or extortion to which any of the party was sub- 
jected; and I strongly incline to the opinion that the charges that 
have been preferred against them in this respect, arose either 
from interested misrepresentation or erroneous information. I 
certainly never experienced any thing like it in my own case, nor 
did I witness or hear of any instance of it in the case of others, 
while I resided among them. ‘Too many that passed through their 
settlement were disposed to disregard their claim to the land they 
occupied, to ridicule the municipal regulations of their city, and to 
trespass wantonly upon their rights. Such offenders were promptly 
arrested by the authorities, made to pay a severe fine, and in some 
instances were imprisoned or made to labour on the public works; 
a punishment richly merited, and which would have been inflicted 
upon them in any civilized community. In short, these people 
presented the appearance of a quiet, orderly, industrious, and 
well-organized society, as much so as one would meet with in any 
city of the Union, having the rights of personal property as per- 
fectly defined and as religiously respected as with ourselves; 
nothing being farther from their faith or practice than the spirit 
of communism, which has been most erroneously supposed to pre- 
vail among them. The main peculiarity of the people consists in 
their religious tenets, the form and extent of their church govern- 


JOSEPH SMITH—BOOK OF MORMON. 138 


ment, (which is a theocracy,) and in the nature especially of their 
domestic relations. 

With regard to the first of these, it is not my design to give 
more than a brief outline, referring the theological student to a 
treatise on this subject, about, as I understand, to be published by 
Lieutenant Gunnison, who was attached to the party, and who has 


paid especial attention to this subject. 


The claim of the Mormons is, that they constitute the only true 
church now upon the earth, that all other denominations of Chris- 
tians, so called, are out of the true path to heaven, which can only be 
attained through the administration of the ordinances of their 
church, by the «« Melchisedec priesthood.” This, they assert, was re- 
moved from the earth some eighteen hundred years ago, since which 
period, as they insist, no true church has existed, until, in 1826, their 
founder, Joseph Smith, was visited by an angel from heaven. This 
favoured man was instructed by the heavenly messenger in the way 
of truth, and led to a spot where, concealed in a stone box buried in 
the earth, were a number of records, written upon golden plates, and 
in a language called by him the “reformed Egyptian.” From this 
box a portion of the records were taken by the angel and given to 
Joseph, upon whom was also conferred the “power and gift of re- 
velation,’’ by which he was enabled to translate the writing graven 
upon the plates. This he did, and gave the result to the world, as the 
“¢ Book of Mormon.” Joseph, they say, was also ordained to the 
‘‘ Melchisedec priesthood,” with the power of knowledge in all lan- 
guages, the gifts of the Spirit, and the authority of «binding and 
loosing.” He and an associate were constituted apostles to preach 
the «gospel,’’ and to establish among the nations the “church of 
Jesus Christ of the latter-day saints.’ In 1830, a church was 


organized, consisting of six members only, which has since grown 


so as to count its disciples by hundreds of thousands. 

The Bible used by the Protestant Christian world is acknow- 
ledged by them to be of Divine origin and authority, but they as- 
sert that it has been much corrupted and interpolated, so much so 
as to require in part a new translation, which has been accordingly 
completed by their prophet Joseph, directly inspired for the pur- 
pose, and the book is soon to be published. ‘They claim for the 
«‘Book of Mormon” the same Divine origin, and hold it to be equally 
authoritative with our Scriptures as a rule of faith and practice. 
In addition, they have the direct revelations which have heretofore 
been made to the seer, and which are recorded in the «Book of 


136 PRIVATE AND DOMESTIC RELATIONS. 


Doctrines and Covenants ;”’ and they also continue to receive, as in- 
timations of the Divine will, such communications as are now made 
to his successor from time to time, for their guidance, not only in 
matters of faith and doctrine, but in those also of worldly policy 
and the concerns of every-day life. In the gift of miracles, and 
healing of the sick by the laying on of hands by the elders of the 
church, they are firm believers; and I have met more than one 
who has assured me not only that they had been eye-witnesses of — 
the miraculous cures thus performed, but had themselves been the 
subjects of them. : 

The mode of worship is, in its general arrangement, the same 
as that adopted by most Protestant denominations who do not use 
printed ritual; to wit, singing, prayer, and a sermon or exhortation 
from the pulpit. A band of music is stationed behind the choir of 
singers, and not only aids in the devotional services, but regales 
the audience before and after the close of the exercises. 

But it is in their private and domestic relations that this sin- 
gular people exhibit the widest departure from the habits and prac- 
tice of all others denominating themselves Christian. I refer to 
what has been generally termed the “spiritual wife system,” the 
practice of which was charged against them in Illinois, and served 
greatly to prejudice the public mind in that State. It was then, I be- 
lieve, most strenuously denietl by them that any such practice 
prevailed, nor is it now openly avowed, either as a matter sanc- 
tioned by their doctrine or discipline. But that polygamy does 
actually exist among them cannot be concealed from any one of 
the most ordinary observation, who has spent even a short time in 
this community. I heard it proclaimed from the stand, by the 
president of the church himself, that he had the right to take a 
thousand wives, if he thought proper; and he defied any one to 
prove from the Bible that he had not. At the same time, I have 
never known any member of the community to avow that he him- 
self had more than one, although that such was the fact was 
as well known and understood as any fact could be. 

If a man, once married, desires to take him a second helpmate, he 
must first, as with us, obtain the consent of the lady intended, and 
that of her parents or guardians, and afterward the approval of the 
seer or president, without which the matter cannot proceed. The 
woman is then “ sealed’ to him under the solemn sanction of the 
church, and stands, in all respects, in the same relation to the man, 
as the wife that was first married. The union thus formed is con- 


PLURALITY OF WIVES. 137 


sidered a perfectly virtuous and honourable one, and the lady 
maintains, without blemish, the same position in society to which 
she would be entitled were she the sole wife of her husband. In- 
deed, the connection being under the sanction of the only true 
priesthood, is deemed infinitely more sacred and binding than any 
marriage among the gentile world, not only on account of its higher 
and more sacred authority, but inasmuch as it bears directly 


‘ upon the future state of existence of both the man and the woman; 


for it is the doctrine of the church, that no woman can attain to 
celestial glory without the husband, nor can he arrive at full per- 
fection in the next world without at least one wife: and the greater 
the number he is able to take with him, the higher will be his seat 
in the celestial paradise. 

All idea of sensuality, as the motive of such unions, is most in- 
dignantly repudiated; the avowed object being to raise up, as 
rapidly as possible, ««a holy generation to the Lord,” who shall 
build up his kingdom on the earth. Purity of life, in all the do- 
mestic relations, is strenuously inculcated; and they do not hesitate 
to declare, that when they shall obtain the uncontrolled power of 
making their own civil laws, (which will be when they are admitted 
as one of the States of the Union,} they will punish the departure 
from chastity in the severest manner, even by death. 

As the seer or president alone possesses the power to approve 
of these unions, so also he alone can absolve the parties from their 
bonds, should circumstances in his Judgment render it at any time 
either expedient or necessary. It may easily be perceived, then, 
what a tremendous influence the possession of such a power must 
give to him who holds it, and how great must be the prudence, 
firmness, sagacity, and wisdom required in one who thus stands in 
the relation of confidential adviser, as well as of civil and eccle- 
siastical ruler, over this singularly constituted community. 

Upon the practical working of this system of plurality of wives, 
I can hardly be expected to express more than a mere opinion. 
Being myself an “outsider” and a “gentile,” it is not to be sup- 
posed that I should have been permitted to view more than the 
surface of what is in fact as yet but an experiment, the details of 
which are sedulously veiled from public view. So far, however, 
as my intercourse with the inhabitants afforded me an opportunity 
of judging, its practical operation was quite different from what I 
had anticipated. Peace, harmony, and cheerfulness seemed to 
prevail, where my preconceived notions led me to look for nothing 


~~. ll ene 


138 PRACTICAL WORKING OF THE SYSTEM. 


but the exhibition of petty jealousies, envy, bickerings, and strife. 
Confidence and sisterly affection among the different members of 
the family seemed pre-eminently conspicuous, and friendly inter- 
course among neighbours, with balls, parties, and merry-makings 
at each others’ houses, formed a prominent and agreeable feature 
of the society. In these friendly reunions, the president, with his 
numerous family, mingled freely, and was ever an honoured and 
welcome guest, tempering by his presence the exuberant hilarity 
of the young, and not unfrequently closing with devotional exer- 
cises the gayety of a happy evening. 

There are many other curious points contained in their religious 
creed, but it is not my purpose here to write a theological treatise 
upon their views. The effect of the system, as may be well sup- 
posed, is to render the people in a high degree separate and pecu- 
liar; and to prevent, not only all amalgamation, but even any 
intimate association, with other communities. 

To this irreconcilable difference, not in speculative opinions only, 
but in habits, manners, and customs necessarily growing out of 
them, may, I think, in a great measure, be attributed the bitter 
hostility of the people among whom they formerly dwelt, and 
which resulted in their forcible expulsion. The same causes of 
social incompatibility which existed then, exist now, and in much 
greater strength—the community being freed from the pressure 
of public opinion that then surrounded them; and, although the 
freest toleration is (no doubt sincerely) proclaimed toward any who 
may choose to settle among them, yet I do not see howit is pos- 
sible for the members of any other Christian societies, all of which 
are theoretically and practically opposed to their views, to exist 
among them, without constant collision, jealousy, and.strife. The 
result, therefore, must be the establishment here of a people of 
one faith, the fundamental principles of whose civil government — 
will, under the lead of the ecclesiastical hierarchy, be framed to 
accord with that faith, to build up and support it, and to exclude 
from all participation in its administration every element that 
does not fully coincide with its requirements. When what is now 
but a Territory shall have become a sovereign State, with the un- 
controlled power of making its own laws, this will undoubtedly be 
done; and we shall then see in our midst a State as different from 
the rest of the Union in faith, manners, and customs, as it is 
widely separated by the vast plains and inhospitable deserts that 
surround it. That such a State will soon be formed, no refiecting 


PROSPECTIVE STATE OF DESERET. 139 


man can well doubt, who has witnessed the indomitable energy, the 
unity and concentration of action, together with the enthusiastic 
spirit of proselytism which seems to possess the entire Mormon com- 
munity. Their zeal for increasing their sect has already filled the 
world with their missionaries; and has, within the space of four 
years, and in defiance of obstacles that would have appalled most 
ordinary adventurers, collected a population of some twenty thou- 
sand souls, all breathing the same spirit, animated by the same 
hope, bound by the same views, and unitedly engaged, heart and 
hand, in providing means by which converts to the faith may be 
transported from all parts of the world to this great head-quarters 
of the church, «the fountain where truth flows from the lips of the 
prophet of God, and where true liberty can only be enjoyed by the 
saints.” 

A large and constantly increasing fund has been created 
among them, called «The Perpetual Emigration Fund,” which is 
devoted exclusively to this object, and receives liberal contributions 
from the “saints,” both in this country and in Europe; it being the 
authorized teaching, all over the world, that it is as much a duty 
binding on every “saint” «to build up the valleys of the moun- 
tains,” by assisting forward those brethren who are too poor to 
provide an outfit for themselves, as it is to be baptized for the re- 
mission of sins. The effects of this widely diffused spirit of propa- 
gandism are already seen in the number of converts that have 
been made in most of the countries of Europe, as well as in the 
Sandwich Islands, and even here in our own country, with all of 
whom it is made a cardinal point to “gather to the mountains.” 

Measures are being taken to open a southern route, by which the 
converts coming from abroad may cross the Isthmus of Panama, 
and, landing at San Diego, may thence reach the land of promise 
by a comparatively short and easy transit, without being subjected 
to the hazard of a sickly voyage up the Mississippi, or to the 
tedious and expensive journey across the plains. In the mean while, 
preparations are industriously making in the valley for the recep- 
tion and immediate accommodation of the coming tide, by the 
building of houses, sowing large quantities of grain, the erection 
of mills, the establishment of manufactures, the importation of 
labour-saving machinery, and the establishment upon a solid basis 
of the means of education. The manifest object of these harmo- 
niously concerted movements is to concentrate, as speedily as 


practicable, in «the valley of the mountains,” a number sufficiently 


SSS 


140 SOIL—WATER—IRRIGATION. 


great to entitle the present Territory of Utah to demand from the 
General Government admission into the Union as one of the sove- 
reign States of the confederacy, and thus to secure to themselves 
unmolested the right to carry out in practice the peculiar princi- . 
ples of their creed. That their wishes in this respect will be 
shortly realized may be considered certain. 

Let us now look for a moment at the sources which can be made 
available for the sustenance of a population so numerous as it is 
thus confidently anticipated will ere long be congregated within 
the limits of the «Basin State.” Situated so far inland, without 
water communication with any part of the continent, and isolated 
by the very nature of the surrounding regions, it will readily be seen 
that the new State must necessarily depend, in a great measure, for 
its support, upon means within itself. Agriculture and the rais- 
ing of stock must therefore be the principal basis of its pros- 
perity. For both these purposes the country which they have set- 
tled is, fortunately, well adapted. The land available for the first 
of these objects, though limited in extent when compared with the 
vast deserts which intervene, is still ample for the support of a large, 
though not very dense population. Owing to the almost total ab- 
sence of rain, from May to October, the dependence of the farmer 
must be entirely upon irrigation. ‘The means for this are supplied 
from the reservoirs of snow which accumulate in the gorges of the 
mountains, furnishing, during the whole of the summer, abundant 
and never-failing streams, which assume in some instances the 
character of rivers of considerable magnitude. 

The soil, formed chiefly from the disintegration of the feldspa- 
thic rock, mixed with detritus of the limestone, of which the 
mountains are principally composed, is of the most fertile cha- 
racter. Owing to its loose and porous texture, it absorbs water 
very readily and in large quantities. Consequently, the streams 
which come rushing down the mountain-sides, when they reach the 


‘plain below, begin to dwindle into insignificant rivulets, and soon 


sink and are entirely lost. Many never reach the base of the 
mountain at all, being absorbed by the soil; and even in the islands 
of the lake there are to be found, near the summits, roaring tor- 
rents, which, ere making half the descent of the mountain, so com- 
pletely disappear as to leave not even a dry bed or channel to 
show they had ever reached the water below. Cultivation is there- 
fore circumscribed within very narrow limits, being generally re- 
stricted to a strip of from one to two miles wide, along the base 


PRODUCTIONS—UTAH VALLEY. 141 


of the mountains, beyond which the water does not reach. The 
extensive plains between the mountain ranges, although composed 
of soil nearly equal in fertility, are at present useless for the pur- 
poses of agriculture, from the want of water. The smallness of 
the area suitable for cultivation is, however, compensated by the 
prodigious productiveness of the soil, which, together with the cli- 
mate, is peculiarly favourable to the growth of wheat, barley, oats, 
and all the cereal grains. I brought with me, for distribution, a 
portion of a crop of wheat, which had produced, upon three and one- 
half acres of ground, the enormous yield of one hundred and eighty 
bushels, from a single bushel of seed. In situations peculiarly 
favourable for watering, the average yield of all lands properly cul- 
tivated may be very safely estimated at forty bushels. Maize, or In- 
dian corn, has not as yet proved so successful, owing to the early 
frosts occasioned by the vicinity of the mountains; but beets, tur- 
nips, melons, and especially potatoes, exceed in increase even the 
most sanguine anticipations. The quality of the latter is fully 
equal, if not superior, to the best Nova Scotia varieties. 

On the eastern side of the Salt Lake Valley, the land susceptible 
of irrigation stretches along the western base of the Wahsatch 
Mountains, from about eighty miles north of Salt Lake City to 
about sixty south of it, the latter portion embracing, toward its 
terminus, the fertile valley of Lake Utah. This is a beautiful 
sheet of pure fresh water, thirty miles in length, and about ten in 
breadth, surrounded on three sides by rugged mountains and 
lofty hills, with a broad grassy valley sloping to the water’s edge, 
opening to the northward. Through this opening flows the river 
Jordan, by which its waters are discharged into the Great Salt 
Lake. The lake abounds in fine fish, principally speckled trout, 
of great size and exquisite flavour, which afford sustenance to nu- 
merous small bands of Utahs. | 
The Jordan, in its passage, cuts through a cross range of moun- 

tains that divides the two valleys, making a deep cajion, in which 
are rapids. At most seasons of the year a skiff can be safely 
floated down these boiling waters, if managed with sufficient skill 
to avoid striking the projecting rocks. The fall continues abrupt 
for one mile, and the river could here be led along the escarpment 
of the western hills as far as to a point opposite the mouth of the 
Little Cotton-wood, and thence on a curve to Spring Point, at 
the north end of the Oquirrh Mountain, thus probably bringing 
under irrigation about eighty square miles of fertile land. 


142 PROGRESS OF NEW SETTLEMENTS. 


Near the eastern shores of Lake Utah, a site for a city has been 
selected on the left bank of the Provaux or Timpanogas River, an 
affluent of the lake, which is to be called Provauz City. From 
Ogden City on the north, all the way to this latter «Stake of 
Zion,” the base of the Wahsatch range is studded with flourishing 
farms, wherever a little stream flows down the mountain-side with 
water sufficient’ for irrigating purposes; while in the gorges and 
cafions of the mountain are erected the saw and grist mills. Of 
the former, sixteen, and of the latter, eleven have been completed, 
and others are in the process of erection. 

To the south of Lake Utah, on one of its tributaries, another 
city has been founded, called Paysan, and a hundred and thirty 
miles farther, on the road to California, another, named Mantz, in 
what is called San Pete Valley. Still farther south, near Little 
Salt Lake, two hundred and fifty miles south of the city, a fourth, . 
called Cedar City, has been laid out, in a spot possessing the ad- 
vantages of excellent soil and water, plenty of wood, iron ore, and 
alum, with some prospect of coal. It is the ultimate object of 
the Mormons, by means of stations, wherever the nature of the 
country will admit of their settling in numbers sufficient for self- 
defence, to establish a line of communication with the Pacific, so as 
to afford aid to their brethren coming from abroad, while on their 
pilgrimage to the land of promise. These stations will gradually 
become connected by farms and smaller settlements wherever 
practicable, until the greater part of the way will exhibit one long 
line of cultivated fields from the Mormon capital to San Diego. 

The mode adopted for the founding of a new town is peculiar 
and highly characteristic. An expedition is first sent out to ex- 
plore the cowntry, with a view to the selection of such points as, 
from their natural’ advantages, offer facilities for a settlement. 
These being duly reported to the authorities, an elder of the 
church is appointed to preside over the little band designated to . 
make the first improvement. ‘This company is composed partly 
of volunteers and partly of such as are selected by the presidency, 
due regard being had to a proper intermixture of mechanical 
artisans, to render the expedition independent of all aid from 
without. In this way the settlement at San Pete was begun, 
sixty families 1eaving in a body, under one of the high officers of 
the church, and that in the month of October, undergoing all the 
rigours of cold and snow, to establish another “stake’’ in the wil- 
derness. In December of the following year, another expedition, 


MANUFACTURES—EDUCATION. 143 


similarly composed and commanded, succeeded, with one hundred 
and thirty men and families, in planting the settlement at Little 
Salt Lake, which is represented as being now in a very flourishing 
condition. The succeeding March, a third party, with a hundred 
and fifty wagons, left the capital for the purpose of establishing a 
settlement in the southern part of California. It was to be 
situated at no great distance from San Diego, and near Williams’s 
ryanche and Cahone Pass, between which and Little Salt Lake it is 
designed to establish other settlements as speedily as possible. By 
means of these successive places of refreshment, the incoming 
emigration from the Pacific will be enabled to “go from strength 
to strength” till they reach the Zion of their hopes. 

At Salt Lake City itself, energetic measures are being taken for 
opening a woollen factory, the raw material being furnished from 
sheep raised in the valley, to the grazing of which the mountain 
slopes are admirably adapted, and whose production has already 
attracted the attention of this energetic and far-seeing people. A 


‘pottery for the manufacture of earthenware is completed; and 


cutlery establishments have been successfully commenced. Ex- 
tensive arrangements are going forward for raising the sugar-beet, 
which, under such favourable circumstances, cannot but prove suc- 
cessful; and ere long it is confidently anticipated that a sufficient 
quantity of sugar will be manufactured from it to meet all their 
wants. At present they are supplied with this article and other 
groceries, as well as with dry-goods and clothing, from extensive 
stocks brought in by enterprising merchants from the States; but 
the policy of the people is to provide for their own wants by their 
own skill and industry, and to dispense, as much as possible, with 
the products of the labour of others. 

While all these exertions are making for the physical develop- 
ment of a new empire among the mountains, the mental elevation 


of the people by education has been by no means lost sight of. - 


Liberal appropriations of land and money have been made for the 
establishment of an university, the grounds. for which are laid 
out and enclosed, being situated on one of the terraces of the 
mountain overlooking the city. A normal school, designed for 
the education of those who desire to become teachers, is already 
in successful operation. School-houses have been built in most 
of the districts, both in the city and country, which are attended 
by old as well as young, and every effort is made to advance the 
mental improvement of the people. 


ee ie 


144 PUBLIC SENTIMENT AMONG THE MORMONS. 


When it is remembered that within the space of four years 
this country was but a wild and dreary wilderness, where the howl 
of the wolf and the yell of the miserable Indian alone awoke the 
echoes of the mountains, and where the bear, the deer, and the 
antelope roamed securely over what is now a compact and popu- 
lous city ; that the physical obstacles to the occupation of a region 
sO unpromising were sufficient to discourage the most sanguine 
imagination and to appal the stoutest heart,—the mind is filled 
with wonder at witnessing the immense results which have been 
accomplished in so short a time, and from a nega apparently 
So insignificant. 

Apprehensions have been entertained as to the expediency of 
giving any countenance to the founding, in our midst, of an asso- 
ciation of men so peculiar in views, and so distinct in principles, 
manners, and customs, from the rest of the American people. 
Serious doubts, too, have been expressed in regard to the policy 
of appointing Mormons to offices of high trust in the administra- 
tion of the affairs of the newly-erected territory; and direct 
charges have been widely published, seriously affecting the patriot- 
ism and personal reputation of the Mormon leaders, as well as 
the loyal feelings of the people toward the General Government. 
Such doubts and apprehensions are, in my judgment, totally 
groundless, and the charges I believe to be either based upon pre- 
judice or to have grown out of a want of accurate information. 
A residence of a year in the midst of the Mormon community, 
during the greater part of which period I was in constant inter- 
course with both rulers and people, afforded much bP a for 
ascertaining the real facts of the case. 

That a deep and abiding resentment of injuries received and 
wrongs endured in Missouri and. Illinois pervades the whole Mor- 
mon community, is perfectly true; and that among many of the 
less informed, and, I regret to add, some even whose intelligence 
and education ought to have enabled them to form more correct 
opinions, this exasperation has extended itself to the General Go- 
vernment, because of its refusal to interpose for their protection 
at the time of these difficulties, is also true; but, from all that I 
saw and heard, I deem it but simple justice to say, that notwith- 
standing these causes of irritation, a more loyal and patriotic peo- 
ple cannot be found within the limits of the Union. This, I think, 
was emphatical!y shown in the promptitude and cheerfulness with 
which they responded to the call of the Government to furnish a 


’ 


LOYALTY OF RULERS AND PEOPLE. 145 


battalion for service during the Mexican war. While in the heart 
of an Indian country, and on the eve of a long and uncertain pil- 
grimage into an unknown wilderness, they were suddenly called 
upon to surrender five hundred of their best men to the hazards 
of a hostile campaign, and to the exposure and vicissitudes of a 
march of two thousand miles across trackless deserts and burning 
plains, to fight the battles of their country. Their peculiar cir- 
cumstances presented almost insuperable objections to a compli- 
ance with the requisition, yet not the slightest hesitation was 
evinced. «You shall have your battalion at once,” was the reply 
of President Young, «if it has to be a class of our elders ;” and 
in three days the force, recruited principally among fathers of 
families, was raised and ready to march. Here certainly was no 
2 ee of a lack of patriotism.* 


* The following extract from a sermon of Brigham Young to his people will, I 
think, confirm the correctness of my views as to the sentiments of the Mormon 


» leaders, at that time, on this subject :-— 


«‘T want to say to every man, the constitution of the United States, as formed 
by our fathers, was dictated, was revealed, was put into their hearts by the Al- 
mighty, who sits enthroned.in the midst of the heavens; although unknown to 
them, it was dictated by the revelations of Jesus Christ, and I tell you, in the name 
of Jesus Christ, it is as good as I could ever ask for.” <‘‘I say unto you, magnify 
the laws. There is no law in the United States, or in the constitution, but I am 
ready to make honourable.” 

Many more expressions of a like character might be quoted, but the above 
are sufficient to show what were the opinions of the rulers. 

The following language, used by General D. H. Wells, at the celebration of the 
fourth anniversary of the advent of the Mormons into the Valley, will show, I 
think, what was the feeling of the people :— 

«It has been thought by some, that this people, abused, maltreated, insulted, 
robbed, plundered, murdered, and finally disfranchised and expatriated, would 
naturally feel reluctant to again unite their destiny with the American republic.” 
* * X «No wonder that it was thought by some that we would not again submit 
ourselves (even while we were yet scorned and ridiculed) to return to our allegi- 
ance to our native country. Remember, that it was by the act of our country, 
not ours, that we were expatriated; and then consider the opportunity we had 
of forming other ties. Let this pass, while we lift the veil and show the policy 
which dictated us. That country, that constitution, those institutions, were all 
ours; they are still ours. Our fathers were heroes of the Revolution. Under the 
master spirits of an Adams, a Jefferson, and a Washington, they declared and 
maintained their independence; and, under the guidance of the Spirit of truth, 
they fulfilled their mission whereunto they were sent from the presence of the 
Father. Because demagogues have arisen and seized the reins of power, should 
we relinquish our interest in that country made dear to us by every tie of asso- 
ciation and consanguinity?” * * * ‘Those who have indulged such sentiments 
concerning us, have not read Mormonism aright; for never, no never, will we 


146 BRIGHAM YOUNG. 


Whether in the pulpit, in public addresses,,in official documents, 
or in private intercourse, the same spirit of lofty patriotism 
seemed to pervade their whole community. At the same time, it 
should not be concealed that a stern determination exists among 
them to submit to no repetition of the outrages to which they were 
subjected in Illinois and Missouri; but, on the contrary, to resist 
by force and to the last extremity, from whatever quarter, any 
such interference with what they consider their civil and religious 
rights, guarantied to them, as to other citizens, by the constitution 
of the United States. Vain-glorious vaunts may indeed have been 
sometimes made by individuals whose knowledge and judgment 
were not equal to their religious zeal, as to the ability of the com- 
munity to maintain itself in the fastnesses of the mountains, even 
against the military forces of the Government; but we know that 
there are in every society men whose valour is ever great in pro- 
portion to the remoteness of the danger. I have no idea that any 
such collision was ever seriously anticipated. 

Upon the action of the Executive in the appointment of the of- 
ficers within the newly-created Territory, it does not become me to 
offer other than a very diffident opinion. Yet the opportunities of 
information to which allusion has already been made, may perhaps 
justify me in presenting the result of my own observations upon 
this subject. With all due deference, then, I feel constrained to 
say, that in my opinion the appointment of the president of the 
Mormon church, and head of the Mormon community, in preference 
to any other person, to the high office of Governor of the Terri- 
tory, independent of its political bearings, with which I have 
nothing to do, was a measure dictated alike by justice and by sound 
policy. Intimately connected with them from their exodus from 
Illinois, this man has been indeed their Moses, leading them 
through the wilderness to a remote and unknown land, where they 
have since set up their tabernacle, and where they are now building 
their temple. Resolute in danger, firm and sagacious in council, » 
prompt and energetic in emergency, and enthusiastically devoted 
to the honour and interests of his people, he had won their unlimited 


desert our country’s cause; never will we be found arrayed by the side of her 

enemies, although she herself may cherish them in her own bosom. Although she 

may launch forth the thunderbolts of war, which may return and spend their fury 

upon her own head, never, no never, will we permit the weakness of human na- 

ture to triumph over our love of country, our devotion to her institutions, handed 

down to us by our honoured sires, made dear by a thousand tender recollections.” 
Such, surely, is neither the language nor the spirit of a disloyal people. 


BRIGHAM YOUNG. 147 


confidence, esteem, and veneration, and held an unrivalled place in 
their hearts. Upon the establishment of the provisional govern- 
ment, he had been unanimously chosen as their highest civil 
magistrate, and even before his appointment by the President, he 
combined in his own person the triple character of confidential ad- 
viser, temporal ruler, and prophet of God. Intimately acquainted 
with their character, capacities, wants, and weaknesses; identified 
now with their prosperity, as he had formerly shared to the full in 
their adversity and sorrows; honoured, trusted, the whole wealth 
of the community placed in his hands, for the advancement both 
of the spiritual and temporal interests of the infant settlement, he 
was, surely, of all others, the man best fitted to preside, under the 
auspices of the General Government, over a colony of which he 
may justly be said to have been the founder. No other man could 
have so entirely secured the confidence of the people; and this se- 
lection by the Executive of the man of their choice, besides being 
highly gratifying to them, is recognised as an assurance that they 
shall hereafter receive at the hands of the General Government 
that justice and consideration to which they are entitled. Their 
confident hope now is that, no longer fugitives and outlaws, but 
dwelling beneath the broad shadow of the national egis, they will 
be subject no more to the violence and outrage which drove them 
to seek a secure habitation in this far distant wilderness. 

As to the imputations that have been made against the personal 
character of the governor, I feel confident they are without founda- 
tion. Whatever opinion may be entertained of his pretensions 
to the character of an inspired prophet, or of his views and prac- 
tice on the subject of polygamy, his personal reputation I believe 
to be above reproach. Certain it is that the most entire confidence 
is felt in his integrity, personal, official, and pecuniary, on the part 
of those to whom a long and intimate association, and in the most 
trying emergencies, have afforded every possible opportunity of 
forming a just and accurate judgment of his true character. 

From all I saw and heard, I am firmly of opinion that the ap- 
pointment of any other man to the office of governor would have 
been regarded by the whole people, not only as a sanction, but as 
in some sort a renewal, on the part of the General Government, of 
that series of persecutions to which they had already been sub- 
jected, and would have operated to create distrust and suspicion 
in minds prepared to hail with joy the admission of the new Terri- 
tory to the protection of the supreme government. 


148 NATIVE TRIBES. 


The native tribes with whom we came in contact in the valley 
were the most degraded and the lowest in the scale of being of 
any I had ever seen. ‘They consisted of the «root-diggers,” a 
class of Indians which seemed to be composed of outcasts from 
their respective tribes, subsisting chiefly upon roots dug from the 
ground, and the seeds of various plants indigenous to the soil, which 
they grind into a kind of flour between two flat stones. Lizards 
and crickets also form a portion of their food. At certain seasons 
of the year they obtain, from the tributaries of both the Salt Lake 
and Lake Utah, a considerable quantity of fish, which they take 
in weirs or traps, constructed of willow-bushes. Those that we 
saw were branches from the Shoshonees or Snakes, and from the 
large and warlike tribe of Utahs, which latter inhabit a large 
tract of country to the southward. ‘They are known among the 
traders by the designation of «‘snake-diggers,”’ and « Utes;”’ those 
of the latter tribe, which inhabit the vicinity of the lakes and 
streams and live chiefly on fish, being distinguished by the name of 
«« Pah Utahs,”’ or «« Pah Utes,’”’—the word Pah, in their language, 
signifying water. 

While engaged in the survey of the Utah Valley, we were no 
little annoyed by numbers of the latter tribe, who hung around the 
camp, crowding around the cook-fires, more like hungry dogs than 
human beings, eagerly watching for the least scrap that might be 
thrown away, which they devoured with avidity and without the 
least preparation. The herdsmen also complained that their cattle 
were frequently scattered, and that notwithstanding their utmost 
vigilance, several of them had unaccountably disappeared and were 
lost. One morning, a fine fat ox came into camp with an arrow 
buried in his side, which perfectly accounted for the disappearance 
of the others. 

After the party left Lake Utah for winter quarters in Salt Lake 
City, the Indians became more insolent, boasting of what they had 
done—driving off the stock of the inhabitants in the southern set- — 
tlements, resisting all attempts to recover them, and finally firing 
upon the people themselves, as they issued from their little stockade 
to attend to their ordinary occupations. Under these circum- 
stances, the settlers in the Utah Valley applied to the supreme 
government, at Salt Lake City, for counsel as to the proper course 
of action. The president was at first extremely averse to the 
adoption of harsh measures; but, after several conciliatory over- 
tures had been resorted to in vain, he very properly determined to 


HVA LadOd NE WeOdLVid NOWAOQS GH YAONA SWANOSIMd NVICNE HYG 
JN APMpED IY GZ YY | UP uiss AY 


ARUP 4, 
op hiphner 


Po; ial ES 
IED lowe, 


ap eh area 
" mara rbaod 


Patan 


St! 


MORMON TROUBLES WITH THE UTAHS. 149 


put a stop, by force, to further aggressions, which, if not resisted, 
could only end in the total destruction of the colony. Before 
coming to this decision, the authorities called upon me to con- 
sult as to the policy of the measure, and to, request the expres- 
sion of my opinion as to what view the Government of the United 
States might be expected to take of it. Knowing, as I did, most 
of the circumstances, and feeling convinced that some action of 
the kind would ultimately have to be resorted to, as the forbear- 
ance already shown had been only attributed to weakness and 
cowardice, and had served but to encourage further and bolder 
outrages, I did not hesitate to say to them that, in my judgment, 
the contemplated expedition against these savage marauders was a 
measure not only of good policy, but one of absolute necessity and 
self-preservation. I knew the leader of the Indians to be a crafty 
and blood-thirsty savage, who had been already guilty of several 
murders, and had openly threatened that he would kill every white 
man that he found alone upon the prairies. In addition to this, I 
was convinced that the completion of the yet unfinished survey of 
the Utah Valley, the coming season, must otherwise be attended 
with serious difficulty, if not actual hazard, and would involve the 
necessity of a largely increased and armed escort for its protec- 
tion. Such being the circumstances, the course proposed could 
not but meet my entire approval. 

A force of one hundred men was accordingly organized, and, upon 
the application of President Young, leave was given to Lieutenant 
Howland, of the Mounted Rifles, then on duty with my command, 
to accompany the expedition as its adjutant: such assistance also 
was furnished as it was in my power to afford, consisting of arms, 
tents, camp-equipage, and ammunition. 

The expedition was completely successful. The Indians fought 
very bravely, but were finally routed, some forty of them killed, 
and as many more taken prisoners; the latter, consisting princi- 
pally of women and children, were carried to the city and distri- 
buted among the inhabitants, for the purpose of weaning them from 
their savage pursuits, and bringing them up in the habits of civil- 
ized and Christian life. The experiment, however, did not succeed 
as was anticipated, most of the prisoners epcapme upon the very 
first opportunity. 

On the 22d of February, about three P. M., a slight shock of 
an earthquake was felt in the southern part of the city, the vibra- 


150 ZOOLOGICAL SPECIMENS—THERMAL WATERS. 


tions being sufficient to shake plates from the shelves and to dis- 
turb milk in the pans. 

Advantage was taken of the confinement of the party to winter 
quarters to observe for the latitude, to arrange and plot the notes 
of the survey as far as it had advanced, and to collect and pre- 
pare specimens of the zoology of the valley. These specimens 
have since been classified and arranged with characteristic ability 
by Professor Spencer F. Baird, of the Smithsonian Institution, 
whose report on that subject is hereto appended. Specimens of 
the different thermal waters, also, were collected and brought 
safely as far as Pittsburgh; but, in their transportation thence by 
the express line, most of the vessels containing them were unfor- 
tunately broken, and their contents lost. This was a subject 
of much regret, as interesting results had been anticipated from 
the analysis. Such as escaped destruction have been carefully 
analyzed by Dr. L. D. Gale, of Washington, and the results will be 
found in Appendix F. 

During the winter, a large boat was built for the survey of 
the Salt Lake. This was an achievement of no little difficulty, 
as almost every stick of timber used in the construction had to 
be procured from the cajions of the mountains, piece by piece; and 
the planking, although of the best material the country afforded, 
was so “shaky” and liable to split and crack, that it was totally 
unfit for the purpose. Had time permitted, it had been my pur- 
pose to procure, before setting out, a couple of Francis’s metallic 
life-boats for this service, which would have saved much time and 
labour. The experience of the exploring expedition to the Dead 
Sea has fully proved the entire fitness of these boats for service 
of this nature; and the ease with which they can be transported 
in sections, and. be put together for instant use, will doubtless ren- 
der them hereafter an indispensable part of the equipment for 
every exploration of a similar character. Where the use of wagons 
is practicable, these boats can readily be mounted on wheels and 
made to answer the purposes of a wagon-box; and where this is 
not the case, their arrangement into sections will allow of their 
being packed and transported on the backs of mules with but little 
inconvenience. 


Stine aici 


% © 
Dini gD OSH 


BARON LA HONTAN. 151 


CHAPTER VIII. 


EARLY KNOWLEDGE OF THE EXISTENCE OF A BODY OF SALT WATER 
IN THIS REGION, BY BARON LA HONTAN.—SURVEY OF THE GREAT 
SALT LAKE. 


THE opening of the spring at length enabling us to prepare 
for a renewal of active operations in the field, the opportunity 
was eagerly embraced, since upon the completion of the survey 
before the setting in of cold weather depended the return of the 
party to their homes before the recurrence of winter. 

The season was now approaching when it would become our 
duty to enter upon a critical examination of this interesting and 
hitherto almost unknown region, and the remarkable body of 
water to which it is indebted for so much of the interest which 
attaches to it. It may not, therefore, be deemed inappropriate to 
look back and see what ideas prevailed in regard to it durimg the 
infant period of our national geography. 

The existence of a large lake of salt’ water somewhere amid the 
wilds west of the Rocky Mountains seems to have been known 
vaguely as long as one hundred and fifty years since. As early as 
May, 1689, the Baron La Hontan, “lord-lieutenant of the French 
colony at Placentia in New Foundland,” wrote an account of dis- 
coveries in this region, which was published in the English lan- 
guage in 1735. 

In the letter, which is dated at «« Missilimakinac,” he gives «an 
account of the author’s departure from and return to Missilimaki- 
nac; a description of the Bay of Puants, and its villages; an 
ample description of the beavers, followed by the journal of a 
remarkable voyage upon the Long River, and a map of the adja- 
cent country.” | 

Leaving Mackinaw, he passed into Green Bay, which he calls 
«the Bay of Pouteouotamis,’”’ and arrived at the mouth of the 
Fox River, which he describes as «a little deep sort of a river 
which disembogues at a place where the water of the lake swells 
three feet high in twelve hours, and decreases as much in the 
same compass of time.” 


152 BARON LA HONTAN—LONG RIVER. 


«¢ The village of the Sakis, Pouteouatamis, and some Malominis, 
are seated on the side of that river, and the Jesuits have a house 
or college built upon it.’ Ascending the Fox River, called «the 
river of Puants,’’ he came to a village of « Kikapous, which 
stands on the brink of a little lake in which the savages fish great 
quantities of pikes and gudgeons.’”’ (Lake Winnebago 2) 

Still ascending the river, he passed through the «little lake of 
the Malominis,” the sides of which «‘are covered with a sort of 
oats which grow in tufts, with a small stalk, and of which the 
savages reap plentiful crops,” and at length arrived at the land 
carriage of Ouisconsinc, which we finished in two days; that is, 
we left the river Puants, and transported our canoes and baggage 
to the river Ouisconsince, which is not above three-quarters of a 
league distant or thereabouts.’’ Descending the Wisconsin, in 
four days he reached its mouth, and landed on an island in the 
river Mississippi. 

So far the journey of the Baron La Hontan is plain enough ; but 
beyond this point it is rather apocryphal. He states that he as- 
cended the Mississippi for nine days, when he “entered the mouth 
of the Long River, which looks like a lake full of bulrushes.” 
He sailed up this river for six weeks, passing through various na- 
tions of savages, of which a most fanciful description is given. 
At length, deterred by the advance of the season, he abandoned 
the intention of reaching the heads of the river, and returned to 
Canada, having, at the termination of his voyage, first “fixed a 
great long pole, with the arms of France done upon a plate of 
lead.”” The following is his description of «the Long River.” 
«You must know that the stream of the Long River is all along 
very slack and easy, abating for about three leagues between the 
fourteenth and fifteenth village; for there, indeed, its current may 
be called rapid. ‘The channel is so straight that it scarce winds 
at all from the head to the lake. ’Tis true ’tis not very pleasant, 
for most of its banks have a dismal prospect, and the water itself 
has an ugly taste; but then its usefulness atones for such incon- 
veniences, for ’tis navigable with the greatest ease, and will bear 
barques of fifty tons, till you come to that place that is marked 
with a flower-de-luce in the map, and where I put up the post 
that my soldiers christened La Hontan’s limit.” 

_ It was at this place that the baron received his information re- 
_specting the lake of salt water. He says, «Two days after, the 
cacick” (of the Gnacsitares) «came to see me, and brought with 


— a en ee ee 


_——. 


aii, z ——— i 


BARON LA HONTAN’S LAKE OF SALT WATER. 153 


_ him four hundred of his own subjects and four Mozeemlek savages, 

whom I took for Spaniards. My mistake was occasioned by the 
great difference between these two American nations; for the 
Mozeemlek savages were clothed, they had a thick bushy beard, and 
their hair hung down under their ears; their complexion was 
swarthy, their address was civil and submissive, their mien grave, 
and their carriage engaging. Upon these considerations I could 
not imagine that they were savages, though, after all, I found my- 
self mistaken. These four slaves gave me a description of their 
country, which the Gnacsitares represented by way of a map upon 
a deer’s skin, as you see it drawn in this map. Their villages 
stand upon a river that springs out of a ridge of mountains, from 
which the Long River likewise derives its source, there being a 
great many brooks there, which, by a joint confluence, form the 
river.’ 

«The Mozeemlek nation is numerous and puissant. The four 
slaves of that country informed me that at the distance of one 
hundred and fifty leagues from the place I then was, their prin- 
cipal river empties itself into a salt lake of three hundred leagues 
in circumference, the mouth of which is two leagues broad; that 
the lower part of that river is adorned with six noble cities, sur- 
rounded with stone cemented with fat earth; that the houses of: 
these cities have no roofs, but are open above, like a platform, as 
you see them drawn in the map; that besides the above-mentioned 
cities, there are above an hundred towns, great and small, round 
that sort of sea, upon which they navigate with such boats as you 
see drawn in the map ;* that the people of that country made 
stuffs, copper axes, and several other manufactures, which the 
Outagamis and my other interpreters could not give me to under- 
stand, as being altogether unacquainted with such things; that 
their government was despotic, and lodged in the hands of one 
great head, to whom the rest paid a trembling submission ; that the 
people upon that lake are called Tahuglauk, and are as numerous 
as the leaves of trees, (such is the expression that the savages use 
for an hyperbole;) that the Mozeemlek people supply the cities 


* The boats, with a drawing, are thus described in the map :—‘‘ The vessels used 
by the Tahuglauk, in which two hundred men may row, provided they are such 
forms as y° Mozeemlek people drew me on y° bark of trees. According to my com- 
putation, such a vessel must be one hundred and thirty feet long from the prow to 
the stern.” ; 


154 BARON LA HONTAN—NATIVE TRIBES. 


or towns of the Tahuglauk with great numbers of little calves, which 
they take in the above-mentioned mountain; and that the Tahug- 
lauk make use of these calves for several ends; for they not only 
eat their flesh, but bring’em up to labour, and make clothes, boots, 
&ec. of their skins. They added, that it was their misfortune to 
be took prisoners by the Gnacsitares with war, which had lasted for 
eighteen years; but that they hoped a peace would be speedily 
concluded, upon which the prisoners would be exchanged, pursuant 
to the usual custom. I could pump nothing further out of ’em, 
with relation to the country, commerce, and customs of that remote 
nation: all they could say was that the great river of that nation 
runs along westward, and that the salt lake into which it falls is 
three hundred leagues in circumference and thirty in breadth, its 
mouth stretching a great way to the southward.”’ «I would have 
fain satisfied my curiosity, in being an eyewitness of the manners 
and customs of the Tahuglauk, but that being impracticable, I 
was forced to be instructed at secondhand by these Mozeemlek ° 
slaves; who assured me upon the faith of a savage that the Ta- 
huglauk wear their beards two fingers’ breadth long; that their 
garments reach down to their knees; that they cover their heads 
with a sharp-pointed cap; that they always wear a long stick or 
cane in their hands, which is tipped, not unlike what we use in 
Europe; that they wear a sort of boot upon their legs which reach 
up to their knee; that their women never show themselves, which 
perhaps proceeds from the same principle that prevails in Italy 
and Spain; and in fine, that this people are always at war with the 
puissant nations that are seated in the neighbourhood of the 
lake, but withal that they never disquiet the strolling nations that 
fall in their way by reason of their weakness—an admirable lesson 
for some princes in the world, who are so much intent upon the 
making use of the strongest hand. This was all I could gather 
upon that subject. My curiosity prompted me to desire a more 
particular account ;* but unluckily I wanted a good interpreter: 


* On that part of the map which is confessedly derived from Indian authority 
is the following note:—‘‘ A map drawn upon stag-skins by y® Gnacsitares, who gave 
me to know y® latitudes of all ye places marked in it, by pointing to y° respective 
places of ye heavens that one or t’other corresponded to; for by this means I 
could adjust y° latitude to half a degree or little more; having first received 
from them a computation of y® distances in fazons, each of which I compute to 
be three long French leagues.” 


+ ign Pr Faity 
ea ‘ 


on it er ane es ie 
Paw sie arraiiaoens) si yl: ee 

OY DBA! 4 Limp yo ty fnalp.oanl ree 
“8 gts ay h eg PVGTO ig na aid Coa ti 
dan Spi" tidutinoy: anh Blivun I 

“ tates Nas sored, vrei 

ae Ye aay ort Iwate #iL Rabahd eo VM i 
) GW Os rf whe) Mie ‘ banc han'e Ate & te eo eee 

. | fo ee Dic «i 

seed din AO seeks Wga's site ane 


aN bie eta Vv ity eae 
oy hiss vy eS BO we eta hy nl 4 re ne 
ae hai p gle 


Mts ¢ “tes 
nix paar , ad hy bie Leva aa 
OMA RR 


Bi azi dsl eae El De a \P Pe ey 
ae PR WEN VG) URE i] an dante 
ghd? pe agers Rly a yal. Mae ta F died ea 
‘es RAS SPE he we Re mee ah ras ae coke 


Be iat tg vd 2 ai bile 


at Wi thy | areas a 5 el it re 


eicaay WSs hiasbin te aah 
De 


BRL has ER at FS 


agkeat. ils we Lp lacks myst: eg 
ath wily pth oak (Wer gy piste nati t Sadaeg this 


sade Dip ile thai lair stub Ih; AIR AH be vil pew wfiavie ai: | 
’ ae oat arate sO 3 ancl Shaky cee Silt Hp nee ip heey: satieciaooge 
Ait SANE; ae foul: DG iN)? nel Pa aay ‘ veal, E 
i. ts one i ne aise nee em nitht es 


. 
tee . » . > 
a 
t 
Pa 
‘ 
“ 
, 
“ 
Ly vy @ ‘ 
v 
. 
“es 
as 
é - 
' 
; 
f 
; 
p> 2 ; 
hs ’ 
. A ‘ ee 
ut Pe >> 
. ' er. a 
t 
J pti P 2 
‘ 94 ai 4 * #5 
é - 
4 ‘ - 
ei ‘ 
, ae 
5 
; a a Rib me me 
; xi fo 


he te se 


ay ee 
ote 9 
, Pa caf tid Pom 
i Se ae 


ANTICIPATED ROUTE ACROSS THE CONTINENT. 155 


and having to do with several persons that did not well understand 
themselves, I could make nothing of their incoherent fustian.”’ 

A detailed map accompanies this imaginative voyage up this most 
imaginary river. Itis represented as flowing due east through 25 
degrees of longitude, numerous streams putting into it on either 
side, with mountains, islands, villages, and domains of Indian tribes, 
whose very names have at this day sunk into oblivion. The map 
was afterward published, in 1710, by John Senex, F.R.S., as a 
part of «North America, corrected from the observations communi- 
cated to the Royal Society at London and the Royal Academy at 
Paris,’ and I have annexed it as a specimen of the geographical 
knowledge of America enjoyed at that period. 

This discovery of Baron La Hontan excited, even at that early 
day, the spirit of enterprise and speculation which has proved so 
marked a feature in the national character. In a work published 
in this country in 1772, and entitled «A description of the Pro- 
vince of Carolana, by the Spaniards called Florida, and by the 
French La Louisiane, by Daniel Cox,” the then proprietary, the 
first part of the fifth chapter is devoted to a «A new and cu- 
rious discovery and relation of an easy communication between 
the river Meschacebe (Mississippi) and the South Sea which sepa- 
rates America from China, by means of several large rivers and 
lakes.” 

The author says :—«It will be of great conveniency to this coun- 
try, af ever it becomes to be settled, that there is an easy commu- 
nication therewith and the South Sea, which lies between America 
and China, and that two ways: by the north branch of the great 
Yellow River, by the natives called the river of the Massorites,”’ 
(doubtless the Missouri,) “which hath a course of five hundred 
miles, navigable to its heads or springs, and which proceeds from 
a ridge of hills somewhat north of New Mexico, passable by horse, 
foot, or wagon, in less than half a day. On the other side are 
riers which run into a great lake that empties itself by another 
great navigable river into the South Sea. The same may be said 
of the Meschaouay, up which our people have been, but not so far 
as the Baron La Hontan, who passed on it above three hundred 
miles almost due west, and declares it comes from the same ridge 
of hills above mentioned, and that divers rivers from the other side 
soon make a large river, which enters into a vast lake, on which 
‘Inhabit two or three great nations, much more populous and civil- 
ized than other Indians; and out of that lake a great river disem- 


The Long River or Dedd River was discover fately by the j {2 ES = 

Baron liahontan as for as uw markd inthe May) that whith as! Cah coors 
sor tsmore Co the ba was drain by the Savages of the Hation / " g Sin, 2 ek ts 
; of Gnaesitares ont car fkins Unlefs the Baron Lahontan has” eka z. ila Q 


Zee \ 
nG<a& °} 


The Glead of th aAMarf{A, 
MY sefsipt ye re Uh 


pvented thefe things, whith ts hard to rofolve Np being the enly 
; = By, 

Person thathas fo into these vaft C ountqes 

/ / 


Mp 


Mpzeemlec 


% Ce 
our 


Th e/N ation 


/ me to the report of the Sa 8 
ee as cet Aap Ma fifacagn-r > 5 
>= Z aa i 
| <2 Ey epi Sh) = 
a == Se. - 
3 > n re The Nation dB ») 
~m ~, Gnadfitares Stove JSNavdvesst 


RM 
= ed 
SCT OU LOUCESTS 


—=s,) 


asfar a5 this Place 
cS! Bans ‘a ee 


LPleine 


ation of Cr 
intons eee 


thenpndering 
ol 


R SPeter~7 


°F Hitler 


Efanap es 


t 
| 
| 
| 
lhe | 
| 
Mahe ( 
| 


Misc onstt OTS } 
} 


m It 
By. nm imn / 
cw Indian Villages | 
€ Op Salt tate IO. Bragucs Fac simile 
whde and 300 about according lo the of « part of a map of - the Panis 
NORTH AMERICA 
corrected feom the observations 
communicaled to the Kayal Seetely 


al London and the Royal Academy 


report of the Savages who allfo Say Lack Fanis 
that the mouth ofa “sat agreal des 
tance, from the South C ‘caft, and ts 
tarry, 2 leaques broad That Were is ae : 5 
above 100 Tons about t /Qnd that ; at Faris. ; . \ Sopmbely 
| thay Sail on it with large Boats. | y hn wee RS ATL | RR Village of | 
Q r | oTamaroa 


10.5. 


a, 

hex 

Lak Pime tout 

Y. F Louis formerly 
cald Chevecndr 


156 COMMENCEMENT OF THE SURVEY OF SALT LAKE. 


bogues into the South Sea, which is doubtless the same with that 
before mentioned, the head of the two rivers being little distant 
from each other.”’ 

In his preface, the writer indulges in the following remarks: 
«<The possibility of a communication by water (except about half 
a day’s land carriage) between the river Meschacebe and the South 
Sea, stretching from America to Japan, which is represented in 
the fifth chapter of this treatise, deserves to be well and duly con- 
sidered.” The work contains what is called «a large and accurate 
map of Carolana, and of the river Meschacebe.”’ 

On the third of April, the parties were fully organized; and 
every preparation being made, teams were despatched to the eastern 
shore of Antelope Island, with the camp and provisions intended 
for that division of our force assigned to the survey of the west 
side of the lake. It was my purpose to form here a small depot, 
accessible by water, and also to herd the mules and cattle during 
the summer. Before commencing the actual survey, I designed 
to make a thorough reconnoissance of the entire lake by water, and 
to erect stations upon the prominent points of the shores and 
islands, sufficient to enable me to cover the whole surface with a 
series of triangles, which would verify the work. 

Although it is not my purpose to give a detailed account of 
every day’s operations while engaged upon this duty, I shall make 
such extracts from the journals as will, perhaps, convey a suf- 
ficiently clear idea of this hitherto unknown sea, and of some of 
the difficulties encountered in its exploration. 

After starting the teams, we embarked on board of our new 
craft, to descend the Jordan to its mouth, where it enters the lake. 
The stream being much swollen by the melting of the snows, the 
current was extremely rapid. The dry, leafless willows on the 
banks were, in some places, black with imnumerable birds’-nests, 
the work of the preceding season, securely built in the midst of 
impenetrable thickets. The day was bright and warm, and all 
felt the exhilaration of spirits occasioned by release from a four- 
months’ confinement to weary and monotonous quarters, and were 
filled with the eager anticipation of ranging over the mysterious 
waters of this far-famed sea! The grass was becoming green, the 
waters began to flow in little rills, and the air was vocal with the 
inusic of myriads of frogs set free from their icy fetters and exult- 
ing beneath the genial rays of an April sun. Every thing betokened 
the opening of spring, although the surrounding mountains 


BEGINNING OF TROUBLES—ANTELOPE ISLAND. 157 


were yet covered with snow, many of them quite down to their 
bases. | 

April 4.—After a row of some twenty miles, we reached the 
mouth of the river, which here spread out into innumerable shallow 
channels, the intervals between them consisting of soft, shiny mud, 
nearly level with the surface of the water, and rendering it im- 
possible to find where the mouth actually was. The channels were 
only a few inches deep, although the current through them’ was 
quite rapid, owing to the great body of water coming down from 
above. The boat soon grounded, and all hands were overboard 
and commenced dragging her forward by main force. On reach- 
ing the broader expanse of water, the channels became less defined 
and more shallow as we advanced. We were therefore obliged to 
unload the boat entirely, and to pile up the baggage upon a plat- 
form constructed of oars and tent-poles, placed upon camp-stools, 
buckets and camp-kettles. After six hours of severe labour, we 
at length succeeded in reaching water that would float our little 
eraft; and, returning to the platform for the baggage, reloaded her, 
and took once more to the oars, directing our course for Antelope 
Island, to rejoin the shore party that had approached it by passing 
over a sand bar which unites it with the mainland. For several 
miles the water was not more than a foot in depth, but gradually 
increased to six and eight feet. 

After a heavy row of six hours we reached the island, and 
found the camp already pitched near the beach. Drift-wood was 
abundant along the shore, and a fire was very grateful, as we had 
been pulling in the teeth of a raw, cold, west wind, which had 
sprung up in the afternoon, and which had chilled those not warmed 
by exercise, to the very bone. A hot stew and plenty of hot 
coffee soon restored circulation and cheerfulness, and we retired 
to our blankets and buffalo-skins, spread upon the ground, well pre- 
pared by twelve hours of incessant toil for deep and welcome re- 
pose. Innumerable flocks of ducks, geese, white swans, and long- 
legged plover were seen during the day, congregated around the 
shallows at the mouth of the Jordan. 

Friday, April 5.—The water being here very indifferent, we 
moved about five miles to the northward, and encamped on a 
rocky point near some large springs of tolerably good water, 
breaking out from the hills above. It is worthy of remark that 
the quality of almost all the springs in this region is dependent, 
in a great degree, upon the season of the year. In the spring and 


158 ANTELOPE ISLAND—FORMATION OF SALT. 


early part of summer, they are supplied by the percolations of the 
melting snows from the neighbouring hills, which at this season 
are generally copious, and render the water sweet and palatable. 
As soon as their supplies from a higher level begin to fail, the soil 
through which they run, or from the depths of which they rise, 
being strongly impregnated with minerals, imparts to the water 
saline or mineral qualities, which in the autumn and winter render 
them totally unfit for use. Such we found to be the case in the 
latter part of the succeeding summer. 

The rocks among which we had now encamped consisted of 
granite, or perhaps an altered sedimentary quartz or siliceous sand- 
stone. Drift-wood is scattered along the shores at an elevation 
of four or five feet above the present level of the lake, which must 
have maintained that height for a considerable period, since in 
numerous spots along the drift line unmistakable evidences of a 
well-defined beach are still to be traced with perfect precision. 
The wood is small and generally sound, but very dry, and must, 
from its appearance, have been deposited there for many years. 
It came, doubtless, from Bear River, the Weber, and the Jordan. 

Antelope Island is the largest of the islands in the lake. It is 
about sixteen miles in length and five miles broad in its widest 
part. Like all the other islands in the lake, and, indeed, all the 
prominences observed west of the Wahsatch range and within its 
valley, it consists of a long rocky eminence, ranging from north 
to south, rising abruptly from the water, and attainmg an eleyva- 
tion of about three thousand feet above the level of the lake. A 
party was sent up the mountain to erect a triangulation station 
upon its highest peak. The officer charged with that duty de- 
scribes the view from this elevation as grand and magnificent, 
embracing the whole lake, the islands, and the encircling moun- 
tains covered with snow—a superb picture set in a framework of 
silver. . 

The southern part of the island is connected with the main 
shore by an extensive sand-flat, which, in the summer, is for the 
most part dry, but is frequently flooded to the depth of eighteen 
inches, the water of the lake being driven over it by every gale 
from the north. Upon the cessation of the wind the water recedes, 
and then the depressions of the beach are filled with pools of shal- 
low water, which, evaporating under the influence of the sun, 
leave extensive deposites of salt upon the sand. The beach is at 
all times sufficiently hard to allow the passage of wagons from the 


GNVISd SLNOWSHEdd FAO Lin Ws LSTHOS VIIN AG SSVW UNV NOLLYIS NOTLWINDNV lab 
APMP OA GR YLT Wem y a: _ ‘ 


FREMONT’S ISLAND. | 159 


main shore to the island, which is constantly resorted to on account 

of its affording on the whole of its eastern slope one of the finest 

ranges for horses and cattle to be found in the whole valley. 

Being insulated from the main shore, it affords great comparative 

security from depredations by the Indians. On account of these 

advantages, and of its being accessible by water, I directed the 
herd, which had been wintered in Tuilla Valley, to be driven to 
this island. They were placed under the charge of the herdsman 
licensed by the Mormon authorities to receive all the cattle which 
may be committed to his care, he giving bond and security for 

_ their safe return, and being held responsible for any loss that may 

occur. The herd remained here until our departure from the 

country. 

Saturday, April 6—The night was windy, and the morning 
cold and raw. Left camp in the boat for an island distant about 
ten miles to the northward, called by the Mormons, Castle Island, 
for the purpose of erecting a station upon its summit. In crossing 
from the camp to the island, the lead was kept constantly going. 
The deepest water found was twelve feet, the depth varying 
generally from three and a half to six. Doubling the northern 
cape of the island, we landed upon a narrow beach, west of a pro- 
jecting little reef consisting principally of green hornblende 
rock. Mica schist also crops out, at an angle of 70°. 

From the drift-wood on the shore, three long poles were selected 
and carried to the summit of the highest peak of the island, and 
a station built with them. This was a work of severe labour, as 
the island was at least eight or nine hundred feet high, the timber 
heavy, and the ascent, in some places, very steep, exceeding an 
angle of forty-five degrees. As we rise to the summit, argillaceous 
schists predominate, filled with cubes of bisulphuret of iron, many 
of which were found among the detritus of the decomposing rock. 
Rolled stones and pebbles of quartz and granite, imbedded in a 
sedimentary rock forming a conglomerate, were also found, with 
tufa. Upon the side of a large and singular mass of schistose rock, 
with three large holes worn entirely through it by the disintegra- 
tion of its softer particles, we found a cross cut into the stone, ap- 
parently with a chisel. From the highest table of the island 
rises an oblong rocky eminence, resembling, from some points of 
view, ruins of an ancient castle, whence it had received from the 
Mormons its name of «Castle Island.” Frémont called it « Dis- 
appointment Island.’’ I deemed it but due, however, to the first 


160 FREMONT ISLAND. 


adventurous explorer of this distant region to name it after him 
who first set foot upon its shores, and have therefore called it 
Frémont Island. While putting up the station here, search was 
made for the cover to the object end of his telescope, which he 
states he had left on the summit of the island, but it could not be 
found, having probably been buried in the detritus of the rapidly 
decomposing rock upon which it had been left. 

The island is fourteen miles in circumference, has neither tim- 
ber nor water upon it, but its sides are covered with luxuriant 
grass, and abound in prodigious quantities of the wild onion, wild 
parsnip, and sego, (Calochortus luteus.) The latter is a small 
bulbous root, about the size of a walnut, very palatable and 
nutritious, and is much used by the Indian tribes as an article of 
food. It abounds on hillsides and in stony ground in great 
quantities. Near the summit of the island, the sage (Sarcobatus 
vermicularis, nees,) grew in great profusion, and to an extraordi- 
nary size, being frequently eight feet high and six or eight inches 
in diameter. Could fresh water be obtained by boring, (and it is 
worth the experiment,) a more admirable range than this for 
sheep and goats could not be desired. Being surrounded by deep 
water, the protection from wild beasts is absolute; an object in 
this country of no small importance, where wolves abound in great 
numbers. ‘The wild parsnip is already up several inches, and its 
vivid green presents a cheerful contrast on the sunny slopes with 
the snow-clad mountains which surround us. A single ground- 
squirrel was seen; but how he got here, and where he obtained 
water to sustain life, is somewhat of a mystery. In all our sub- 
sequent examinations not the least indication of a spring was dis- 
covered. Our men picked up quite a number of the eggs of the 
blue heron, now just beginning to lay, in the tall grass along bb 
shore. 

In approaching the island from the water, it presented the ap- 
pearance of regular beaches, bounded by what seemed to have 
been well-defined and pene eey horizontal water-lines, at different 
heights above each other, as if the water had settled at intervals 
to a lower level, leaving the marks of its former elevation dis- 
tinctly traced upon the hillside. This continued nearly to the 
summit, and was most HEP aro on the north-eastern side of the 
island. 

On our return to camp, we spread our sails merrily to the breeze, 
and although our boat was heavy and by no means a clipper, yet 


EGG ISLAND—WATER-FOWL—CARRINGTON ISLAND. 161 


we moved along in all the dignity and complaisance of a first-rate, 
persuaded that no other craft of equal pretensions had ever floated 
on the bosom of these solitary waters. After no little consulta- 
tion, she was finally called «« The Salicornia,” or «Flower of Salt 
Lake,’ which euphonious appellation the men very soon dispensed 
with for the more homely but more convenient one of «The 
Sally.” A small skiff had been procured as a consort to our 
frigate, and after being fitted up and caulked, proved a very valu- 
able addition to our marine. 

Tuesday, April 9.—Morning very cool. Heavy blow all last _ 
night from the north. Senta team to the city for an additionaY 
supply of provisions and equipage. Started with the boat and 
nine men to explore and erect stations on the islands in the west- 
ern portion of the lake, taking two days’ provisions and water, and 
a blanket for each man. 

Rounding the northern point of Antelope Island, we came to a 
small rocky islet, about a mile west of it, which was destitute of 
vegetation of any kind, not even a blade of grass being found upon 
it. It was literally covered with wild waterfowl; ducks, white 
brandt, blue herons, cormorants, and innumerable flocks of gulls, 
which had congregated here to build their nests. We found great 
nuthbers of these, built of sticks and rushes, in the crevices of the 
rock, and supplied ourselves, without scruple, with as many eggs 
as we needed, principally those of the herons, it being too early 
in the season for most of the other waterfowl. 

Having erected a station on this island, we started for another, 
apparently about twenty miles to the westward.- The wind was 
fresh, and we carried away the step of one of our masts in the blow. 
The crew were all entirely unaccustomed to the water, and were 
no little alarmed at the heavy swell caused by the gale, which was 
much greater than I had anticipated, and made most of them quite 
sea-sick. We arrived at the desired point without accident, and in 
time to erect a station upon the summit of the island before dark. 
The island was between six and seven hundred feet high, and six 
and a-half miles in circumference. As we ascended the slope of 
the hill, which is much more gentle than that of any other island 
in the lake, small rolled stones, sand, and gravel are first met 
with, then slate, covering the ground in broken laminez; and the 
summit consisted of ledges of excellent roofing-slate, of which any 
quantity can be obtained. The latter was filled in places with 


cubes of bisulphuret of iron, which frequently penetrated several 
11 


162 CARRINGTON ISLAND—HAT ISLAND. 


lamins. I searched diligently, but could find no cubes free, al- 
though the rock was full of the small cavities from which they had 
either been dislodged or had decayed under the influence of the 
weather. Abundance of the slate can be procured free from this 
objection; and by trial I ascertained that a nail could be driven 
through the layers almost as easily as through a shingle. On the 
shores were large quantities of a deposite resembling hard clay, 
which had formed when soft upon the rolled stones of the beach, 
and, when hardened by the sun or other causes, had been broken 
off, retaining, like a hollow mould, the ‘shape of the stone upon 
which it had been deposited. The island is surrounded by exten- 
sive shoals. The beach is gradually making to the south, and will 
doubtless join with the wide sand-flats to the south and west before 
many years. 

At sundown we returned to the beach, where we bivouacked on 
some soft sand, partially protected from the searching wind by a 
thick growth of grease-wood, which was abundant. Our fires were 
plentifully supplied from the drift-wood piled up on the shore. 

Wednesday, April 10.—Up by sunrise. Breakfast, cold fried 
bacon, roasted heron’s eggs, and cold water. Morning cool—wind 
from east, afterward shifted to north-east and north. Started for 
a small island lying about five miles to the northward, to erect a 
station upon it. We found it be a mere islet, one hundred feet in 
height, and about a mile in ‘circumference, having a long, narrow 
sand-spit running off from it in a south-east direction for a mile 
and a-half. It is merely a pile of granitic conglomerate, with tufa 
in large masses. Grease-wood seems to be the principal growth, and 
the whole island abounds in the wild onion, now vividly green, 
filling the air with its odour. Two species of cactus were also seen. 
A cliff of slate rock occurred, preserving to a certain extent its 
laminated structure, but so burned, altered, and filled with pebbles 
as to be useless. The water, for a long distance around this islet, 
is shallow, more especially to the westward. 

Having completed the station at this point, we returned to Fré- 
mont’s Island to cover the station there with cloth, so as to render 
it visible from a distance. After a row of twelve mee we landed 
on the south-west beach at noon. The water crossed was at first 
quite shallow, but gradually deepened to eighteen, twenty-four, 
twenty-seven, and thirty-three feet, and then moderately shoaled to 
Frémont’s Island, being eighteen feet deep within a hundred yards 
of the shore. 


FREMONT ISLAND—MUD ISLAND. 1638 


The west point of the island presents a bold escarpment, one 
hundred feet in height, of talcose slate, overlaid by granite and 
gneiss, occasionally traversed by seams of white ferruginous quartz, 
and containing cubic crystals of iron pyrites. The ascent of 
the southern slope of the island in this part is much more gra- 
dual than from the western point where we first landed, and a 
beautiful beach, covered with clear, white quartz pebbles, lined the 
shore of a pretty little bay, now glistening in placid beauty under 
the rays of the setting sun. The slope on this side presents the 
same appearance of benches or lines of what must have been water 
levels or beaches, parallel and horizontal, though apparently not 
so near to each other as on the north side. Ledges of mica and 
talcose slate crop out at different heights, with a dip to W. N. W. of 
about 40°. The slate is soft, slightly unctuous: laminz regular, 
parallel, and quite thin. 

At some twenty feet above the water, I observed two protruding 
ledges, in which, lying upon the slate, (which in this case was of a 
much lighter colour than the rest,) was a dark-brown rock, much 
vitrified, tinged with iron, and burned so hard that it sounded, when 
struck, like delf-ware. It had, while in a state of fusion, flowed 
around the neighbouring rocks, forming a sort of mould or casing 
over them. These having perished by gradual disintegration, 
have left the moulds connected, but empty. In this lava, quartz, 
some white and some tinged with iron, is freely interspered; in ~ 
some of the moulds, occasionally seamed with the white quartz 
veins, was a brown, hard sandstone, which, where exposed, was 
rapidly disappearing. The vegetation on this side of the island 
was similar to that on the other: the bunch-grass was especially 
fine andabundant. After a long and fatiguing row, reached camp 
at nine o’clock at night. 

Thursday, April 11.—Morning bright and warm, with gentle 
breeze from the south. Got under way early, for the purpose of 
putting up a station on Mud Island, distant about eight miles. A 
line of soundings was run until midway, when the boat grounded 
on a shoal which extends quite to the northern extremity of 
this part of the lake. The deepest water found on the line was 
eleven and a-half feet. The skiff was sent ahead with an officer, 
but it was soon left on the flat, and the party waded through soft 
mud and water to the shore. 

After dragging the large boat half a mile, a sufficiency of water 
was found to float her, within a hundred and fifty yards of a point 


164 LARVE OF INSECTS—SINGULAR FORMATIONS. 


of rocks projecting from the mud-plain which surrounds the island. 
In wading to the shore, we struggled through a deep, soft, dark- 
coloured mass of what at first appeared to be ooze and slimy mud, 
but which, upon examination, proved to consist almost solely of 
the larve of insects lying upon the bottom, producing, when dis- 
turbed, a most offensive and nauseous odour. The mass was more 
than a foot in thickness and extended several yards from the shore. 
A belt of soft, black mud, more than knee-deep, lay between the 
water and the hard, rocky beach, and seemed to be impregnated 
with all the villanous smells which nature’s laboratory was capa- 
ble of producing. 

The point where we had effected our landing was found to be 
a protrusion of an isolated pile of metamorphic rock above the 
vast mud-plain, which latter extended to the northward and east- 
ward, without a shrub or a bush or a blade of grass to be seen 
upon its surface. This protrusion consisted of various kinds of 
rock, pushed up from beneath, with a dip to the west from nearly 
perpendicular to 45°. Slate, almost vertical, was found lying side 
by side with a dark rock filled with pebbles and stones as large as @ 
man’s head, consisting of what appeared to me to be granite altered 
and burned by intense heat. This dark rock presented some indis- 
_ tinct traces of a laminated structure, and may be slate very much 
fused. Large boulders of granite and feldspar or quartz, with 
scales of mica, lay strewn about, and I observed one with several 
well-defined cubes of iron pyrites imbedded in it. The slate 
seemed to be completely filled with pebbles and small broken frag- 
ments of granite rock, with here and there a cube of iron pyrites. 
Boulders of feldspathic rock, seamed with white quartz, and con- 
taining thin veins of jasper of a brick-red colour, are occasionally 
found in the slate. Near the western extremity of the point is a 
different kind of rock—the direction nearly perpendicular. It is 
of a more sandy structure, but is filled with the same pebbles. 
The whole has been in a state of fusion. 

The mud-flat, where above the level of the water, is thickly 
covered with round, dark-coloured, circular cakes, precisely resem- 
bling, in form, colour, and appearance, the excrement of cattle 
dried in the sun. Underneath the dry surface of these cakes is a 
soft, black, and sometimes greenish mud, which, when the cake is 
moved by the foot, and the dry covering pushed aside, emits a 
most fetid, sulphurous odour, poisoning all the surrounding air. 
The substance of which these lumps are formed appears to have 


ha 
etcicaisandods 


iam 


Spm pelea eae SS Sat saree oid 


Se le 


E 


Howe 


ered 


ee ae 


BEAR-RIVER BAY. 165 


boiled up from beneath, through numerous small orifices in the 
sand, and to have spread itself over the surface of the flat, in a 
semifluid state, to the thickness of from half an inch to three 
inches, with various diameters from three inches to a foot. The 
exposed surface has been indurated slightly by the action of the 
sun, and has formed a thin, tough, and slightly elastic covering or. 
skin, which retains the substance within in a moist state for a long 
time. By long exposure, these lumps seem to dry up entirely, 
although, upon removing them, they are found still to be supplied 
with moisture from the small orifice or tube in the centre beneath, 
which latter apparently extends to a considerable depth in the 
ground. 

Having erected the station, we returned to camp, which we did 
not reach until nearly midnight, all hands being completely worn 
out by incessant labour of nearly eighteen hours at the oars, and 
in wading through mud and water. The distance passed over 
amounted to about thirty miles, and the exposure in the water, at 
this early season, was peculiarly severe. The team from town, 
with provisions, &c., returned in the afternoon. 

Friday, April 12.—Broke up the camp on Antelope Island, 
and started for the north end of the lake, to complete that portion 
of the survey as early in the season as possible. It had already 
been foreseen that one great obstacle to the rapid prosecution of 
the work would be the want of fresh water, and means had accord- 
ingly been provided for carrying in the boats as much as was pos- 
sible. Such was the limited means of transportation, that a sup- 
ply for more than two or three days could not be carried without 
overloading the boats, already burdened with camp-equipage and 
provisions. - Being uncertain of finding any water at our next 
contemplated encampment, all the vessels were filled. 

After passing Frémont’s Island, the water of the lake continued, 
as on yesterday, very shallow, the deepest being six and a-half 
feet. We were now in the Bear-River Bay, and the shallowness 
of the water is no dotibt owing to the deposite of immense quanti- 
ties of alluvion brought down by that river at every freshet, in a 
state of suspension. 

After coasting along the eastern slope of the promontory range, 
which puts into the lake from the north, and seeking in vain for a 
point at which to land, we were at length forced to drag the boat 
to within a mile and a-half of the shore, where we left her stuck 
fast in the mud. The tents, water, and provisions were placed in 


166 ROCK-GATE CAMP. 


the skiff and dragged as far as possible, when the whole had to be 
packed upon our shoulders and carried to the shore, a distance of 
rather more than half a mile, through a black, tenacious, and fetid 
ooze, which rendered wading an excessively fatiguing task. 
The camp was pitched in a wide rocky ravine, which had cut 
entirely through the southern point of a low rocky peninsula, at 
the foot of and parallel with the main promontory, and we gave 
it the name of the «Rock Gate” camp. 

The formation here was a compact, massive, blue limestone, 
thickly and irregularly marked with close seams. Numerous 
brackish and sulphur springs percolate from beneath the foot of 
the cliffs, forming a black oozy mud, which filled the air with its 
nauseous odour. Water was found in small quantities at the foot 
of the mountains to the westward, half a mile distant; and wood 
for cooking was furnished by the wild sage which grew in scattered 
patches on the sides of the hills. 

Tuesday, April 16.—The survey of Bear-River Bay had been 
carried on by two parties; that on the eastern side being under 
the command of Lieutenant Gunnison. In the afternoon a violent 
storm came up suddenly from the westward, accompanied by thun- 
der, and a gale which instantly prostrated our little encampment. 
A copious fall of rain, mingled with hail, wetted my party to the 
skin before reaching camp. The damage was soon repaired, and 
the tents repitched amid rain, hail, and snow. 

As the storm continued with unabated violence, I began to en- 
tertain serious apprehensions for the other party, under Lieutenant 
Gunnison, who were engaged on the flats on the eastern side of the 
bay; lest, m the darkness, they should miss their way, and be 
unable to return. At dusk a large signal-fire was built on the 
hilltop, and guns were fired at intervals to attract their attention. 
But the night passed without their appearing. 

Karly the following morning they came into camp, covered from 
head to foot with salt and mud, cold, wet, hungry, and thoroughly 
exhausted. A more forlorn-looking group it has seldom been my 
lot to behold. Anticipating their arrival in some such plight, I 
had had an early breakfast prepared, with plenty of hot coffee; after 
partaking of which, they were immediately wrapped in blankets, 
and a sound sleep restored them to their accustomed strength. 
The following extract from the journal of Lieutenant Gunnison 
will give an idea of what they endured in the course of the night. 
The storm overtook them in the midst of the extensive mud-flats 


dV LIVS O°¢ oNdNVO HLINd AMOOU AUNV BONVY AMOLNONOUd NAEMLYAA AWTIVA 


AN Atmpeoig EG) Yi] UbWAsyoy 
: s i 


Py 
o 
io 
7 


A NIGHT ON THE MUD-FLATS. 167 


bordering the eastern side of the lake, without a bush or a shrub 
to shelter them from its fury. Lieutenant Gunnison says—“ The 
skiff was dragged for half a mile into three inches water, when the 
wind suddenly shifted to the north, and blew a tremendous gale. 
Our course was north, and we endeavoured to force our way with 
four men wading by the sides of the boat, but the gale was too 
powerful for them, and drove the whole company off the course, so 
that by the time we were in one foot water, we were obliged to 
stop ; the spray dashed over the boat in showers ; the rain and hail 
came down in torrents; and soon all hands were drenched to the 
skin. The mist shut down upon us, and we could only see a few 
steps around. It was nearly sunset, and rapidly growing dark. 
The men became bewildered, and despaired of reaching camp, de- 
claring that they could not survive till morning. The snow began 
to fall fast and the air to become more chilly and raw. Our 
course was now completely lost, and it was rapidly becoming too 
dark to read the compass much longer. We therefore determined 
to turn back for the mud-flat from which we had started, so that 
we might at least have solid ground upon which to pass the night. 
After wading and dragging our boat for about a mile, we came 
upon our trail of yesterday. The men becoming too stiff and 
benumbed to proceed any farther, we managed to turn the skiff up 
on its side, as a shelter from the piercing wind, and laying down 
the oars and thwarts to keep us as much as possible out of the 
mud, (which was about four inches deep,) we huddled together be- 
hind it. In a couple of hours the wind lulled, the skiff was turned 
again upon its bottom, the muddy boards arranged as a sort of 
platform, and we prepared for our night’s lodging. I placed two 
of the men edgewise on the bottom of the boat, and crawled in 
alongside of them. The two extra men, (there were five of us,) 
laid themselves down on the lower tier, taking care to break joints. 
_ We had a bit of an old sail, which, in turning up the skiff, had 
been well trampled in the mud, and was in rather a sorry condi- 
tion. This we dragged over us as a covering from the snow, which 
was falling fast. Although we were almost freezing, the heat from 
our bodies was sufficient to melt it and cause it to trickle down 
upon us, to our great discomfort. 

«¢ About eleven o’clock, as nearly as we could judge, the snow 
ceased to fall, ‘but the piercing wind howled over us till daylight. 
Nearly frozen to death, we hailed the first streaks of day, and 
jumped cheerfully into the icy mud, pushed our boat a couple of 


168 MOVING CAMP—SEARCH FOR WATER. 


miles, until the water was deep enough to float her, and in two 
- hours found ourselves once more in camp, where Captain Stansbury 
anxiously awaited us with dry clothes and a hot breakfast to re- 
fresh us after our night’s adventure.” 

Strange to say, no very serious consequences followed this night 
of severe exposure. | 

Preparations were now made for removing the camp to the south- 
ward. The “Sally,” although empty, was found to be nearly high 
and dry from the effects of last night’s norther; and it took all our 
disposable force to shove her out some half mile into water deep 
enough to float her. The baggage was then carried out to the skiff, 
which was hauled and pushed out to the larger boat, then lying 
nearly two miles off. The water was very cold, and the chilling 
wind swept down from the Wahsatch Mountains, which were in many 
places covered with snow nearly to their base. 

We coasted along the promontory, as near to the shore as the depth 
of the water would permit, which was generally within a mile or 
more, until the afternoon, when we again grounded on an extensive 
shoal, and were occupied a couple of hours in dragging the boat 
over it. A small party, in the mean time, waded to the main 
shore to search for fresh water, evidences of which had been dis- 
covered from the boats. After several ineffectual attempts to land, 
we bore away for Frémont’s Island, which we reached about nine 
o'clock, enlightened by the rays of a young moon. A large fire 
from drift-wood soon illumined the beach and rocky cliffs. Fried 
bacon, hard bread, and a single gallon of coffee constituted our 
supper, no water having been procured during the day, and our 
supply having been reduced to that quantity. The men being very 
tired with rowing and wading in the cold wind and water since 
sunrise, only a single tent was pitched, which was assigned to 
Lieutenant Gunnison, who had had quite enough of “«lying-out”’ 
the night previous, and was somewhat unwell in consequence. The 
rest of us made a spacious and airy bedchamber of’the open 
canopy of heaven. ‘The wind freshened during the night, and 
rolled in a heavy sea upon the rock-bound shore; and the roar of 
the waves, as they dashed against it, reverberated among the cliffs 
like thunder, reminding some of us of scenes far, far away, where 
more mighty billows paid their unceasing tribute to the strand. 

_ Thursday, April 18.—There not being a drop of water in camp, 
we got under way without breakfast, and made for a cove just 
east of the southern extremity of the promontory, where appear- 


Ackerman Lyin. 278. Broadway NY- 


LAKE. 


SALT 


GREAT 


PROUNOUNTORY Orr 


1 


4. NEAR 


N°? 


CAMP 


CLIFF OF ALUM SLATE. 169 


ances promised a chance of landing. The water was bold to within 
three or four boats’-length of the pebbly shore, but the men were 
obliged to get into the water and carry the baggage to land, and 
thence nearly a quarter of a mile to the foot of the mountain, be- 
fore finding ground upon which to encamp. Several of them were 
quite unwell from previous exposure, and all were jaded, stiff, and 
sore. 

The nearest water was two miles distant, and had to be carried 
in India-rubber bags and kegs upon the shoulders. No breakfast 
was to be had until near noon, and the remainder of the day was 
devoted to rest and to drying our clothing and bedding. The sun 
was bright for most of the day, but the wind was keen and cold. 

The presence or absence of the sun in this climate has a marked 
influence upon the temperature. The day may be oppressively 
warm, but as soon as the sun disappears behind the western moun- 
tains, a fire is indispensable to comfort. The morning early may 
be very cold, but soon after the appearance of the sun, cloth coats 
become uncomfortable. Should the sun become overclouded for 
any length of time, they are resumed. | | 

Segos are here very abundant, and of a large size, and are found 
in every tuft of bunch-grass. In the vicinity, a cliff was discovered 
of alum slate, nearly a mile in length, and about sixty feet in 
height above the lake. It iy traversed by several dikes of 
trap, with cubical crystals of iron pyrites, and by one dike of 
quartz rock. The slate contains numerous veins of very pure 
fibrous alum. Between this cliff and Promontory Point there 
occur strata of mica slate, fine grindstone-grit, sandstone, and 
albite. 

Friday, April 19.—Rain during the night. Morning wet, cold, 
and raw. Nevertheless, we started in the boat for a large island 
west of Antelope Island, which the officers of the party had done 
me the honour to call by my name. After erecting a station upon 
its northern point, it was my intention to proceed to Black Rock, 
a large isolated rock on the southern shore of the lake, midway 
between the two islands, to erect a station upon it: thence I de- 
signed to accompany Lieutenant Gunnison to town, and fit out a 
party, to be placed under his command, for the survey of the east- 
ern shore of the lake, as well as of that part of the valley lying 
between it and the foot of the Wahsatch range. It rained and 
snowed several times in the course of the morning, and the day 
was very cold and unpleasant. We erected the station upon a pro- 


170 . BLACK ROCK. 


minence of the northern extremity of the mountain ridge form- 
ing the island, and encamped upon the white-sand beach of a 
lovely little bay indenting its eastern shore. 

The whole of this part of the ridge consists of large masses of 
quartzose and sandstone rock, variously tinged with oxide of iron, 
and conglomerates. Heavy squall, with rain and snow, during 
the night. Having provided ourselves with only a single keg of 
water, in expectation of finding some during the day, we discovered 
before night that it was nearly exhausted; so all hands were put 
upon short allowance, and a very small cup of coffee was served 
out for supper. ‘There was no discontent, however, though all 
went to their blankets almost famished for water, and fervently 
hoping we might find a supply early in the morning. 

Saturday, April 20.—A fresh gale was blowing from the north- 
west, which continued to increase during the day. The wind was 
excessively cold, and the men were obliged to wrap themselves in 
buffalo-skins to keep warm. Setting the foresail, we ran to Black 
Rock, a distance of more than twenty miles, in a little more than 
three hours. A station was framed from timbers which had been 
previously cut in the mountains and hauled to the spot for the 
purpose; but the force of the party was not sufficient to raise it. 

Orders were sent to the herdsmen in Tuilla Valley, near by, to 
bring a team in the morning, to transport Lieutenant Gunnison’s 
party to the city. A beef was killed and divided between the 
companies. The herd was also directed to be removed from 
Tuilla Valley to Antelope Island for the season. 

The station was raised the following day, and we started for the 
city, leaving the boat’s crew encamped on the shore of the lake. 
The interval, until the 25th, was employed in fitting out the party 
destined for the survey of the eastern shore of the lake, which 
being completed, I rejoined my people at Black Rock. 

Friday, April 26.—A violent blow from the north prevented 
our moving before the afternoon; but, just before sundown, the 
wind having lulled, we loaded the boat and started for Promontory 
Point, where the camp had been left under charge of Mr. Carring- 
ton. A southerly breeze struck us about dark, and continued all 
night. The weather was clear, but extremely cold. None of my 
crew had the least knowledge of managing a boat, and I was 
therefore always obliged to take the helm myself whenever the 
sails were set. As the wind had now become fair, and there was 
no necessity for rowing, each man wrapped himself in his blanket, 


“UNW'TS 


SLNOWAHA NO 


WAWO 


LAKE-WATER A PRESERVATIVE OF MEATS. 171 


and sought repose in the bottom of the boat, while I guided our 
little craft during the night, until, at the break of day, I sank 
down, benumbed by the cold and overcome by the fatigues of the 
preceding day, followed by an incessant watch of more than twelve 
hours. We were, however, not far from our encampment, hav- 
ing run during the night nearly thirty miles. 

The party under Mr. Carrington returned to the camp in the 
afternoon from Frémont’s Island, having completed its detailed 
survey during my absence. They were almost famished for water, 
and had «laid out,” or bivouacked, for four successive nights. 
The fresh provisions and vegetables I brought with me were received 
with much satisfaction. i 

Before leaving Black Rock, I made an experiment upon the 
properties of the water of the lake for preserving meat. A large 
piece of fresh beef was suspended by a cord and immersed in the 
lake for rather more than twelve hours, when it was found to be 
tolerably well corned. After this, all the beef we wished to pre- 
serve while operating upon the lake, was packed into barrels with- 
out any salt whatever, and the vessels were then filled up with the 
lake-water. No further care or preparation was necessary; and 
the meat kept perfectly sweet, although constantly exposed to the 
sun. I have no doubt that meats put up in this water would 
remain sound and good as long as if prepared after the most ap- 
proved methods. Indeed, we were obliged to mix fresh water with 
this natural brine, to prevent our meat from becoming too salt for 
present use—a very few days’ immersion changing its character 
from corned beef to what the sailors call « salt junk.” 

Monday, April 29.—Moving camp from Promonotory Point. 
The morning was bright and clear, but the wind blew a gale from 
the west, by which the boats, being deeply laden, were so much 
retarded that with all our exertion we could not get farther than 
seven miles around the western side of the Promontory, when, to 
avoid being driven ashore by thé violence of the gale, we were 
obliged to land and encamp for the night. The baggage, as usual, 
had to be carried about half a mile to obtain ground upon which 
to pitch the tents. This was found under a high cliff of quartzcse 
rock, with masses among the detritus of nearly pure feldspar. 

In the cliff was quite a large cave, which had been used as a 
shelter by Indians—the marks of their fires being still visible 
within it. Our men took possession of it, built a large fire, and 
consoled themselves for the fatigues of the day with a dance to 


172 FLAT-ROCK POINT. 


the strains of one of the most execrable fiddles I ever listened to 
anywhere. 
The water all the way from Promontory Point is very shallow 
for a long distance from the shore, which is generally sandy, with 
pebbles; the detritus from the range being mainly quartzose pud- 
ding-stone, hornblende, quartz, and granite. | 
Tuesday, April 30.—Moved camp again to-day in search of 
the springs of water I had passed on horseback in October last. 
The day was calm and warm, and the atmosphere balmy and de- 
lightful. In the course of the morning we descried from the 
boats a patch of reeds on the shore, which our experience had 
taught us to be an indication of fresh water. We accordingly 
landed, and found some very indifferent brackish water by digging. 
Fearful lest we should obtain none better, we filled our vessels and 
embarked. Crossing a shallow reef of flat rocks, extending from 
the shore for several miles, we struck at once into deep water— 
the southern extremity of a large bay with a bold shore, encircled 
by high and picturesque mountains. The water was twenty feet 
in depth, and gradually increased to thirty feet, which continued 
to a projecting rocky point, crossed by a well-defined Indian trail. 
Near the extremity of the spur is a remarkable cliff, or projection, 
which towers above the surrounding mountains, forming a promi- 
nent and impressive feature in a landscape full of wild and peculiar 
beauty. 
- The ridge of which this formed the apex is composed of black, 
blue, and ash-coloured limestone of a very close texture, seamed in 
all directions by small veins of white carbonate of lime, producing 
avery pretty appearance. It would make beautiful mantels and 
tops for tables, could it be quarried in slabs of sufficient size. 
- Calcareous tufa is forming in large quantities near the base of the 
mountain. The hillsides contain numerous caves, some of which 
are of considerable size. Stretching off from the point to the 
‘ south-west is a ledge of flat sandstone rock, lying in from one to 
two feet water, which makes a large shoal in that direction. The 
lower portion of the point itself, extending from the base of the 
cliff, is also of this rock, lying horizontally about two feet above 
the level of the water. From the prevalence of this rock here we 
called it Plat-rock Point. Rounding the prominence, the boat 
was run into a little rocky cove, and the camp pitched in a thicket 
of grease- -wood and artemisia, just above the storm-hne of the lake— 
which is here very determinate. | 


TNTOad= Sod Ty ist AGOLIN GINO da HO aes LS aie 


KN Sem pearg (6 Wry mesrias py v 


: , 
> 
(2 
3 
3 
= 
(“a 
o> 
~ 
oOo 
z 
: = 
: 5 
cS 
4 = 
OO cts 
4 
as 
. sy 
- - 


PH _ Pre eS ee eee ee he hilar tte ee ait i ie ital 


Ackerman th319 Broadway NY. = e : ast eee su " . Were , Poa ee: ; 4 ste ie . 


VIEW OF PART(OF THE WESTERN SLOPE OF PROMONTGRY RANGE. GREAT SALT LAKE 


ROCK ABBEY—INDIAN SPRINGS. 178 


To the north, the rugged and rocky eminences gradually recede 
from the shore; to which, a few miles farther on, they again return 
in a superb semicircular sweep; forming a landscape in the highest 
degree picturesque and beautiful, to which nothing is wanting but 
trees. In the centre of the arc a fantastic mass of rock reared 
its gigantic outline against the sky, presenting from the camp a 
striking resemblance to an old ruined abbey, glowing and glisten- 
ing in the rays of the setting sun. 

Immediately behind us rose a rounded knob, some two hundred 
feet in height, composed of one solid limestone rock, in the cre- 
vices of which grew, in graceful luxuriance, the everlasting bunch- 
grass, so characteristic of this region. Upon the shore, among 
the boulders with which it was thickly strewn, I obtained some 
fine specimens of dark compact limestone, containing well-defined 
fossils of Cyathophyllum and Strepletasma. 

The shore party reached camp about dark. They reported find- 
ing abundance of the finest water at the head of the bay we had 
crossed to-day, which was easy of access. This was joyful news, 
as the water we were using was very brackish and nauseous—so 
much so as to impart a most disagreeable flavour to the strongest 
coffee. 

Wednesday, May 1.—The survey of the shore-line being in 
arrears some miles, I availed myself of the opportunity to overhaul 
the large boat, which was a miserable, lumbering affair, and make 
such alterations in her as would admit of more convenient and 
expeditious stowage of the baggage. This having been accom- 
plished, I started in her for the watering-place. After rounding 
Flat-rock Point, the water was bold and deep to within a boat’s 
length of the shore opposite the springs. I at once recognised 
the spot as being the same passed by me on horseback during my 
reconnoissance of the previous October. 

Several springs here gush out, fresh, clear, cold, and convenient 
to the shore. As I anticipated the necessity of feernn: visits to 
this spot, for a supply of water for the party when engaged on the 
western side of the lake, all hands were at once set to work to run 
out a pier of stones, alongside of which the boat could lie without 
danger. This was soon done; some of the springs were cleared out 
and deepened, and the casks filled, when we returned to camp with, 
_a fine fresh breeze. In our progress we passed innumerable flocks 
of young ducks, which upon our approach dived beneath the sur- 
face and disappeared. The gnats have at this camp been almost 


174 SHOSHONEE INDIANS—SPRINGS. 


insufferable during the day, settling on the forehead and ears, get- 
ting into the eyes and hair, producing pain and irritation by their 
bite, and rendering the manipulation of instruments extremely diff- 
cult under the tormenting infliction. Fortunately, they always 
disappear upon the going down of the sun, affording a grateful re- 
lief from their annoying attacks. 

The following day we attempted to move camp, but after getting 
under way, the wind blew so hard that we could make but little 
progress, and were at length obliged to come-to on a sandbar only 
two miles from our morning encampment, and to carry our camp 
to land upon our shoulders. Here we met three Shoshonee Indians 
on horseback, who had been watching our motions for some time. 
They belonged to a small lodge encamped among the hills a few 
miles to the northward. They seemed friendly, and expressed 
much curiosity at the size of the boat, and by signs informed me 
that they would very much like to borrow it, to pay a visit to the 
islands in the lake, some of which they gave me to understand 
they had never reached. Indeed, I doubt if they had ever placed 
a canoe upon the waters. We saw no sign of any thing of the 
kind, even upon Utah Lake, although its waters abound with the 
finest fish. A hearty supper, which they devoured as if they never 
expected to get another, sent them away delighted with their 
Visit. 

In wandering among the artemisia, to-day, I was struck at by a 
most villanous-looking adder, rather more than two feet in length, 
and was very near being bitten. It was the first snake I had seen 
in the country, and I think we only saw one other—a rattle snake 
—while we were upon the lake. 

Friday, May 3.—In about two miles we came upon a number 
of fine springs, near to which I had encamped last fall. The 
water then was brackish and bad; but now, owing to the supply 
from melting snows in the mountains, it was fresh, soft, and delight- — 
ful. The springs were beautifully clear, and situated amid a large 
growth of reeds—an invariable indication, here, of fresh water. 
The one from which we obtained our supply was about fifteen feet 
long, ten feet wide, and four deep, with overhanging banks of sod. — 
Some curious insects inhabited it, several of which were caught 
and preserved. 

The adjacent hills are composed of gray limestone, overlying 
the same kind of black marble as that seen at Flatrock Poit. 


SILK PLANT—-SPRING BAY. 175 


Both are of a very close texture, and would make excellent build- 
ing-stone. 

Around the springs was found, growing in considerable quantity, 
a plant I had not before seen—called by some of the men sik- 
plant. It somewhat resembles the sumac in appearance, and has 
at its top a cluster of long slender pods, which, when ripe, split 
open longitudinally, disclosing a number of seeds, each attached 
to a fascicle of long silky fibres, resembling very fine threads of 
spun glass. The bark is tough, strong, and very much like that 
of flax. The root and the plant, when broken, exudes a milky 
viscous substance—that from the root is intensely bitter. The 
Ottoes and Omahas make lariats of the bark, which are said to be 
stronger and better than those made of hide. It is said to grow 
abundantly near Council Bluffs, in Missouri. My Frenchman 
called it vache @ lait. The Mexican negro cook calls it capote 
des acarte. He says that the Pueblo Indians call it noche. They 
cut it down when ripe, rub it so as to separate the fibres, and make 
of it beautiful and very strong fishing-lines and fine sewing-thread. 
They also use a decoction of the root for medicinal purposes—the 
root itself is put into liquor to make bitters. 

Beyond these springs the lake makes a wide, deep bay, stretch- 
ing far to the northward, near the head of which the promontory 
range bounding it on the east seems to sink away. On the west, 
the bay is bounded by a low range of rocky hills, stretching to the 
north. Between these hills the country is much lower, and ascends 
northwardly, by a gentle rise, to a belt of low hills far off in the 
distance; over which, farther to the north-west, is seen a range 
of snow-capped mountains. 

Starting from the springs with the boats, we attempted to make 
our way to the point where the eastern range seemed to terminate ; 
but found the water so shallow that it was impracticable. After 
many fruitless attempts, the boat was brought as near the shore 
as possible, part of the baggage loaded into the skiff, and pushed 
toward the land. It was impossible, however, to bring even this 

light barque nearer than within a quarter of a mile of the beach, 
and the luggage was transported to shore on our shoulders. On 
landing, we found ourselves on the margin of an immense flat of 
sand, destitute of the least sign of vegetation, and only a few 
inches above the level of the water, which covers a large portion 
of it whenever a fresh wind prevails from the south. It stretches 
across the head of the bay from near the springs to the southern 


176 CAMP ON THE SANDFLATS. 


point of the western promontory, and doubtless at one time formed 
the northern boundary of the lake in this direction. 

At the point where we landed, it was upward of a mile in breadth. 
There being no wood to be obtained for cooking purposes nearer 
than the foot of the mountains, all hands were despatched to gather 
and pack upon their shoulders sufficient for that purpose; and the 
camp was pitched upon the naked sand. ‘The land party came in 
at sundown from the line of survey, which had been carried to the 
point of the mountain at the head of the bay. They reported the 
discovery there of some springs with a small supply of good fresh 
water and abundance of excellent grass. 

Saturday, May 4.—Morning calm, bright, and balmy. De- 
spatched most of the hands to the springs found yesterday, for 
water, which they brought on their shoulders to the boat—a dis- 
tance of over two miles. On their return, we packed the baggage 
in the skiff, and dragged it about half a mile to the boat. After 
much difficulty, we extricated ourselves from the shoals, and made 
for the south point of the range forming the western boundary 
of the bay. At this point the water is deep, quite to the shore, 
which is iron-bound by boulders of black rock from the surround- 
ing hills. A small ledge or reef of rock was seen above water 
some three hundred yards to the south. Rounding the point of 
this little promontory, we encamped on a small bluff, a short dis- 
tance from the edge of the water. 

The hillsides are covered with broken masses of a very dark- 
coloured rock, containmg much iron, and which, from its peculiar 
tint, as we approached them from the water, gave the slopes the 
appearance of being covered with a forest of cedar. ‘These rocks 
we found, upon landing, to consist of a dark compact limestone, 
stained, and in many cases apparently vitrified, on the surface by 
iron. The top of the ridge, which is three or four hundred feet 
above the water, is crowned by a stratum of light cream-coloured 
limestone, in a cliff thirty feet in height. All the varieties are 
of a close texture, and very hard, especially the white, which will, 
I think, quarry easily. If so, it will be quite valuable, as the 
shores at this point being accessible for boats, a continuous navi- 
gation is practicable to Black Rock, at the southern end of the 
lake, whence the land-carriage to the city is only twenty miles. 

Small quantities of bitumen were found on the shore, in the 
masses of a substance which lined the beach to the depth of six 
inches, and resembled in appearance the brown, dried seaweed 


LARVA OF INSECTS. 177 


of the ocean. Under the magnifying glass, these masses were 
found to consist almost entirely of the larvee or dried skins of a 
dipterous insect, adhering together. They had apparently been 
driven upon the shore at different periods; some appearing fresher 
and of a different texture from others, the insects having been of 
a larger size. The question where these larve originated presents 
a curious subject of inquiry. Nothing living has as yet been 
detected in the lake, and only a few large insects in the brackish 
springs, which do not at all resemble these remains either in shape 
or size. That they have existed in almost incredible numbers is 
evident, as the shores are lined with their skins, and the bottom, 
in many instances, for a long distance out from the shores, is covered 
with them. ‘This is especially the case in the north-east part of 
the lake, where they lie on the bottom a foot thick, mingled with 
the oozy mud, of which they form a large proportion. Yesterday 


I observed from the shore several dark patches far out in the lake, 


apparently of dull reddish water, looking very much like large 
shoals. Upon rowing over some of them, however, I found that 
this change in the colour of the water, here very shallow, was not 
occasioned by any marked difference in the depth, but from the 
bottom being covered with these larve, which the oars raised at 
every stroke, rendering the water turbid and offensive. Some of 
the deposites were very large, and must have covered many hun- 
dred acres. 

Monday, May 6.—Morning clear and warm. About a mile to 
the westward of camp the forces of the land and water party were 
unitéd, to erect a triangulation station upon a high, rocky knob, 
near the southern extremity of the range, commanding a view to 
the south and west. ‘The distance from the water was about a 
mile, and the elevation of the station about six hundred feet. 
Some long sticks were selected from a pile of drift-wood on the 
beach, and transported on the shoulders of the men, over very 
rocky, ascending ground, covered with large boulders of a close- 
grained, compact limestone. The labour was very great, and was 
much enhanced by the annoyance of swarms of gnats. They have 
become an intolerable nuisance. Our faces and hands are covered 
with scars and blotches from the incessant irritation occasioned by 
their bites. . : 

Having erected the station, we proceeded along the shore in 
search of a camping-place, which was found just south of a little 


rocky: butte, having in front of it the first well-defined reef I had 
12 


178 HORNED FROGS—SAND OF THE LAKE SHORE. 


yet seen. Leaving the baggage on the camp-ground, I followed 
up the shore in search of water, and happily, in little more than 
a mile, found an abundant supply and very good, in some small 
ponds, formed by the snow and spring rains, which are, however, 
dry in the summer. ‘This was a fortunate circumstance, as, other- 
wise, we should have been obliged to cross the lake during the night 
to the springs on the western side of the main promontory, as our 
supply was by this time reduced to half a barrel. 

Several horned frogs were found to-day running upon the shores. 
The sand which forms the beach in this vicinity is worthy of re- 
mark. Under a magnifier it is found to consist of rounded glo- 
bules, chiefly of calcareous rock, worn doubtless by attrition into 
their present form, not an angular particle being found among them. 
It is variegated by different and brilliant colours, and reminded 
me of the sand I had once seen which was brought from the Great 
Desert of Zahara. Its conformation makes it very fatiguing to 
walk in, presenting no firm resistance to the foot, but yielding at 
every step. A piece of bitumen was found buried in the sand, 
which had adhered to it when softened by the sun, and completely 
frosted it over, so that it very much resembled one of the small 
chocolate lozenges of the shops, covered with miniature sugar- 
plums. : 

Tuesday, May T.—After moving camp some few miles above, 
started for an island in the lake, apparently fifteen or twenty miles 
to the southward, to place a triangulation station upon it. The 
wind had been southerly during the night, and had raised such a 
sea that I found it impossible to force the boat through the water, 
whose ponderous waves struck upon our bows with a power that 
was irresistible. After rowing some eight miles, we gave it up 
and returned to camp. ; 

One of the party, in attempting to cut across the country to-day, 
got lost, and as he did not return by dark, signal-fires were lighted 
upon one or two of the neighbouring eminences to guide him to 
camp. He returned by bedtime, very much exhausted by his wan- 
derings, having been without food or water since sunrise. 

Wednesday, May 8.—The day being calm and the water smooth, 
renewed the attempt made yesterday to reach the island to the 
southward, taking with us blankets, provisions, and water, prepared 
to encamp upon it, if necessary. We reached the island after a 
row of four hours. The water.was bold and deep nearly the whole 


AN Aeepeodg 616 4" the 


UBLILIOyoWy 


GUNNISON’S ISLAND—-GULLS AND PELICANS. 179 


-distance—fifteen, seventeen, and twenty feet; and ten feet, within 
a hundred and fifty feet of the shore. 

_ There are two islands here, one of them quite small, and lying 
within one hundred yards to the northward of the larger one, of 
which it has.at one time formed a part. We landed at the head 
of a beautiful little sandy bay, on the eastern side, which has its 
counterpart on the western, the two being separated by a low, 
_narrow neck of land, forming a delightful little nook, and separat- 
ing the lofty pile of rock forming the northern part of the island 
from the rocky cliffs which extend to its southern extremity. 

The whole neck and the shores on both of the little bays were 
occupied by immense flocks of pelicans and gulls, disturbed now for 
the first time, probably, by the intrusion of man. They literally 
darkened the air as they rose upon the wing, and, hovering over 
our heads, caused the surrounding rocks to re-echo with their dis- 
cordant screams. The ground was thickly strewn with their nests, 
of which there must have been some thousands. Numerous young, 
unfledged pelicans, were found in the nests on the ground, and 
hundreds half-grown, huddled together in groups near the water, 
while the old ones retired to a long line of sand-beach on the 
southern side of the bay, where they stood drawn up, like Prussian 
soldiers, in ranks three or four deep, for hours together, apparently 
without motion. 

A full-grown one was surprised and captured by the men, just 
as he was rising from the ground, and hurried in triumph to the 
beach. He was very indignant at the unceremonious manner in 
which he was treated, and snapped furiously with his long bill to 
the right and left at everybody that came near him. On the top 
of his bill, about midway of its length, was a projection about an 
inch long and half an inch high, resembling the old-fashioned sight 
of a rifle: in the female this is wanting. We collected as many 
eggs as we could carry. That of the gull is of the size of a hen’s 
egg, brown and spotted; that of the pelican is white, and about as 
large as a goose egg. The white of the latter, when cooked, is 
translucent, and resembles clear blanc-mange. 

After much searching, we found among the scanty drift-wood 
along the beach, two indifferent sticks with which to build a station. 
We set them up on the highest peak of the island, at its northern 
extremity, where a nearly perpendicular cliff of dark-gray limestone 
rises from the water to the height of five hundred feet. 

It was a work of great fatigue to transport these heavy timbers 


180 NORTH END OF THE LAKE. 


over a rough, and in many places almost impassable cliff, to the top 
of this towering peak; and we were no little rejoiced when it was 
accomplished. 

Having built our station, we set out on our return to camp, with 
a fair wind, which, however, soon died away, and the tired crew 
were obliged to take to their oars. When within two or three miles 
of camp, a most furious gale of wind suddenly broke down upon us 
from the north-west, which soon raised such a sea as rendered the 
progress of our heavy boat so slow that we did not reach home 
until ten o’clock, wearied, cold, and hungry. But hot coffee soon 
restored us, and we enjoyed a deep and most welcome slumber. 

Friday, May 10.—Moved again to-day, with the intention of 
encamping at the head of this arm of the lake, but the water be- 
came so shallow that not even the skiff would float, and we had to 
resort to the usual process of transporting beds and baggage on our 
backs to the shore. Here we found that we had still half a mile 
farther to go to reach the nearest artemisia-bushes, which have now 
become our sole dependence for firewood. The bushes afforded us, 
it is true, the means of making fire for cooking, but at the same 
time they gave shelter to shoals of gnats, that drove us almost dis- 
tracted. 

We are now near the head of the northern extremity of the 
lake, and expect soon to turn our faces to the south, along its 
western shore. The character of the country has changed some- 
what: the hills in the vicinity of the lake have become lower, and 
abound less in rocks, although the projecting points still consist 
of that material. Range behind range is seen to the westward 
across the water, while to the northward the ridges seem to be at 
first low, but rising in the distance, one above the other, like the 
seats of an amphitheatre. The following day we erected a large 
Station on a hill south of the camp. 

The two last nights we have been regaled by the varied notes 
of a little brown bird, the only songster I had yet heard in the © 
valley. It sang all night; but I in vain attempted to secure it as 
a specimen. A couple of blackbirds, also, hung around the camp, 
and enlivened us with an occasional note. Abundance of a fine 
small pepper-grass grew in the sandy soil, which, in the absence 
_ of other vegetables, was very refreshing. 

Monday, May 13.—Finding it impossible for the boats to ad- 
vance any farther north, on account of the extensive flats occupy- 
Ing the whole of this end of the lake, I determined to cross to the 


MOVING CAMP—TURRET ROCK. 181 


western side, leaving the shore party to follow the old storm-line, 
or that indicating the ancient limits of the water. 

A small station was put up upon a little rocky projection just 
north of the camp, which had to be constructed of stone, filled in 
with earth, as no timber of any kind could be procured for the 
purpose. A fewslender poles of drift-wood were gathered, erect- 
ed into a small lodge on the top of the mound, and covered over 
with white cloth. The camp was broken up, and every thing being 
loaded into the boats, the customary process of wading com- 
menced. 

This mode of moving was one of the most disagreeable and 
onerous of our duties. The boat was nearly two miles from the 
camp, and the men made their way through a soft, tenacious blue 
clay, ankle deep, and exposed to the insufferable annoyance of 
myriads of gnats, which the occupation of both hands in grasping 
the burden did not allow them to repel; and although, after get- 
ting out upon the water, we left for a time our inveterate little 
enemies behind, it was with the knowledge that the same torment 
must be again encountered on our next approach to land. In cross- 
ing we were repeatedly driven to the southward by the shallow- 
ness of the water; and, upon at length reaching the shore, we 
found, to our dismay, that an extensive flat of sand still lay be- 
tween us and the line of grass and bushes where alone we could 
obtain fuel for cooking. Over this the camp was carried upon 
weary shoulders, and pitched among some artemisia, near “ Turret 
Rock,” as we named a large pinnacle toward which we had direct- 
ed our course in the morning, and which had frequently before 
attracted: our notice. 

The soil, after passing the flat, is a hard yellow clay, with some 
pepper-grass—its only production. We here found a large hole 
that had been dug by the wolves. It was full of brackish water, 
which, although unfit to drink, was sufficiently good for mixing 
bread and for cooking generally. We had learned, by dire ex- 
perience, not to despise any water, however indifferent, as we did 
not know at what period we might be reduced to its exclusive use. 

The place had long been a resort for the few Indians that occa~ 
sionally pass through this inhospitable region. Remains of old 
lodges constructed of sage-bush, beds of the same, collected under 
the cover of projecting ledges of rock and little caves in the cliff, 
together with a well-defined horse-trail, showed that this had been a 
favourite camp-ground. The trail came from the south-west, and led 


182 SAND-FLATS—NORTH END OF THE LAKE. 


to the northward, probably to the more favoured region of the Pan- 
nack and the Port Neuf. 

The shore party did not reach camp until nearly ten o’clock 
to-night, having been obliged to cross the extensive flat, some six 
or eight miles wide, after the termination of a day’s work of seven 
miles of chain-line. They describe the country over which they 
have passed to-day as a mud-flat, studded with slight elevations, 
like islands, six or eight feet high, and covered with grease-wood 
and artemisia. One single stream was passed, four feet wide and 
one in depth, with gravelly bottom, the water of which was drink- 
able. ‘The rest was one barren, dreary waste, over which the 
water of the lake had at one time flowed, and its gradual reces- 
sion. had left behind it nothing but solitude and desolation. 

Tuesday, May 14.—Engaged all this day in erecting two tri- 
angulation stations; one upon a projecting peak west of Turret 
Rock, and another upon a high rocky cliff, about a mile north of 
the camp. As no timber could be obtained within twenty miles, 
both of these stations were constructed wholly of stone, laid up in 
a conical form, upon the highest point of their respective peaks, 
and covered with white cotton. This was a work of great labour, 
as the stone was difficult to be obtained, and had to be laid with 
great care to enable the station to resist the force of the violent 
winds to which it would be exposed. I added to-day the accom- 
plishment of laying stone-wall to my numerous other avocations, 
and returned to camp thoroughly wearied, and with the skin of 
my hands nearly worn through. 

The rock upon which the higher station was erected is part of a 
ridge coming from the north, and terminating at this point in a 
bold, rugged escarpment, commanding an extensive view to the 
southward and westward. It consisted of a light-gray limestone, 
very friable, and rapidly disintegrating. The rocks in the vicinity 
are completely covered over with a casing of tufa, which has formed 
upon them, adapting itself to all their irregularities. It is very 
hard, and as the limestone underneath is constantly crumbling 
away from it, numerous caves are forming, the roof and sides of 
which consist of the tufa alone. The covering is from four to six 
inches thick. In one spot I found the rock for a small space 
thinly veneered with jade. Some broken specimens of Indian 
pottery were found at this camp, and also some pieces of obsidian. 
The latter, probably, had been procured elsewhere by the pebble 
for ene arrow-heads. . 


SURVEY OF THE WESTERN SHORE. 183 


The pelican captured upon Gunnison’s Island was consigned to 
the kettle to-day, and boiled, for the purpose of preserving his 
skeleton. We had not been able to induce him to eat or drink a 
morsel, so that, finding he would starve to death, I had, a few days 
before, set him at liberty. He followed the boats for a long dis- 
tance, but we lost sight of him, and the next morning he was found 
dead upon the shore. I had made some little progress in taming 
the fierceness of his nature. He would suffer me to assist him in ~ 
pluming his feathers; but to all others he was sullen and intract- 
able, snapping violently at every one who approached him. 

Thursday, May 16.—We are now to turn our faces to the south- 
ward, and to explore the western shore of this solitary sea. From 
the most elevated points the prospect before us was dreary and 
discouraging. ‘To the south, as far as the eye could reach, lay a 
plain of arid white sand, stretching many miles westward to a lofty 
range of mountains, which had, no doubt, ages ago, formed the 
barrier to the waters in that direction. A high and rocky hill rose 
occasionally from this level flat, like an island from the sea, while 
the scene was here and there relieved by patches of green arte- 
misia—which alone flourished in this inhospitable region. The 
uncertainty of finding water was the only question which caused 
the least apprehension in encountering this desert of sand. Labour 
and fatigue all were willing to undergo; but, without water, it was 
impossible to live. We, however, determined to venture; knowing 
that, at the worst, we could procure a sufficient supply by crossing 
to the eastern shore of the lake. 

The shore party was provided with as many canteens as were 
to be found, and we commenced to move the camp; previously 
establishing a system of signals, by which they might be enabled 
to find our rendezvous at night. The wading operation had to be 
undergone as usual, and by noon we were fairly under way. Two 
long sandspits, or bars, had to be doubled before we could hope 
to reach any point that would be near the shore party at the ter- 
mination of their day’s work. This was finally accomplished by 
-dint of dragging and wading, and we at length turned in to what, 
from the boat, we supposed to be the shore. But when within 
something less than a mile of it, the boat again grounded, and the 
usual consequences followed. When we reached the beach, before 
us lay a boundless flat of white sand, only a few inches above the 
water-level of the lake. A line of artemisia-bushes appeared to be 
about half a mile distant, and we bent our steps toward it for the 


184. WESTERN SHORE OF THE LAKE. 


purpose of encamping. But the mirage was so great that we 
found oursélves much deceived in the distance. Instead of half a 
mile, the bushes were more than two miles off; and after travel- 
ling upward of a mile, I concluded to encamp where we were, and 
to go to the fringe of green for wood enough to cook with. 
This was accordingly done, and in our search we stumbled on two 
very pretty little streams of fresh, cool water, within a half-mile 
of the camp, but which, after flowing a short distance, sank in 
the sand and disappeared. Upon the banks of one of the creeks 
was a patch of long, dry, matted grass, which had been beaten 
down by the winter snows. ‘To this I set fire, as a signal to the 
shore party of our whereabouts. A huge column of smoke imme- 
diately rose to the heavens, and completely answered the purpose. 
The party, nevertheless, did not get into camp before ten o’clock 
at, night, having been perplexed in the dark by salt creeks and 
marshes. There was but little joking or music in camp to-night, 
as the unwearied fiddle had been left in the boat, and the men were 
thoroughly tired out. 

The shore party, to-day, in running their line, crossed several 
quite large streams of good fresh water; and upon the termination 
of the day’s work, came upon one, eighty feet wide and ten feet 
deep, by measurement, and flowing with a full current. All of 
these spring-branches burst forth on the old storm-line of the lake, 
but none of them ever enter it. They doubtless take their rise in 
the mountains to the north and north-west, and percolating through 
the sands, or passing in veins underneath the surface, break forth 
suddenly into bold streams, with abundance of water, which in a 
short distance fork and spread out into shallow channels, form a 
sort of marsh, and are finally altogether absorbed by the sand, 
long before they reach the lake. In the channels, and on the 
margin of the streams, grow reeds, dagger-grass, and some. cat- 
tail flags. Numerous insects congregate in the brackish waters of 
the marshes, affording food for plover, gulls, and innumerable 
waterfowl. These streams afforded the last fresh water that we 
found on the western shore of the lake. . 

Friday, May 17.—Moved camp, taking with us in the boats the 
shore party, who wished to be landed south of the marshes and 
mud-flats they had waded through yesterday. The flat where we 
- landed was six miles wide, and covered, in many places, with salt. 
The nearest wood was at a point of bluff which bounded the plain 
to the westward. ‘Transporting fuel on men’s shoulders, this dis- 


SEARCH FOR FRESH WATER. 185 


tance, after a hard day’s work, was not to be thought of; so we 
supped on some hard bread and a small piece of dried beef, which 
fortunately remained, and laid our blankets down on the sand. 

The night was fine, notwithstanding a heavy gust in the after- 
noon from the south-west, accompanied by rain, the principal part 
of which fell in copious showers upon the mountains around us. 
The shore party did not arrive until near ten o’clock, being guided 
to our bivouac by the light of a lantern elevated upon a spade- 
handle stuck in the sand. They were not a little disappointed at 
getting no supper, as they were wet, hungry, and tired. 

Saturday, May 18.—As it was manifestly impossible for us to re- 
main here any longer without the means of cooking, I determined 
to move, although the survey was considerably in the rear. The 
eyes of my assistant, Mr. Carrington, from exposure to the intense 
glare of the sunlight from the white sands, had become much in- 
flamed, and I was fearful, at one time, that he would be unable to 
continue the performance of his duties. He was much better to- 
day, however, and returned to the line of the survey. 

The skiff was despatched along the shore to the northward to 
search for some indication of the large stream of which mention 
has been made, and which we hoped might, perhaps, approach near 
enough to the shore to be made available for the purpose of wa- 
tering. It should be borne in mind that the line of survey, as 
has been remarked, followed the storm-line of the lake, which 
(since leaving Turret Rock,) was many miles west of the lake 
shore. 

After getting the camp on board, I determined to go myself in 
the yawl, and, if water could be found, to fill up every vessel we 
could spare, as I was entirely ignorant when we could again pro- 
cure this indispensable article without crossing the lake to the In- 
dian Springs. A scaffold or platform was made of spare oars and 
tent-poles, elevated upon camp-stools, placed near the boat in a 
few inches of water, and all the baggage piled upon it, to lighten 
the boat. We started after the skiff, passed it, and advanced 
farther to the north, toward a point on the shore where the ap- 
pearance of some reeds seemed to indicate the presence of water. 

Anchoring our boat to keep her from drifting off, we waded 
some half-mile to the shore, and proceeded nearly three miles in- 
land on our bare feet, over a sandy flat, and plunged through thick, 
oozy mud nearly knee-deep until we reached the growth of reeds 
we had seen from the boat. We here found where one of the 


186 NIGHT ADVENTURE. 


streams, formerly passed on the survey, had now spread far and 
wide over the surface of the plain, making a broad marsh with 
connecting channels, furnishing plenty of water very nearly fresh. 
Insects of various kinds abounded in great numbers in the shal-’ 
lows, and had attracted vast flocks of wildfowl from the lake, 
whose waters afforded them literally nothing either to eat or to 
drink. We filled up our kegs and commenced our return, when a 
violent gust arose from the north-west. Fearing the consequences, 
we hastened our steps; but upon arriving in sight of the beach, 
we found, as we had anticipated, that the water had receded before 
the wind full half a mile, and our boat, which we left afloat, was 
now high and dry upon the sand. As she was very heavy and flat- 
bottomed, we found it beyond our strength to move her, and con- 
sequently made up our minds to spend the night where we were, 
unless the wind should again change and blow from the south- 
ward or eastward, and thus cause a reflux of the water, which would 
set us afloat once more. 

The skiff was despatched with some water to the point agreed 
upon for the night’s encampment, and directed to call at the plat- 
form for some provisions and blankets for the shore party, while 
we remained seated in our boat awaiting the movement of the 
waters. After watching some hours for a change of wind, the 
men were directed to coil themselves down in the bottom of the 
boat, and we all tried to get some sleep, which, tired and weary as 
we were, soon came to our relief. While we were thus unconscious, 
the wind veered round to the south, and we were driven by the 
rising waters farther on the shore than ever. It soon ceased, 
however, and upon rising, at daylight, we were again hard and 
fast upon the bottom. As the sun rose, a gentle breeze came up 
from the south, bringing back with it the refluxed waters, which 
gradually began to deepen around us. It will thus be seen that 
the rise and fall of the water of any particular locality is de- 
pendent in great measure upon the force and direction of the 
wind, making a difference of nearly a foot in a very short period 
of time. This of course makes a corresponding difference in 
the extent of the sand-flats, amounting, in many cases, to miles in 
width. 

The skiff was descried about sunrise coming in search of us; 
and when she joined us we succeeded, by the united force of 
both crews, in dragging the boat into deeper water, and set out on 
our return to camp, little refreshed, as may readily be believed, by 


PROGRESS OF THE WORK. 187 


a night’s repose in the bottom of the boat, without even a blanket 
to cover us, and a supper and breakfast of raw bacon. 

The shore party had shared but little better than ourselves. 
Having neglected to provide themselves with the means of making 
a fire when they left camp yesterday morning, they could cook no 
food, and were consequently restricted to the same delicate fare 
as that upon which we had so sumptuously regaled. 

Owing to the character of the ground, the past week has been 
one of unusual exposure and fatigue to both parties, while the 
progress of the survey has been by no means commensurate with 
our exertions. ‘The difficulty of approaching the shore, the im- 
mense extent of the arid sand-plains, together with the distance 
of water and fuel, and the labour of obtaining them, have made 
the duties of the boat’s crew, who have been constantly wading in 
the brine of the lake or plunging and floundering through the 
deep, soft mud of the marshes, extremely arduous and harassing. 
The task of the shore party has not been less onerous, owing to 
the great distances between the camp and their work, the glare 
from the dazzling white sand, the incessant annoyance of the 
gnats, and the want of proper food and shelter. We console our- 
selves, however, with the hope that the worst has been overcome, 
and that the causes of so much vexation and delay will continue 
to diminish as we advance to the southward. 

Our stock of flour being diminished to little more than one sack, 
the coffee to sufficient for but a few days’ consumption, and the 
fresh beef having entirely disappeared, we were reduced to fat 
salt pork and fried bread. It became necessary to renew our sup- 
ply; and preparations were made for a start early in the morning 
for Antelope Island, sixty miles distant, where the cattle and 
mules were herded, and whence a team could be despatched to the 
city for what we needed. 

Monday, May 20.—Morning warm and cloudy. The strength 
of both parties was required to drag the yawl out into water deep 
enough to float her. After much wading and hauling, this was at 
length effected, and we set sail. We continued rowing and sailing 
all the day. At sundown, the wind hauling into the north-west 
and blowing quite fresh, the crew were sent to their blankets, and 
Isat at the helm until daylight, occasionally calling the men to 
the oars when the wind failed. 

I shall never forget this night. The silence of the grave was 
around us, unrelieved by the slightest sound. Not the leaping of 


188 A NIGHT ON THE LAKE—WATER-FOWL. 


a fish nor the solitary cry of a bird was to be heard, as, in profound 
darkness, the boat moved on, plunging her bows into the black 
and sullen waters. As we passed within the shadows of the ob- 
scure and frowning mountains, the eye was strained in vain to 
catch some evidence of life. The sense of isolation from every 
thing living was painfully oppressive. ven the chirp of a cricket 
would have formed some link with the world of life; but all was 
stillness and solitary desolation. 

At daylight we were still fifteen miles from the island, and the 
boat was “hove-to”’ for breakfast, which consisted of bread and 
cold bacon, without even a drink of water. Before we passed 
around the point of Antelope Island, we stopped for a few moments 
at the little islet near it, where the number of gulls and pelicans 
was, if possible, greater than we had seen on Gunnison’s Island. 
The whole islet was covered with eggs, chiefly those of gulls, and 
with innumerable young birds, just hatched, the most of which on 
our appearance instinctively concealed themselves among the 
crevices of the rocks, while the parent birds, in countless numbers, 
anxiously hovered over us, filling the air with their discordant 
cries. Some young herons and cormorants were also found amid 
this colony of gulls—the former fierce and full of fight, the latter 
timid and alarmed, running from their nests to the water, where 
they endeavoured to conceal themselves by. persevering but abor- 
tive attempts to dive. We filled half a barrel with the eggs, but 
most of them proved to be bad. © ! 

Stopping for a short time to quench our burning thirst, at a 
spring breaking forth under a ledge of rocks at the foot of a 
picturesque little bay on the eastern side of Antelope Island, we 
reached our point of destination at five o’clock in the afternoon, 
quite worn down by the incessant toil of nearly thirty-six sleepless 
hours. 

I was much gratified to learn from the herdsman the safe ar- 
‘rival of the party I had despatched to the Uintah Mountains last 
November on a trading expedition, and that they had brought with 
them nineteen fine horses. As the party on the western shore — 
were nearly out of provisions, a couple of horses were immediately 
caught, and, accompanied by a single attendant, I started at once 
for the city, distant twenty-five miles, leaving directions that a 
team should follow early in the morning, for the purpose of haul- 
ing provisions out to the island. We reached our quarters in the 
city at two in the morning, and retired to rest, thoroughly wearied 


BIVOUAC ON DOLPHIN ISLAND. 189 


out.—The Jordan was over its banks from the melting of the 
snows in the mountains. 

Thursday, May 24.—Having laid in all the necessary supplies, 
we returned to the island to-day, killed a beef, and made all pre- 
parations for an early start next day. 

The range for cattle on this island is now very fine, and the 
herd appears in excellent condition. The drove of Indian horses, 
which had suffered much from the hard winter at Fort Bridger, 
begins to improve, and many of them’are very handsome animals. 

Monday, May 28.—Owing to head winds and heavy blows upon 
the lake, we have been constantly occupied for the last four days 
in endeavouring to reach our companions on the salt-plains of the 
western shore—a place which the men had, not inappropriately, 
distinguished by the title of «Tophet.’”’ An hour before sundown 
we descried the smoke of their camp-fires near a small island on 
the west coast, called Dolphin Island, and shaped our course ac- 
cordingly. When'it became dark, we made out the camp-fire 
itself, and hoisted a signal-lanteryn to let them know that fresh 
beef, vegetables, and water were near at hand. 

At two in the morning we landed, or rather grounded, near 
what we supposed to be the sand-flat of the main land, north of 
the island, and apparently some half-mile from the camp. The 
moon was shining bright and clear, and, anchoring the boat, lest 
she should drift off before morning, we shouldered our blankets, a 
keg or two of water, and some pieces of fresh beef, and commenced 
our tramp for the encampment. Reaching the shore, we trudged 
on with bare feet for about a mile, over sharp incrustations of salt 
and sand, when we most unexpectedly came again to water. 
Judging it to be a mere pool or washing up of the lake, we disre- 
garded so petty an impediment, and continued plodding our weary 
way through it for about a mile farther, when, finding that it began 
to grow deeper and deeper, it being then nearly up to the waist, 
we returned to the flats, and, kindling a fire of sage-brush, lay 
down upon the sand until daylight. The night was uncomfortably 
chilly, and a single blanket was but a sorry protection against the 
cold, damp sand and the searching winds from the neighbouring 
mountains. 

Tuesday, May 28.—A little after sunrise, we perceived some 
of the shore party approaching in the skiff; and we now found that, 
instead of landing last night, as we had supposed, on the main, we had 
struck a wide sand-flat. extending northward from the island, had 


190 FORMATION WEST OF THE LAKE—FIELD OF SALT. 


crossed it, and attempted also to cross the channel which separated 
it from the main shore, when we were driven back by the depth 
of the water. 

After a hearty breakfast on fresh beef-bone soup, which the 
poor fellows from the region of «Tophet’’ enjoyed exceedingly, 
they were despatched after the rest of their company, with direc- _ 
tions to join me at once, in order to survey the island upon which 
we were at present encamped. They were in rather a deplorable 
condition ; their coffee being exhausted—their flour almost gone— 
and their stock of water reduced to a single five-gallon keg. 

The rest of the day was spent in erecting a station upon the 
highest peaks of the island, and in unloading the yawl upon a plat- 
form placed upon kegs and barrels set up in the water, preparatory 
to hauling her up, she having become so leaky, from thumping on 
the rocks and being dragged over sand-bars, as to endanger the 
safety of the provisions. The gnats here were perfectly intolera- 
ble, and drove us almost mad. 

The character of the country passed over by the line of survey 
during my absence was much the same as that which had prevail- 
ed more to the northward; viz. extended plains, with grease-wood 
and sage. Water, there was none; and our sole dependence was 
now upon the supply to be furnished by the boat from the eastern 
side of the lake. 

The detritus from the mountains, whenever encountered, con- 
sisted principally of compact sandstone, abundance of calcareous 
tufa, coarse sandstone, and conglomerate.. From the report of 
Mr. Carrington, of a partial examination made by him of the range 
west of Dolphin Island, it consists of calcareous tufa and conglo- 
merate, overlying which was argillaceous limestone of various 
colours and texture, extending to the base of the short spurs of the 
ridges ; above this was found black, bluish, and gray limestone, 
very compact and fine grained, veined with calc spar; above this 
again, was an inferior or earthy limestone, overlaid by a brownish- 
gray fossiliferous limestone, capped by an argillaceous limestone, 
which extended to the summit of the range. The vegetation was 
the ever-recurring artemisia, bunch-grass, and a few scattering 
dwarf cedars not more than ten feet high. 

On the flats near the lake, and immediately west of the island, 
a large field of solid salt occurred, beautifully crystallized upon 
the sand, about half an inch thick; and the crystals, from one to 
two inches in diameter, glittered in the bright sunshine like a bed 


‘HLIMON HHL NOUS NOILVLS GNV GNVISI SNOSINNGS 


- 


AN Aempeosg e.g WT uedayy 


iS 


GUNNISON’S ISLAND—GULLS AND, PELICANS. 191 


of diamonds. The evaporation of the shoal water between the 
island and the main shore has left this beautiful deposite of salt, 
which must rapidly increase in extent as the season advances. 
Wednesday, May 29.—Before breakfast, all hands hauled up 
the yawl into a few inches of water, and turned her over to dry, 
preparatory to caulking and pitching. The survey of Dolphin 
Island was completed to-day. The summit is some seventy feet 


above the level of the water, and the island consists mainly of con- 


glomerate in horizontal strata, and varying much in the size of 
the cemented stones. 

To day I made my first essay as a caulker :—caulked, payed, 
and painted the boat; and althoughit may not have been very 


artistically done, yet it fully answered the purpose. I found that 


the brine of the lake had acted very sensibly upon the iron fasten- 
ings of the boat, corroding them much more rapidly than ordinary 
sea-water would -have done. 

This day one year ago the expedition left Fort Leavenworth. 
Of the original members of the party only four now remain: the rest 
having broken their engagements and gone to the gold-mines. 
Those that left last autumn by the Little Salt Lake route, were, 
as we heard, stripped by the Indians of all they possessed, and 
left to find their way to the land of promise as they could. 

Friday, May 30.—Morning calm and pleasant, all hands up 
early to move camp. The yawl was turned over and launched, 
and found, to our great satisfaction, to leak but little. As we were 
again short of water, I determined to run over to Gunnison’s 
Island, and make a survey of its shores, while the boat crossed to 
the eastern side of the lake to obtain a supply. We landed at two 
P. M. upon the shores of the same beautiful little bay, with its 
rocky points and white-sand beach, which had so much delighted 
us when we first visited the island. Our friends, the gulls and 
pelicans, had by no means decreased in numbers; the former fill- 
ing the air with their interminable chattering, which continued the 


whole night, and formed a most striking contrast to the chilling and 


deathlike silence of the surrounding shores. The little bay is 
covered with their forms floating lightly and gracefully upon the 
undulating waters, and unceasingly engaged in earnest conversa- 
tion; while flocks of the more dignified pelicans drew off in 
Separate groups, as if in silent contempt of their more garrulous 
neighbours. 

The island is surrounded by bold, clear, and beautifully trans- 


192 GUNNISON’S ISLAND—PERILOUS ADVENTURE. 


lucent water, and is, for its size, one of+the most pleasant upon 
which we have landed. Immediately north, and distant but a 
hundred yards, is a, small rocky islet, which has at one time formed 
a part of the main island, the bar by which they are connected 
being plainly visible beneath the water. The space between them 
forms a beautiful and romantic little bay, with deep, blue water, 
clear as crystal. The northern end of the main island is a nearly 
perpendicular cliff of black and gray limestone, between five and » 
six hundred feet in height. 

Immediately upon landing, the yawl was unloaded upon the 
beach, as I intended to despatch her for water across the lake to 
the Indian Springs—the nearest point where it could be ob- 
tained without transporting it from the interior to the boat on the 
men’s shoulders. The skiff, which had started with us from Dolphin 
Island, had not yet arrived, and I only waited her coming up to 
take from her some empty kegs, which it was desirable to fill. To- 
ward sundown, the wind began to rise from the south-west and to 
blow very fresh. It gradually hauled into the west and north- 
west, and blew most furiously, but favourably for the destined point. 
Desirous of taking advantage of the gale, I directed the boat to put 
off without waiting any longer for the skiff, and she was soon lost 
to the view in the mist and gloom to leeward. The distance to the 
springs was about twenty miles, which may give an idea of some 
of the difficulties which had to be encountered in prosecuting the 
examination of these solitary waters. 

Soon after the commencement of the gale, it began to hail vio- 
lently, accompanied by a darkness and mist which made me uneasy 
for the safety of the skiff, especially as she had been seen a short 
time previous about two miles to the westward. Having waited 
until nearly dark, I concluded that she had been blown past the 
island by the sudden fury of the gust, and must consequently be in 
great danger of being lost. We were, however, utterly powerless 
to aid her, and our only hope “for her safety was that she might 
encounter the yawl, which was seen, shortly after leaving the island, . 
to haul up suddenly to the northward. One of the lookouts re- 
ported that he thought he had seen them approaching each other 
far to leeward. A man sent to a commanding peak with a tele- 
scope was ‘almost sure’ that he had seen the two boats together 
before the thickening mist hid every thing from view. This afforded 
us some relief, although we were filled with deep anxiety for the 
fate of our companions, all of whom were young men, and inex- 


GUNNISON’S ISLAND—BLIND PELICAN. 193 


perienced in the management of boats, although perfectly compe- 
tent to guide one over the calm and tranquil waters upon which we 
had put forth in the morning. All the tent-poles and cooking 
utensils were in the missing skiff; and as the night threatened to 
be cold and stormy, a temporary shelter from the weather was 
hastily erected with spare oars and some drift-wood, which, being 
covered with a couple of tents, proved a very tolerable substitute for 
our usually more comfortable quarters. 

In aramble around the shores of the island, I came across a 
venerable-looking old pelican, very large and fat, which allowed me 
to approach him without attempting to escape. Surprised at his 
apparent tameness, we examined him more closely, and found that — 
it was owing to his being perfectly blind; for he proved to be very 
pugnacious, snapping fiercely, but vaguely, on each side, in search 
of his enemies, whom he could hear, but could not see. As he was 
totally helpless, he must have subsisted on the charity of his neigh- 
bours, and his sleek and comfortable condition showed that, like 
beggars in more civilized eommunities, he had «fared sumptuously 
every day.’’ The food of these birds consists entirely of fish, which 
they must necessarily obtain either from Bear River, from the 
Weber, the Jordan, or from the warm springs on the eastern side 
of Spring Valley, at all of which places they were observed fishing 
for food. The nearest of these points is more than thirty miles 
distant, making necessary a flight of at least sixty miles to procure 
and transport food for the sustenance of their young. Immense 
numbers of the young birds are huddled together in groups about 
the islands, under the charge of a grave-looking nurse or keeper, 
who, all the time that we were there, was relieved from guard at 
intervals, as regularly as a sentinel. The goslings are an awkward, 
ungainly mass of fat, covered with a fine and exceedingly thick 
down of a light colour. 

The island, which is an irregular ridge of rock, protruded from 
beneath, consists of the same compact limestone as that observed 
at Flat Rock. Between the water and storm line, ‘on the western 
beach, coarse and fine conglomerate or pudding-stone is found in 
broad laminz, very hard, and generally inclined to the slope of the 
beach. The layers, or slabs, are of various thickness, and would 
make excellent pavements. 

Saturday, June 1.—As the watering party was still absent, I 
took such force as remained on the island, to the summit of the high 


peak, to rebuild the station previously erected there, which, from 
13 


194 HAIL-STORM ON THE LAKE. 


a distance, I had found to be not sufficiently conspicuous. As the 
morning was exceedingly hot and sultry, and the ascent of the 
cliff difficult and toilsome, we imprudently left our upper garments 
in the camp, and continued our labour upon the station until the 
middle of the afternoon; when a squall, which had been threaten- 
ing all the day, burst upon us from the north-west, with great fury, 
accompanied with vivid lightning and heavy thunder. The sky 
was black as midnight, and the reverberations of the awful peals 
of thunder from the surrounding mountains was solemn and sub- 
lime. Before the storm came up, the boat with water had been 
descried to the northward, with both sails set, and I had watched 
her with great anxiety, as I had but little confidence in the skill 
of those who had her in charge, and was fearful the squall would 
come upon them unprepared. I bitterly reproached myself for 
allowing her to go without me; as, in case of her loss, which, at 
the time, seemed highly probable, not only would the lives of her 
crew be sacrificed, but we, who remained on this barren rock, with- 
out the means of leaving it and destitute of water, must have mise- 
rably perished. To my great relief, however, I saw, through the 
glass, the sails furled, the masts taken down, and the boat brought 
to anchor just as the gust struck her, burying every thing in mist 
and darkness. I thought, too, that I made out the skiff in tow, but 
was not certain, as the distance was too great. The storm lasted 
more than an hour—the wind blew a hurricane, and it hailed with 
great violence, covering the ground in a few minutes with hail- 
stones as large as peas. We sheltered ourselves as well as we 
could behind some rocky cliffs; but, being thinly clad, we became 
uncomfortably cold; and, as soon as the hail ceased, we hastened, 
wet to the skin, to descend to our camp, glad to escape from the 
piercing wind which swept in furious gusts over our elevated 
pinnacle. | \ 

The yawl, soon after, came into the bay, with the missing men 
on board, but without the skiff, which, though recovered, had 
broken adrift in the gale, parting an inch cable, doubled, by which 
she had been fastened. It appeared, as we had hoped, that those 
in the yawl had seen the skiff yesterday in the storm, and had suc- 
ceeded in picking her up. The men in her were without their 
coats, exposed to the peltings of the hail-storm, sea-sick, almost 
frozen, and nearly scared to death. Had they not been rescued, 
they must have suffered terribly, if indeed they had not perished, 
as they had nothing to eat; their boat at the time was half-full. 


STRONG'S KNOB—CLOTH CAP. 195 


of water, and they incapable, from sea-sickness and fright, of 
making any further exertion for their safety. Had they used ordi- 
nary diligence in the morning, they might easily have reached the 
‘island before the storm commenced, and saved themselves much 
suffering and their companions no less anxiety. The loss of the 
skiff is severely felt—the nature of the shore being such that the 
large boat, when loaded, cannot, in many places, approach within 
two or three miles of the land, and the lighter vessel is therefore 
indispensable. I was rejoiced that no lives were lost, which there 
was, at one time, serious reason to apprehend. 

We found that the hail had killed a large number of the young 
pelicans, as, upon the approach of the gust, they had been deserted 
by their parents, who had betaken themselves to the protection of 
_ the neighbouring rocks, leaving their tender offspring to «bide the 
peltings of the pitiless storm.’”’ So much for the ancient fable of 
their feeding their young from their own veins! 

Sunday, June 2.—Last night the wind was from the south- 
east, which mduced me to hope that the skiff might have been 
drifted somewhere into our neighbourhood, and a man was de- 
spatched to the top of one of the cliffs, with a glass, to look out for 
her. He soon returned, and reported that he had made out what 
he believed to be the boat, four or five miles to the southward. 
This was joyful intelligence, and we immediately started after her. 
We found her in the spot indicated, nearly full of water, but with 
all the baggage undisturbed. After picking her up, we directed 
our course still farther to the southward, toward a high peak, 
which appeared to be either an island or a peninsula. We found 
it to be the latter, connected by a broad sandbar with a broken 
rocky range, which forms the northern termination of the ridge 
bounding Spring Valley on the west, and which I had crossed in 
November last, on my reconnoissance of the western shore of the 
lake. We gave it the name of «Strong’s Knob.” 

After strolling a couple of miles on the sand-flat of the main 
shore, beneath high, broken, perpendicular cliffs of black lime- 
stone, being desirous of obtaining a view of the south-western 
portion of the lake, we ascended one of the highest peaks, from 
the summit of which rose a perpendicular oblong mass of rocks, 
which, from its singular resemblance to that article of dress, we 
called «Cloth Cap.” The day was warm, and the mountain up 
which we clambered from eight hundred to a thousand feet in 
height. With considerable labour we reached the top, and were 


196 STRONG'S KNOB—CLOTH CAP. 


richly rewarded for our toil. Beneath us, to the west, lay an ex- 
tended plain of bare, arid sand—stretching, apparently, to the 
great range of mountains forming the ancient barrier of the lake- 
waters in that direction. The water-line of the lake bounded this 
flat on the north. South-west from the water, and bounded on the 
east by the range upon which we stood, an immense flat stretched 
far to the southward, until its termination was lost in the haze 
which pervaded the atmosphere. This, doubtless, comprised within 
its dreary waste the desert over which we had passed the preced- 
ing autumn. 

To the east and north-east lay the lake in calm and placid 
beauty; while to the south stretched the broken and rugged moun- 
tain upon which we were standing, whose fantastic peaks stood out 
in bold relief against the pure and azure sky. But, beautiful and 
interesting as was this vast panorama of mountain, plain, and wa- 
ter, the view to the northward and westward was any thing but 
encouraging. In continuation of the survey, the distance from this 
point to Dolphin Island must be traversed; but how, was a ques- 
tion much more easily asked than answered. 

The cliff, or conical peak, which we climbed to-day, consists of 
black and gray limestones, of various textures, and all of it highly 
fossiliferous, its character, in this respect, becoming more marked 
as we ascended, until we reached the summit, called the « Cap,” 
which is almost entirely formed of a mass of cyathophylle, im- 
bedded in limestone, constituting a complete conglomerate of these 
fossils. About halfway up the ascent, was a dark, coarsely granular 
limestone, crystalline, and filled with minute fossils of Ortris. 

The cliffs were veined in many places with fine white, fibrous cale 
spar, half an inch thick, some of which was beautifully variegated 
and stained by iron. At the western base of the mountain, we 
found quite a large cave, the walls of which were incrusted, in 
places, with salt, half an inch thick; and cyathop hy lle projeenll 
abundantly as its sides. ) 

On our return to camp, a line of soundings was taken from the 
knob to the island: the deepest water found was ten feet. 

Tuesday, June 4.—The last two days have been occupied in 
surveying the island, in finishing the station upon its summit, and 
reconstructing that on Dolphin Island, which had been destroyed © 
by the storm of Saturday. Our stock of water being reduced to 
less than one day’s supply, I started at sundown for the eastern 
side of the lake, to renew it, determining to run all night, so as to 


NIGHT VOYAGE FOR WATER. 197 


return to-morrow, if possible. We left but five gallons in the 
camp, and took the same quantity with us for our voyage. 

The water in the lake, from Gunnison’s Island to «the springs,”’ 
is bold and deep, averaging from fifteen to twenty feet, within a 
hundred yards of the shore, and reaching in some places thirty-six 
feet. After rowing till midnight, a slight breeze sprang up, which 
enabled us to set our sails, and advance, though slowly, on our 
course. ‘The men had been much fatigued before we started, by a | 
hard day’s work in climbing the rocks and rebuilding the stations; 
so they were sent to their blankets in the bottom of the boat, 
an order which they most promptly obeyed, and were soon buried 
in profound repose. The stillness of this beautiful night, as I sat 
at the helm, guiding our little bark over the solitary waters of this 
mysterious sea, was most impressive. 


‘¢ Silence how dead! and darkness how profound! 
Nor eye nor listening ear an object finds.”’ 


The moon rose bright and clear over the rugged cliffs of the pro- 
montory, as, an hour before daybreak, we landed at our little pier 
of stones; and ere long the gray tints of dawn began to appear, 
followed by the blush of a most lovely morning. 

A fire was soon kindled, coffee-pots and camp-kettles made their 
appearance, and in a short time a smoking breakfast was spread 
upon a little patch of grass, of which all partook with a keenness 
of appetite little dreamed of by more refined but less favoured 
mortals. In less than an hour we had filled our vessels, increased 
the length and stability of our pier, washed our faces and hands— 
the first time for ten days, (as water was too precious an article to 
be wasted for any purpose other than drinking and cooking,)—and 
were on our way back to camp, where, favoured by a noble breeze 
from the south, we arrived at two o’clock. The station on the 
summit was entirely completed and covered with cloth; and the 
survey of the island being finished, every preparation was made 
for an early start for the main shore on the morrow, there to 
renew the dismal scenes of salt-plains, mud-flats, gnats, and mus- 
quitoes. 

I noticed this afternoon that the gulls’-eggs, which, when we 
arrived upon the island lay so thick upon the ground that we could 
_ scarcely avoid treading upon them, had now entirely disappeared 
from the vicinity of the camp. They had undoubtedly been re- 
moved by the gulls themselves to some safer place of deposite, but 


198 RETURN TO THE WESTERN SHORE. 


how or when the removal had been effected it was impossible for us 
to discover. 

Thursday, June 6.—We left this encampment with reluctance, 
as it was the most pleasant one we had yet made in our peregri- 


nations around the lake, and pitched our tents once more upon the 


inhospitable flats of the main western shore. As it was necessary 
to get a full view of our present position, which it was impossible 
to obtain unless from some elevation, I started on foot, in com- 
pany with Mr. Carrington, for a peak some seven miles to the 
southward, crossing a broad mud-plain, bordered on the right by 
a range of hills running off to the north-west. Upon reaching the 
eminence, it was found to be part of a ridge or rocky projection 
putting down to the border of the lake from the north-west. It 
rose abruptly from an immense flat of sand and mud, extending 
some ten miles westward to the base of another similar ridge, at 
the northern termination of which we had halted in October last, 
the day previous to crossing the field of salt and reaching Pilot 
Peak. ‘To the southward the flat continued unbroken by the least 
elevation for an apparently indefinite distance. 

The question which now presented itself was in what way 
this sterile desert was to be surveyed. Apart from the con- 
sideration of time and expense, water was only to be procured 
by crossing the lake, bringing it to the shore, and then pack- 
ing it on the backs of my crew for the chain party. This was 
obviously impossible, as they could not carry enough in that 
way to supply both the shore party and themselves while pas- 
sing to and fro over the plain. In addition to this difficulty, 
how were the provisions to be carried and cooked? These con- 
siderations induced me to hesitate in risking the lives of my 
people by attempting to penetrate this desert, where the slightest 
derangement of the measures by which they were to be supplied 
with water might prove fatal. The appearance of the plain indi- 
cated that the lake had not been over it for very many years, for 
it was thickly grown up with grease-wood; and the great proba- 
bility, if not positive certainty was, that, as the waters were 
evidently in a state of subsidence, they would never again over- 
flow it. As, therefore, my object was to survey the shore of the 
lake in its present stage, I determined to abandon, in this instance, 
the storm-line, and to run the line of survey to a point west of 
the water, as it then was, and thence to strike across the flat to 
Strong’s Knob, triangulating upon the prominent points of the 


Ackerman Lith 379Broadway. NY 


NNISON’S 


U 


G 


ele 


ae: Poa: Ne 


les 


f 


J 


INE 


( 


Ceci esos 


Se 


angie: laa? ee . 


Dc on ili aes 


WESTERN SHORE—STRONG’S KNOB STATION. 199 


different ranges, so as to obtain their general shape and distances, 


and sketching in the intervening ground. This course would 
secure all the ends of practical utility, without the hazard and 
delay to be incurred by penetrating the desert. 

The hill from which we made our reconnoissance was about three 
hundred feet high, and consisted of coarsely granular and earthy 
limestone, terminating to the northward in a perpendicular cliff of 
the same formation, in horizontal strata of only a few inches in 
thickness from top to bottom, the whole of which was in a state of 
rapid disintegration. 

Friday, June T.—As it was not expected that the line could 
reach Strong’s Knob before the following day, and there was no 
intervening point that could be reached by the larger boat, pro- 
visions ready cooked and the blankets of the shore party were 
transferred to the skiff, whose crew was directed to coast along the 
shallow water as far south as they could get, and then to land in 
the bight of the bay and await the coming up of the line. Some 


drift-wood was cut up and loaded into their boat, to enable 


them to boil coffee for supper and breakfast. The main camp was 


taken to Strong’s Knob, and pitched at the base of the lofty rocky 


peak which composes it, and which is about seven hundred feet 


high. 


Saturday, June 8.—Morning warm and sultry. A station was 


erected to-day upon the highest peak of this peninsula. A circular 
. stone enclosure was built up about five feet high, within which the 


feet of a tripod, made of drift-wood poles, were placed, after the 
area had been filled in with stones and gravel; the wall was then 
continued, and the feet of the tripod secured by being built therein. 


‘The whole was covered with cotton cloth of different colours, and 


presented an object that could be easily distinguished in clear 
weather at a distance of twenty miles. 

After completing the station, and while taking a series of angles 
upon the surrounding peaks and stations, a most furious gale, with 
low muttering thunder, came up suddenly from the south, which 


‘made it difficult to stand erect in our exposed position. With the 


gale came a mist, which shortly enveloped the lake and surround- 


ing mountains, irenderint objects a few miles distant so indistinct 


as at once to put an end to my observations; and the gale at 
length rose to such a height that the instrument hem to be removed 


to the shelter of a neighbouring cliff to save it from destruction. 
‘The skiff, with the camp-equipage of the shore party, came im 


900 LOST SURVEYING PARTY. 


about noon. The party itself had bivouacked on the sand the night 
previous, and were seen from the station, just before the storm 
came up, making good progress across the flat toward the Knob. 
As they did not make their appearance at dark, a couple of 
men were sent to the west end of the Knob to make signal-fires 
to guide them in. These fires were kept up until nearly midnight, 
and supper postponed in expectation of their arrival. The look- 
outs, on their return, reported that they had seen fires to the south- 
west, and that after waiting for the party more than an hour, had 
concluded that they must have been unable to find their way in, 
owing to the darkness. I was quite uneasy about them, as I knew 
they must be suffering for water, having with them only what they 
could carry in their canteens. There was no help for it, however, 
and, about midnight, we took our supper and retired to rest. A 
number of specimens were added to the herbarium to-day. 

Sunday, June 9.—Mr. Carrington came in with his party a 
little after sunrise. They had struck from the flats to the north 
point of the range, instead of the peninsula to the north of it, 
which, intervening between them and our camp-fires, had concealed 
us from their view. Not finding the camp, as they expected, they 
had followed along the shore (which here turns to the southward) 
for five miles in search of it, but being disappointed, had returned 
to the point which they had first reached, kindled a fire, and lain 
down on the sand for the night, without either blanket .or food. 
They did not see our signal-fires before reaching this point, as 
their faces were turned to the south, and, when they did desery 
them, were too much exhausted to come in. They had suffered 
much from want of water, but were in good spirits. 

A heavy thunder-gust came up in the afternoon, with violent 
wind from south-west, and more rain than we had seen since we 
left Salt Lake City. Just before dark, the yawl was despatched 
across the lake to the springs for water, with instructions to cover 
the station near them with white cloth, and return as speedily as 
possible. Evening dark and threatening. The gusts here are 
short, but the wind very violent, driving the sand before it with 
great force and velocity. 

Monday, June 10.—Dark and lowering early in the morning, 
with some rain, and the prospect of a wet day. But it soon cleared 
off, with a brisk cool wind from the north, which anywhere else 
would have given a clear transparent atmosphere. It seems, 
however, a striking peculiarity of this climate, that no matter 


HAZINESS OF THE ATMOSPHERE. 201 


from what quarter the wind may come, nor how great its force, 
the air continues to be hazy, so that it is impossible to see any 
distant object with distinctness. Whether it be owing to the 
elevation and consequent rapid evaporation from the lake, I 
cannot say, but it occasions great vexation and delay in the ob- 
servations upon distant objects, and renders the work lable to 
uncertainty and error. 

About two P. M. a most furious gust came up from the south-west, 
with heavy thunder, and a copious shower of rain and _hail-stones 
as large as hazel-nuts. The squall lasted about twenty minutes, 
forcing the water up over the beach, overthrowing the tents a 
second time, completely flooding us with brine, and forcing us to re- 
pitch our camp higher up the beach. In the course of an hour a 
fresh gust came up from the west, but it soon hauled into the 
north, and blew with fury all the night. The weather was quite 
cold, and the wind piercing, so that we were obliged to bank up 
the bottoms of the tents, inside and out, with large stones and 
sand, to keep them from being blown away, and to exclude the 
chilling blast, which rendered great-coats indispensable to com- 
fort. The yawl is doubtless out on the lake to-night, and the 
crew exposed to all the inclemency of the storm. 

Tuesday, June 11.—The gale still continues, the temperature 
resembles that of an October morning much more than one in 
June. Closed tents at breakfast, and the survey party running 
their line buttoned up to the throat. The morning is bright and 
clear with flying clouds. The haze, however, envelops the sides 
of the distant mountains in a mist which renders their outlines dis- 
torted and indistinct. The yawl arrived in the course of the morn- 
ing under oars, having dragged her anchor and drifted to leeward. 
The night upon the lake, in the open boat, was any thing but 
agreeable: the gale had raised a heavy sea, which frequently broke 
over the bows, drenching every one to the skin, wetting their 
blankets, and rendering sleep impossible. They had reached the 
promontory on Sunday night, but, in the darkness, had struck it 
too far north. Yesterday they clothed the station, (which is now 
quite visible,) filled up their water-vessels, and were about leaving 
the springs when the storm overtook them. They had very little 
hail, although in camp it fell very abundantly. They represent 
the lake as being much higher than when we were last there; 
owing doubtless to the increased melting of the snows in the moun- 
tains, consequent upon the advance of the season. The water, 


202 CAVE AT STRONG'S KNOB—UTAH DIGGER. 


under the influence of the northern blast, rose upon the beach 
crossed by the line a few days since so as to extend some six or 
seven miles to the south of it ; but this morning it had returned to 
its old boundaries, upon the subsidence of the gale. 

The rock composing Strong’s Knob is almost entirely block 
limestone, very hard and close-grained, veined with spar, and very 
brittle. Tufa occurs near the base of the hill in large masses, 
several feet thick, some of which, having formed around large rocks, 
upon which it had deposited itself, had been precipitated with them 
from the cliffs above. In other cases, it has formed around the 
masses after they had fallen, encasing them completely with a 
shell, frequently two feet thick, and had filled up large interstices 
between them. Frequently the rock itself has disappeared, leay- 
ing the tufa behind, somewhat like a hollow mould after the cast- 
ing has been removed. At the north-west end of the peninsula is 
an outcrop of compact sandstone and millstone grit, fifty feet 
thick, capped by black fossiliferous limestone, slightly inclined. 
All attempts at taking angles from the station to-day were render- 
ed abortive by the haze which filled the atmosphere and obscured 
and concealed every distant object. In ascending the mountain, 
quite a large cave was accidently discovered in the hillside, pene- 
trating about sixty feet, with a width of twenty-five feet and 
height of ten. It had been the resort of deer and antelope. The 
rock is black and gray limestone,:with some calcareous conglome- 
rate. 

Wednesday, June 12.—Moved camp about five miles to the 
southward. The ridge continues parallel with the shore, and de- 
scends by a gentle slope nearly to the water. The shore is rocky 
with scarcely any sand-flat. 

As we were rowing along the shore, we espied an old Indian, 
with his squaw and papoose, running down the mountain to hail 
us. We landed, to inquire of him as to the prospect for water 
ahead of us; but he could give us no information on this subject. 
He was a Utah digger, and proved to be the same old fellow who 
had come to us last autumn, in Spring Valley, and who had en- 
gaged to bring in a “give-out”’ mule which we had left behind, for 
the promised reward of a new blanket. I questioned him about 
the mule, but he only laughed and would give me no satisfaction. 
The poor donkey had doubtless furnished his lodge with meat for 
the winter. He was an old man, nearly sixty, quite naked, ex- 
cept an old breech-cloth and a tattered pair of moccasins. His wife 


/MHroadway, NY 


Arkona, bith. 


Avkernin, Lath a/) Broadway NY 


VIEW FROM STRONG S KNORR, TOOK ING | SO Ure RAT SAIC WAKE 


DIGGER SQUAW AND CHILD. 203 


was in the same condition precisely, minus the moccasins, with 
a small buckskin strap over her shoulders in the form of a loop, 
in which, with its little arms clasped around its mother’s neck, 
sat a female child, four or five years old, without any clothing 
whatever. She was a fine-looking, intelligent little thing, and as 
plump asa partridge. The mother seemed to evince much affec- 
tion for it, and was very much pleased when I threw over its 
shoulders an old piece of scarlet flannel which had been torn from 
one of the stations by the wind. I noticed, however, that after 
they left us, and she thought herself out of our sight, the cloth 
was fluttering from her own person, and the baby was as destitute 
as ever. I gave them something to eat, and, what I suspect was 
more welcome, a hearty draught of water. The poor child was 
almost famished. The old man was armed with a bow and a few 
arrows, with which he was hunting for ground-squirrels. 

June 13-14.—Moved camp yesterday and to-day, aiming to 

reach what appeared to be the southern extremity of the spur, at 
the base of which the line is being run, but were much bafiled 
by shallow water in the attempt. After advancing eight or ten 
miles, we dragged the boat as near as was possible to the edge of 
a wide sand-flat, lying between us and the ridge, made a scaffold 
in the water, and upon it deposited all the provisions, and every 
thing not needed for immediate use, so as to lighten the boat 
preparatory to a trip to Antelope Island for water and’ for another 
beef from the herd. 
_ Sending the surveying party’s camp ashore with three men to 
pitch it and to cook, I started in the yawl, intending to pass be- 
tween Stansbury’s and Carrington’s Islands, which would have 
made the course very direct. The wind was fresh from the west, 
and we bowled merrily along over the dense and briny waves, un- 
til, upon approaching the passage between the islands, we ground- 
a upon a sandbar, which seemed to stretch from one island to 
the other, forbidding, apparently, all farther progress in that di- 
rection. We then tried to pass to the north-west of Carrington’s 
island, and, for a time, with every prospect of success, when we 
suddenly found ourselves embayed in a cul de sae formed by an 
extensive sand-flat, which stretched from the island an indefinite 
distance to the westward. 

There was nothing to be done now other than to coast along the 
edge of this bar until we should be able to get around it to the 
northward. Night was approaching; the wind was ahead, and 


204 | NIGHT VOYAGE. 


rising fast; while lowering clouds spread their black and gloomy 
pall over the dark, tumultuous waters. With our heavy flat-bot- 
tomed boat, rowing against a head wind and a very considerable 
sea was hard work, especially after a day already spent in severe 
toil; but we had either to continue on, or to anchor, as there was 
no shore that we could approach in the dark, on account of the 
shallowness of the water. We accordingly followed around the 
edge of the bar, being forced thus to make a circuit of some ten 
miles, when we finally succeeded in getting to the northward of 
the shoal, and turned our faces in the proper direction. By this 
time it was ten o’clock at night, and we had been constantly en- 
gaged since daylight. The wind now blowing favourably from the 
north-west, we again set our sails, the crew was sent to rest in the 
~ bottom of the boat, and I continued at the helm during the night. 
The western and northern part of this extensive flat (for it is all 
just above the level of the water) forms, as well as I could judge 
in the darkness, a hard tufaceous reef, against which a north-west 
wind dashed the heavy water with great violence. Indeed, for a 
part of the night, I was guided in my course by the roar of the 
breakers beating against the reef, reminding me forcibly of similar 
adventures. upon the iron-bound coast of New England, or of the 
heave of the surf upon the coral-reefs of Florida. . 
Nothing occurred during the night, except grounding upon the 
tail of a sand-spit making out to the southward from a little island 
a few miles north of Carrington’s, to which the boys had given the 
name of « Hat Island.” This might easily have been avoided had 
not the night been so very dark and the lofty range of the Wah- 
satch Mountains ahead enveloped us in a mantle of such profound 
blackness that it seemed at every heave of the sea as if we were 
plunging into the very mouth of Avernus. After shoving the boat 
over the bar with handspikes, we struck immediately into deep 
water, and as I now knew every inch of the way, the people again 
retired to their blankets, being very weary. The night soon began 
to clear away and the stars to appear, their beams reflected bril- 
hantly in the dense water of the lake. Flashes of vivid lightning 
blazed up occasionally from behind the mountains, and several me- 
teors, some of great size and dazzling brilliancy, shot down the sky 
to the north-east. This was the third entire night I had thus spent 
upon the lake, sitting quietly at the helm, guiding my little bark 
qver its solitary waste. Again was I struck with the deep and pro- 


BLACK FLIES—ANTELOPE ISLAND COVE. 205 


found silence that reigned around me. The night was cold, and I 
found two great-coats exceedingly welcome. 

While passing from camp over the sand-flats, this morning, I ob- 
served a quantity of translucent, white, pink, and blood-coloured 
matter, of a gelatinous, or rather mucilaginous character, spread 
about in coagulated masses upon the sand, whither it had apparently 
been washed up from the lake by yesterday’s gale. The quantity 
was considerable, and, if the whole shore was similarly lined, must 
have been very great. An incredible number of small black flies, 
also, perfectly covered the white sand near the shore, changing its 
colour completely—a fact only revealed as the swarms rose upon 
being disturbed by our footsteps. They, too, had apparently been 
driven in by the storm; for I afterward discovered that they were 
almost as thick upon the water as upon the land, moving over its 
surface with great ease and swiftness. In the shallows left by the 
receding waters, I noticed also quite a number of ants, (the first I 
had seen,) drowned seemingly by the overflow. Both of these in- 
sects doubtless furnish food for the gulls and snipes, which are 
almost the only birds found along the shores. 

Saturday, June 15.—Daylight found the boat at the mouth of 
the passage between Frémont and Antelope Islands, and, shortly 
after, we entered the beautiful little cove on the north-east side of 
the latter, from the banks of which several springs trickle down 
from the base of a small cliff of protruding rocks. 

The scene was calm and lovely in the extreme. The rays of the 
rising sun, glancing brightly over the eastern mountains, shone 
upon the tiny ripples of the placid little bay, upon whose 
bosom a flock of snow-white gulls was calmly floating; while the 
green and gently sloping shores, covered with a luxuriant growth 
of rich and waving grass, contrasted strongly in our minds with the 
dreary and desolate waste of sand over which we had been roaming 
for the last month. Several little mocking-birds were singing gayly 
on the shore, and the shrill, cheerful whistle of the curlew resound- 
ed along the beach. Four graceful antelopes were quietly grazing 
on the grassy slope, while the cry of the wild duck, and the 
trumpet-note of the sandhill crane were heard in the distance. The 
whole formed a picture which, in this desolate region, was as wel- 
come as it was rare. 

I found, this morning, that my conjecture respecting the food of 
the gulls had been correct. Across the little bay ran a broad 
streak of froth or foam, formed by the meeting of counter currents, 


906 SURVEY OF EASTERN SHORE OF THE LAKE. 


and driven in by the wind. Passing through it, I found it filled 
with the small black flies, such numbers of which I had noticed 
yesterday. In the midst of these were flocks of gulls, floating upon 
the water and industriously engaged in picking them up, precisely 
as a chicken would pick up grains of corn, and with the same ra- 
pidity of motion. 

We landed at our first camp-ground near the box-elder tree, 
about two hours after sunrise, making twenty-four continuous hours 
that I had sat at the helm, without a moment’s respite. A mes- 
senger was despatched to the person in charge of the cattle, with 
directions to drive up the herd, out of which one was selected and 
killed. The rest of the crew were engaged in filling the water-ves- 
sels from an excellent spring near the shore. Here I enjoyed the 
exquisite luxury of washing my face and hands, for the first time in 
more than two weeks—water being too scarce an article in the camp 
to allow that privilege to any other person than the cook. 

About eleven o’clock I had the pleasure of meeting my friend 
and efficient assistant, Lieutenant Gunnison, whom I found busily 
engaged in pushing forward the field-work of his portion of the 
survey, with his accustomed industry and energy. He had com- 
pleted the survey of the eastern side of the lake, and was at the 
time engaged in that of Antelope Island. He brought news of the 
arrival of the first mail this year, and a large packet of letters from 
home. After concerting measures for meeting on the western shore 
of the lake, where our surveys were to join, he returned to his camp. 
Every preparation being made for an early start in the morning, I 
retired to read my letters and to refresh my weary spirits by repose. 

Sunday, June 16.—As the party on the flats was nearly out of 
water when we left them, and the weather was so uncertain as to 
render the time occupied in our return to them equally so, I de- 
termined to leave to-day. 

- Rounding the north point of Antelope Island, we called at the 
little islet to which we had given the name of Egg Island, to look 
after our old friends, the gulls and pelicans. The former had 
hatched out their eggs, and the island was full of little, half-fledged 
younglings, who fled at our approach, and hid themselves under the 
first stone they could find. We caught several of them, and amused 
ourselves by putting them into the water, when they immediately 
followed the instinct of their natures, and paddled away with their 
little black feet most assiduously. One poor fellow, about four 
inches long, driven by the extremity of his fear, took to the water 


WATER-FOWL ON EGG ISLAND. 207 


of his own accord, when he was swept out by the current to the 
distance of two or three hundred yards, and seemed quite be- 
wildered by the novelty of his situation. As soon as he was dis- 
covered by the old birds, who hovered over our heads by thousands, 
watching our proceedings with great anxiety and noise, one—the 
parent, we judged, by its greater solicitude—lighted down by his 
side, and was soon joined by half a dozen others, who began guiding 
the little navigator to the shore, flying a little way before him, and 
again alighting, the mother swimming beside him, and evidently 
encouraging him in this his first adventure upon the water. The 
little fellow seemed perfectly to understand what was meant, and, 
when we sailed away, was advancing rapidly under the convoy of 
his friends, and was within a few yards of the shore, which he 
doubtless reached in safety. 

The young herons had grown, since our last visit, to nearly their 
full size, although they were not sufficiently feathered to fly. They, 
too, fled as fast as they could, and “cached”’ themselves in the re- 
cesses of the rocks. When closely pursued, however, they would 
turn and fight most fiercely—striking furiously with their long 
sharp bills as well as with their claws—screaming all the while with 
a shrill, discordant, and angry note. Those that were too small to 
leave the nest were equally pugnacious—standing on the defensive, 
with a watchful and determined eye, which evinced any thing but 
a disposition to succumb, if attacked. A large number of young 
cormorants (Phalacrocoraz) were also seen, who exhibited the same 
combative spirit when hard pressed; but the greater portion of 
them ran from the nest to the water, where they gave instant evi- 
dence of the peculiar instinct belonging to the species, by desperate 
attempts to dive, and thus conceal themselves beneath the water. 
This they were unable to do, owing, I suppose, partly to the great 
density of the water, and partly to their want of strength. The 
stench was very offensive, from the quantity of fish brought by the 
parent birds for the support of their very numerous progeny. 

We reached camp about ten o’clock at night, after dragging over 
the shoals and wading about a mile through soft mud, half-leg 
deep, and filled with little, sharp rocks, which cut our feet until they 
bled. A rain-storm came on just felune our arrival, with violent 
wind, which lasted all the night. 

Monday, June 1T.—As it was manifestly impossible to prosecute 
the survey any farther south by means of the boats, it was deter- 
mined to complete the examination of the islands, and then to finish 


208 CARRINGTON ISLAND. 


the remainder of this part of the work from the eastward, employ 
ing mules, if possible, to furnish the party making it with provisions 
and water. 

Both parties, therefore, proceeded to Carrington’s Island, . 
which we reached late in the afternoon, effecting a landing on its 
southern shore. It rained heavily and was very cold, with a gale 
from the northward; and we landed wet and almost frozen, haying 
spent one of the most disagreeable days we had yet endured upon 
the survey. Continually baffled by shoals, which could not be seen 
until the boat grounded upon them, the whole day had been con- 
sumed in making a distance which, under ordinary circumstances, 
might have been accomplished in a few hours. 

The two following days were occupied in the survey of ‘ki 
island, and of a small one about five miles to the northward of it. 
The water between them is quite shoal, the deepest being only six 
and a-half feet. 

The station previously erected upon the summit of Carrington’s 
Island had been torn down, doubtless by some wandering Indians, 
as we saw the remains of their fires in the immediate vicinity. 
They were probably attracted by the cloth with which it was 
covered, and must have reached it by wading and swimming to the 
island from the mainland. 

The slate found when we first landed upon this island abounds 
also in various localities. Quartzose rock, generally with a dip of 
5° to the south-east, was observed in large boulders on the southern 
slope, veined with thick seams of white quartz. Limestone was 
also found on the south-west portion of the island, near the base of 
the hill. On the north-east point was an outcrop of quartzose rock 
plentifully seamed with white and ferruginous quartz. Striated 
talcose slate, very much contorted, occurred in the centre of the 
island, and, to the west, gray granite, with quartzose conglomerate. 
The island is about eight miles in circumference, exclusive of the 
flats, which stretch out from it to the southward and westward, and 
which are more extensive than the island itself, being terminated 
on the west by the rocky reef passed on Friday night. 

It abounds in the sego, (Calochortus luteus,) which is beginning 
to seed, and, with its beautiful white, lily-like flowers, whitens 
and enlivens the gentle slopes of the island. A large number of 
other plants was also collected here, among which Cleome lutea, Si- 
dalcianeo mexicana, Malvastrum coccineum, Stephanomeria mimor, 


HGS tsa GNVISt SAUNESNVLS WOU ONV ISI SNOLONIEUYo 
XK N Aempeosg ¢/.¢ yi] UeULLayOYy 


STANSBURY’S ISLAND. 209 


a new species of Malacothiz, and Grayia spinosa were the most 
prominent. 

Thursday, June 20.—Moved camp to the north point of Stans- 
bury’s Island, and commenced the survey of it, which occupied us 
until the 26th. This is the second island in point of size in the 
lake, being twelve miles long and twenty-seven in circumference. 
Like Antelope Island, it is a high rocky ridge, rising abruptly 
from the plane of the lake, and reaches, in its greatest elevation, 
the height of nearly three thousand feet. It is, at this time, in 
fact, a peninsula, the space between it and the mainland, which 
formerly was covered by the water, being now occupied by a broad, 
level plain of sand, thickly overgrown in places by artemisia. The 
scenery, especially on the eastern side, is in many places wild, 
rugged, and grand. Peak towers above peak, and cliff beyond 
cliff, in lofty magnificence, while, crowning the summit, the 
‘‘dome’’ frowns in gloomy solitude upon the varied scene of bright 
waters, scattered verdure, and boundless plains of arid desolation 
below. ‘The eastern shore, in many parts, affords springs of excel- 
lent water, and the numerous tracks of wolves, deer, and antelope, 
added to the frequent remains of Indian fires, indicate that these 
spots have long been the favourite haunts of both man and beast. 
In the vicinity of these springs, the grasses are rich and abundant, 
and the range for cattle the best I have seen in the country. 
Both this and Antelope Island have been reserved by the saga- 
cious Mormon authorities for grazing purposes. 

In skirting the shores, several plants were collected for preser- 
vation; among which were the Comandra umbellata, a new genus 
of Hlymus, Stipa juncea, and the Elymus striatus. Various seeds 
were also gathered. 

The western shore of the island is, at this season, so far as we 
could discover, entirely destitute of water, although, while the 
snows on the summit are melting, some must doubtless reach it. 
The party, while engaged here, had to be supplied from the oppo- 
site side of the island. 

According to previous concert with Lieutenant Gunnison, a 
beacon-fire was lighted on a commanding eminence on the evening 
of the 22d, which was immediately responded to by a similar 
signal from Black Rock. I crossed over to his camp on the fol- 
lowing day, to borrow some provisions and arrange for the further 
prosecution of the survey. It was agreed that while I was coin- 


pleting the survey of this island, the line on the flats, which we 
14 


210 STANSBURY S ISLAND—THE DOME. 


were obliged to abandon for a time from want of water, should be 
resumed and completed by Lieutenant Gunnison, his provisions 
and water being transported upon mules. 

On our return from his camp, we had not proceeded more than 
a mile, before a sudden and violent gust, accompanied with light- 
ning and thunder, drove us to the shore for safety, our little skiff 
being too small and fragile to withstand the fury of the waves. I 
have had frequent occasion to remark in what a very short time a 
gust, sweeping down from the mountains, will lash the heavy 
water of the lake into waves of very considerable magnitude. The 
seas are short, and the commotion as suddenly subsides with its 
exciting cause. The rise and fall of the waters on the shores is 
thus occasioned, and frequently precedes the current of air by 
which it is produced. 

Wednesday, June 26.—The survey of the island shore being 
completed, the whole party ascended to the highest peak to erect 
a triangulation station upon it. The day was intensely hot, and 
the sun beat down upon the eastern slope, as we climbed up its 
rugged sides, with a force that was very oppressive. Every man 
was packed, like a mule, with tools and provisions; but principally 
with water, in India-rubber bags and canteens. 

After a severe climb of some three hours, through rich bunch- 
grass near the base, artemisia and grease-wood higher up, and, 
still higher, over rocky projections covered with stunted cedar, we 
at length reached the summit of the ««dome.”’ From this point, 
the highest within the circuit of the lake, we had expected to 
enjoy a noble view of both it and the surrounding islands and moun- 
tains; but, unfortunately, the atmosphere was filled with so thick 
a haze that our hopes were wholly disappointed. In our ascent, 
quite a variety of plants were collected and carefully preserved. 
Among these, several have been ascertained by Professor Torrey, 
to whom the whole collection has been submitted for examination, 
to be new species; among others, a Heuchera, Peretyle, Cowania, 
and Chenactis. 

After resting under the shadow of some wide-spreading cedar- 
trees, (the first shade we had enjoyed for months,) the summit 
of the peak was cleared, and a circular wall built, five feet high, of 
stone, upon the top of which was erected a triangulation station of 
wood, covered with cloth. An attempt was made to take some ob- 
servations here with the theodolite, but the atmosphere was so 
filled with vapour that they were not at all satisfactory. 


ASN Aempeorg 5 ae UU [ae wis 2a y 


Sigate 
ek 


cant Lil 


ee 


s 


Tet 


ee 


23 


WA Ages 


© 


AN Aemproag gle uy weUaYY 


V2 Ta 
' woo * 
ay ie 


% 


COMPLETION OF THE SURVEY—TRIANGULATION. P11 


The island where we descended is a ridge composed of a stratum 
of white siliceous sandstone, two hundred feet thick, underlying 
one of black and gray limestone, which latter constituted its sum- 
mit, and was filled with fossils of Cystiphylinm, Syringipora, Favo- 


sites, Fenestella, Streptilasma, and erinoidal joints, ima in crystal- 


line limestone, and cyathophylle. In our return from the top of 
the mountain, we followed the narrow, dry bed of a ravine or 
cajion that had been formed by the rush of a torrent of melting 


snows in the spring of the year. It was amazing to see what huge. 


masses of rock had been moved by this agency—many of several 
tons weight having been carried far into the plain below. As we 
descended, the gorge, which had at first been almost shut up be- 
tween perpendicular cliffs of white sandstone, opened out into a 
superb, wide, and gently sloping valley, sheltered on each side by 
beetling cliffs to the very water’s edge, effectually protected from 


all winds, except on the east, and covered with a most luxuriant. 


growth of rich and nutritious bunch-grass. 
Near the shore of the lake, abundant springs of pure, soft water 
gush forth, amply sufficient for the consumption of all the stock 


the valley could supply with food. As a range for cattle, it was 


all that could be desired; and is superior to either Tuilla Valley 


or Antelope Island, on account of the complete protection it affords 


from the storms of winter, here both long and severe. 

To-day has been one of severe suffering, from the insufficient 
supply of water we were able to carry with us, every drop of which 
was consumed long before we commenced our descent from the 
mountain. 

Thursday, June 27.—The survey of the lake was finished yes- 
terday, having occupied the incessant labour of nearly three 
months. Nothing now remained but to complete the observations 
upon the different triangulation stations that had been erected in 
the course of the examination. The camp was broken up, and we 
returned to our depdt on Antelope Island; discharged such of the 
hands as were no longer required, procured a fresh supply of pro- 
visions from the city, and, on Wednesday, July 3d, started on our 
tour around the lake, leaving three men upon the island, to take 
charge of the herd and to prepare the pack-saddles required for 
our return to the States. 

The triangulation of this part of the survey occupied us until 
the 16th, during which time the lake was again traversed in every 
direction, and observations were taken from the various stations 


Ae 


ee a ee a 


ESS 
ae ee OM eee te a 


ea 


912 SALT WATER BATHING—ANALYSIS OF THE LAKE WATER. 


that had been erected upon prominent headlands on the shore and 
on the summits of the several islands. 

While engaged upon this duty, we frequently enjoyed the luxury 
of bathing in the water of the lake. No one, without witnessing — 
it, can form any idea of the buoyant properties of this singular 
water. A man may float, stretched at full length, upon his back, 
having his head and neck, both his legs to the knee, and both 
arms to the elbow, entirely out of water. If a sitting position be 
assumed, with the arms extended to preserve the equilibrium, the 
shoulders will remain above the surface. The water is nevertheless 
extremely difficult to swim in, on account of the constant tendency 
of the lower extremities to rise above it. The brine, too, is so 
strong, that the least particle of it getting into the eyes produces 
the most acute pain; and if accidentally swallowed, rapid stran- 
gulation must ensue. I doubt whether the most expert swimmer 
could long preserve himself from drowning, if exposed to the 
action of a rough sea. 

Upon one occasion a man of our party fell overboard, and, al- - 
though a good swimmer, the sudden immersion caused him to take 
in some mouthfuls of water before rising to the surface. The effect 
was a most violent paroxysm of strangling and vomiting, and the 
man was unfit for duty for a day or two afterward. He would in- 
evitably have been drowned had he not received immediate assist- 
ance. After bathing, it is necessary to wash the skin with fresh 
water, to prevent the deposite of salt arising from evaporation of 
the brine. Yet a bath in this water is delightfully refreshing and 
invigorating. 

The analysis of this water by Dr. Gale has shown that it con- 
tains rather more than 20 per cent. of pure chloride of sodium, 
and not more than 2 per cent. of other salts, forming «one of the 
purest and most concentrated brines known in the world.” Its 
specific gravity was 1.17, but this will slightly vary with the sea- , 
sons, being doubtless affected by the immense floods of fresh water 
which come rushing down into it from the mountains, in the spring, 
caused by the melting of the snows in the gorges. 

Thursday, July 16.—To-day we took a final leave of this sin- 
gular lake. The difficulty of finding water fit for the ordinary 
purposes of life—the necessity of transporting, by means totally 
inadequate, every pound of provisions and every drop of water 
needed for the daily consumption of a large party of men—the un- 
avoidable distance of our depot, and the barren, savage inhospi- 


EASTERN SHORE OF THE LAKE. 913 


tality of the region we were obliged to traverse, have made this 
survey one of unusually arduous and protracted toil. But the 
salubrity of the climate is such that, notwithstanding our constant 
exposure to the vicissitudes of the elements, a large portion of the 
time without the protection of tents, not a man was seriously un- 
well, and most of the party were in the uninterrupted enjoyment 
of robust health. 

The survey of the eastern side of the lake had, in the mean time, 

been completed by the party under Lieutenant Gunnison. The 
following extract from his report to me, will sufficiently ae the 
character of this portion of the valley :— 
«To recapitulate and give the result of the field-notes. Two 
lines have been located, the shore of the lake and base of the 
hills, in order to give the flat occupied by the farmers. These 
lines are determined by the three-point problem, as numerous 
points of the triangulation afforded facilities, and we had no boat 
on the lake. 

«¢ The land on the north of Bear-river Bay, ten miles wide to 
the base of the hills, is a clay barren. Numerous springs issue 
from the hills, which soon sink. They are all more or less brackish, 
but seem to answer well for cattle. There is fine pasturage in the 
high grounds. The clay-flat has numerous buttes about six feet 
above the lower plain: these are islands left by the washing down of 
the original level, and have nearly perpendicular sides. The lake 
waters are driven by storms over the flat and wash off from the 
buttes, which will soon disappear. Drift-wood is found some miles 
from the present shore. Light carriages can be taken over the 
flat near the Bear River outlet and along the shore: the shore in- 
tervening, to the hills, is soft and impracticable. In the Salt Lake 
Valley, on the Bear and Malade rivers, is some excellent soil for 
grain. Fifty miles could be irrigated, but the expense would be 
heavy in constructing a dam at the “ gates’’—that is, where Bear 
River breaks through the Wahsatch range. The river at the ferry 
was two hundred feet wide and twelve deep.” 

«¢ Near the river, and twelve miles below the “ gates,” are the hot 
and cold springs. They issue at the foot of the flanking terrace 
of hills, and have excavated for themselves a circular hole, fifteen 
feet deep, with sloping sides and a deep channel leading into the 
meadow. ‘There are currents issuing between different strata of 
conglomerate and limestone, within a few feet of each other, of 
which one is a hot sulphur, a second warm and salt, and the third 


214 EASTERN SHORE—“‘ BULL-BOAT.”’ 


cool, drinkable water. At numerous places fine salt is brought up 
and jets of gas emitted: the salt forms an incrustation around the 
hole, and is pure enough for table use. Some small rivulets de- 
scend from the mountains as you proceed south; two of which, fed 
by springs, could be used for irrigation and ee farms of one 
hundred acres each.” 

«« Box-elder is a beautiful stream of clear, sparkling water, except 
when swollen in the spring by the rapidly melting snow. It was 
swelled into a large river during the survey, and overflowed all its 
banks. This stream, and the two south, to Bright Creek, can be 
carried over some excellent land, and made to water ground enough 
for five thousand people.” 

« Red Springs are hot waters, impregnated with iron, which is 
deposited and colours the ground crimson red: hence the name. 
The salt-flats extend from the lake to this point, and a low flat 
sweeps round to Weber River, of the clay barren character, between 
the north branch of the river and Mud Island. The river was now 
swollen, and filled several channels over this flat, entering the lake 
to the northward; and many were ten feet deep, with a swift cur- 
rent. At the mouths they shoal, but were very uncomfortable to 
ford, as we did, by wading, as the water was of the temperature of 
melting ice and snow.” 

«In the angles of Red-spur and Main range are small streams for 
farming purposes; but the Ogden river, as it bursts out of a narrow 
caiion, furnishes an opportunity for mill privileges and irrigating 
canals seldom equalled. On its bank is the site of Ogden City. 
Beautiful meadows, the river-level of which is subject to overflow, 
are below the junction with the Weber; and I estimate forty square 
miles on these streams as capable of cultivation.” 

««The freshets had swept off all the bridges, and embarrassed us 
much. ‘To survey the delta of Weber and cross the rivers re- 
quired boats. We obtained hides, and, by the aid of some Flat- 
head Indians, constructed a «bull-boat,”” by taking willow rods and 
laying a keel and longitudinal ribs between two stakes driven into 
the ground, marking the length, and then cross-sticks, tied with 
thongs, making the skeleton of a canoe. Three hides were sewed 
together; the sides of the centre one and one end of each of the 
others being joined and then softened in water: they were now 
stretched over the willow-work, and the seams made tight by a 
composition of melted tallow and ashes. Our wagons were taken 
into parts, a rope was stretched over a foaming, tossing, boiling 


Ackerman Lith379 Broadway NY. 


mtn GH Ores, VALLEY OF EES W Gis bh Bey Be. 


VALLEY OF THE JORDAN AND OF LAKE UTAH. 215 


torrent, at a narrow chasm, where the banks were high enough ° 
for landing, and then the frail, bending boat, by repeated journeys, 
carried us all over safe, with our baggage and instruments. It 
was a severe day’s work, however, and the risk of life and property 
very great; but the only insurance Ee be had was in the Saal 
of prudence, skill, and perseverance.”’ 

«« Between Weber and Ogden the land is too high for sengnian 
except by some rivulets which afford but a small supply. Along 
the Weber, below the cajion, is a narrow meadow strip, and to the 
south a flanking spur from the mountains, of six successive terraces, 
ten miles broad, and sloping to the lake. These are grazing lands 
for Weber settlement. South of the terrace pastures commences a 
series of creeks of bright, sparkling water, that irrigate a strip of 
land averaging two miles in width, and extending to the vicinity 
of Hot Spring, three miles from Salt Lake City. This includes 
Miller’s and Session’s settlements, and is covered with lovely fields 
and gardens. From the hot spring to the city are numerous warm 
fountains that deposite gypsum and other sulphates. These waters 
give delightful baths, but destroy the fertility of the soil. The 
south shore-line of the lake, from the mouth of Jordan River, was 
measured by the chain on the soft, sandy beach and barrens. ‘The 
line of the Jordan river was previously chained, and I passed over 
the traverse range with the triangulation in the valley of Lake 
Utah. From the Jordan caiion to Dry Cotton-wood is a grazing 
range. At the outlet of the lake there is a reed marsh which, by 
early cutting the dense growth, a pretty hay can be made. It 
will be difficult to obtain irrigable land until we reach the Spring 
ereek, and we have to rely on the American Fork for water to 
irrigate with. A beautiful and wide bottom land les along the 
lake shore, for some miles under the control of this stream; and 
from the crossing to the heads of Pomont-quint is a rich alluvial 
soil, mixed with vegetable mould. A series of rolling, round hills 
now occur between the Pomont creek and Timpanogas, well grassed 
for cattle ranges. On the Timpanogas bottoms wheat grows most 
luxuriantly, and root-crops are seldom excelled. A continuous 
field can be made thence to the Wa-ke-te-ke creek, and the lovely 
Utah valley made to sustain a population of more than a hundred 
thousand inhabitants. The west of the lake is grazing land, and 
a road on this side to the southern settlements is the natural line 
of travel from the capital city below. Warm springs issue near 
the outlet, and the mists from the cooling waters give at all sea- 


216 SUMMARY OF THE WORK. 


~ sons the appearance of rising smoke and steam from a manufac- 
turing hamlet. <A limestone quarry is located here. The water on 
the west side of the lake is bold, and much of the way a towline 
might be used to propel. boats. The Jordan is too crooked and 
shallow in places for boating. A magnificent water-power exists 
at the caiion, very accessible for teams; and here the river could 
be led out along the western or eastern base of the hills, for either 
manufacturing or irrigating purposes.” 

The following summary exhibits the amount of work done in 
prosecuting this examination: 


1. The selection and measurement of a base line, six miles in 
length. . | 

2. The erection of twenty-four principal triangulation stations, the 
lumber for many of which was hauled a distance of upward of 
thirty miles. Many of these, put up in the fall of 1849, had 
to be renewed in the summer of 1850, having been torn down 
and used for fuel by the Indians during the winter, as well as 
by some of the inhabitants, who probably supposed they had 
already fulfilled the purpose for which they were erected. 


Miles. 
3. The survey of the Great Salt Lake, the shore line 
of which, exclusive of offsets, extends to.............. 208 
4, The survey of the islands in the lake...............000. 96 
oe Snevarrey of Utalt lake.7i. F405. 202.052, eae 16 
6. The survey of the River Jordan connecting the two 
lakes and some tributaries..........cccccceccscccscececs 50 
Making. in all... c.temuscien por bastecnnsnd-eh sae 513 


7. The observations from different triangular stations, extending 
from the northern extremity of the Salt Lake to the southern 
boundary of the valley of Lake Utah, comprising an area of 
more than five thousand square miles, and involving the neces- 
sity of traversing a large extent of country, both by land and 
by water. 


The triangulation of the valley south of the Salt Lake and the 
observations for the azimuth of the base line were finished on the 
12th of August, and the time until the 28th busily occupied in 
preparations for our return. 


I had. determined, if possible, to find a practicable route to the 


MR 


“dined SHS WHL #0 BLNOW 4H Woud N¥CMOr SIL 42s 


arf na ares ryt c mae 
KeMnnoal(G/,¢ yu | imeutiaxy : | 


a 


MORMON HOSPITALITY—HOMEWARD JOUNREY. D1T 


southward of that now crossing the mountains by the South Pass. 
I therefore disposed of all my wagons, and such instruments and 
public property as were no longer necessary, by selling them to 
the Mormon authorities; and arranged,for the transportation of 
the baggage of the party entirely by pack-mules. The horses 
purchased from the Uintah Indians the past winter enabled me to 
mount a force sufficiently numerous for self-protection upon the 
projected route—part of which lay through the common battle- 
ground of the Sioux, Snakes, Utahs, Blackfeet, and Crow Indians. 

Before taking leave of the Mormon community, whose history has 
been the subject of no little interest in the country, I cannot but 
avail myself of the opportunity again to acknowledge the constant 
kindness and generous hospitality which was ever extended to the 
party during a‘sojourn of rather more than a year among them. 
The most disinterested efforts were made to afford us, both per- 
sonally and officially, all the aids and facilities within the power 
of the people, as well to forward our labours as to contribute to 
our comfort and enjoyment. Official invitations were sent by the 
authorities to the officers of the party, while engaged in distant 
duty on the lake, to participate in the celebration of their annual 
jubilee, on the 24th of July, and an honourable position assigned 
them in the procession on that occasion. Upon our final de- 
parture, we were followed with the kindest expressions of regard, 
and of anxious hopes for the safety and welfare of the party upon 


its homeward journey. 


CHAPTER IX. 


HOMEWARD JOURNEY.—EXPLORATION OF A NEW PASS THROUGH THE 
ROCKY MOUNTAINS. 


Wednesday, August 28.—Having completed our arrangements, 
we left the city of the Great Salt Lake for home, and encamped 
at the mouth of the «« Big Kanyon,” which affords a pass through 
the Wahsatch range of mountains to the plains beyond. 

Thursday, August 29.—Morning fine and cool. A train of 
Mormon wagons, just arrived from the States, is encamped near us. 
Our road to-day is up the Golden Pass, through a caiion formed 


218 PASS THROUGH THE WAHSATCH MOUNTAINS. 


by Big Kanyon Creek, and which has lately been opened and 
worked by the Mormons, who demand a small toll on each animal, 
to be devoted to its improvement. The ascent is not so abrupt 
as I had anticipated. ‘The valley is very narrow, with bold escarp- 
ments on either side, scarcely affording space between them for 
the passage of the turbulent little mountain-stream, which, with 
its bright, flashing waters, comes tumbling down the pass with a 
cheerful, murmuring sound, producing, after the dead silence of 
the barren plains and dreary sand-flats of the lake, a sensation 
peculiarly pleasant and refreshing. 

The road, which is very crooked, and in many places even 
dangerous, passes over a friable sandstone, underlying a heavy 
stratum of limestone. Cedar, oak, maple, service-berry, aspen, 
bitter cotton-wood, and willows are found in the pass; and I ob- 
served several fine vines of the wild hop, loaded with fruit. Much 
heavy grading, expensive side-cutting and walling, besides inclined 
planes, would be required to render this pass at all eligible for the 
passage of a railway. A good wagon-road, however, can here be 
made, and at a moderate expense. The great obstacle to the use 
of all these mountain passes is the vast accumulation during the 
winter of snow, which, drifting over from the bordering heights, 
effectually blocks up the valleys, not unfrequently to the depth 
of thirty feet; thus rendering them impassable from five to six 
months in the year. : 

As the sun went down, the temperature became sensibly lower, 
and at nine o’clock the thermometer stood at 46°. Observations of 
Polaris gave for latitude 40° 45’ 40’".5. Day’s march up the pass, 
seven miles. 

Friday, August 30.—Morning clear and cool. Thermometer 
at sunrise, 52°. Our road continued up the Big Kanyon Creek, 
(crossing its south fork) for five miles, when we reached the summit 
of the range, and struck upon Bauchmin’s Creek, a branch of Hast 
Kanyon Creek, which latter is a tributary of the Weber. Lati- 
tude by meridian observation at summit, 40° 44’ 48’. The val- 
ley is here from two to three miles broad, and near the summit 
several large pines are growing, as yet undisturbed by the 
emigration. Scrub-oak and aspen constitute the predominant 
growth. | 

The road continues up the valley of Bauchmin’s Creek, crossing 
several small affluents, until reaching a main fork coming in from 
the right. This stream, which is six feet wide and two feet deep, 


PARLEY S PARK—CAMASS PRAIRIE. 919 


( 


heads in a range of hills three miles to the south-west, whence it 
issues with a beautifully clear and rapid current, and, crossing the 
valley, joins the main stream and flows west and north-west into 
the Weber. Where it issues from the hills, it enters a lovely and 
fertile circular meadow, about three miles in diameter and skirted 
with trees. A couple of miles north-east, a trail passes over the 
hills to the Provaux, a tributary of Lake Utah, six or eight miles 
distant. Crossing this beautiful little prairie, which is called 
Parley’s Park, and passing around the head of a noble spring on 
our left, we crossed Bauchmin’s Creek, here about twenty feet 
wide and two deep, with a rapid current and clear, cold water. 
Following up a dry channel for two miles, we encamped upon a 
little spring branch with plenty of fine grass. 

The country is becoming more level and the valley much wider. 
The ascent for the last two miles is quite gentle, and the land ex- 
cellent. Wheat could be raised in large quantities on the prairie 
land which lies on our left, all the way from Bauchmin’s Creek. 
Distance from Salt Lake city, thirty miles, Latitude, 40° 43’ 04’7.8. 

Saturday, August 31.—Starting the train on the road, with direc- 
tions to encamp at the ford of the Weber, I made a detour to the 
right, with a small escort, to examine a prairie called Camass Prai- 
rie, through which a level and practicable route was said to exist 
between the heads of the Weber and the Timpanogas. 

Following for about five miles a south-easterly course up the val- 
ley of Silver Creek, a tributary of the Weber, we left it at a land 
where it comes from the south-west, and ascended a ridge, or 
divide, and descending it on the opposite side, through a ravine on 
its eastern slope, about a mile in length, we came into the wide 
valley of a small stream flowing into the Timpanogas toward the 
south-east. Crossing this, we ascended a long slope to the top of a 
broad level ridge, on the eastern side of which the Timpanogas 
itself flows southwardly, and finally discharges itself into the Utah 
Lake. From this point, Timpanogas Peak, in the vicinity of the 
lake, bears south, 10° west. At the head of a dry ravine putting 
into the Timpanogas, a meridian observation was taken for the 
latitude. Duchesne’s Pass bears south, 70° east, and the heads of 
Timpanogas and Weber rivers, north, 70° east. 

Leaving this summit, we struck north-east about four miles, and 
descended the bluffs bordering the south-western side of Camass 
‘Prairie until we reached the plain, which we found to be a most 
lovely, fertile, level prairie, ten or twelve miles long, and six or 


220 HEADS OF THE TIMPANOGAS, WEBER, AND BEAR RIVER. 


seven wide, and extending north-west and south-east from the Tim- 
panogas to the Weber. At the south-eastern end of the prairie, 
the Timpanogas breaks forth from a range of lofty mountains, and 
skirts the edge of it, passing near the base of some high hills on 
the south-east ; while from near the same point, the most southerly 
branch of the Weber issues also and crossing to the western end 
of the same prairie, discharges its waters into the main stream. 
This latter, coming through a deep cafion from the north-east, 
bounds the prairie on the north-west, and winding its sinuous 
course through a wilderness of willow and cotton-wood thickets, 
pursues a north-western direction for about fifteen miles to the 
point where it is crossed by the road. | 

The pass made by the Weber through the mountain, although 
narrow, is said to be practicable by one of the guides, who passed 
through it in former years with a train of pack-mules. 

Several little streams of pure, clear water wind through this fer- 
tile prairie, cutting small, deep channels for themselves in the rich 
alluvial soil; their existence being only discernible from the in- 
creased height and luxuriance of the grass upon their borders, and 
occasional clumps of willows flourishing along their edges. The 
prairie, from one end to the other, from the Weber to the Timpano- 
gas, 18 one level plain, covered ih a heavy growth of rich grasses, 
and affords a passage from one stream to the other as perfect as 
could be desired. Bear River is said by our guides to take its rise 
in the same mountain with the Weber and the Timpanogas. Should 
such be the case, (and there is no reason to doubt it,) an easy com- 
munication can be obtained by means of the valley of that stream 
into this prairie, and thence down the valley of the Timpanogas 
into that of the Utah Lake. 

The grade down the Timpanogas is disnaived to be easy, and 
the cafion through which it descends to the level of the Salt Lake 
basin to be sufficiently wide for the construction of a road. Sucha 
route would obviate many difficulties which must be encountered in 
descending either of the only other two practicable cafions through 
the mountain in the vicinity of Salt Lake City, or in going 
through that of the Weber River. When once the level of the 
basin is attained, the way from Utah Lake, either north or south, 
appears to be open. 

In the event of any exploration for a railroad to California or 
Oregon, upon a route so far north, a careful examination of the 
country from the point where the main emigration-road strikes 


ROUTE OVER THE WAHSATCH MOUNTAINS. 221 


Bear River, up that stream to its source, or at least to a point 
where a crossing could be obtained over to the heads of the Weber, 
thence down that stream into the Camass Prairie, and thence, cross- 
ing the prairie, down the Timpanogas to the Utah Valley, offers, I 
think, the most feasible mode of crossing the formidable obstruc- 
tion presented by the Wahsatch range of mountains. 

Fording the Weber at the north-west end of the prairie, we fol- 
lowed down its right bank, recrossed, and struck once more into 
the beaten track, which we followed in pursuit of the train. 

The weather had been very threatening during the afternoon, 
and the lurid clouds and muttering thunder gave token of the ap- 
proach of a‘heavy storm. About dark the tempest burst upon us. 
Rain fell in torrents, intermingled with hail; and, as it increased 
in violence, was accompanied by a darkness so profound that it 
became impossible to distinguish the road, or even the horses upon 
which we were riding. The lightning blazed with such intense 
brightness around us that we became completely blinded, and the 
storm driving furiously in our faces, it was with the greatest diffi- 
culty we could keep the track for a few yards at a time. We were 
ten miles from camp, and felt ourselves in a rather uncomfortable 
situation. The guide, who had preceded us, and had arrived at 
the encampment before the storm began, surmising what must be 
our condition, with commendable foresight despatched a couple 
of men to meet us with a lantern, by the aid of which we succeeded 
in reaching the tents about ten o’clock, thoroughly drenched, cold, 
and exhausted, having been in the saddle, without food, for more 
than fifteen hours. A basin of hot soup and a cup of strong coffee. 
soon revived us, and, wrapped in our blankets, we slept soundly, 
regardless of the tempest, which flashed and raged around us the 
whole night. 

It may be remarked here, that the Camass Prairie consists of 
most excellent land, and can be irrigated over its whole extent 
with comparatively little labour. Water for stock is abundant, 
and timber for ordinary farming purposes is plentiful and con- 
venient. The broad-leafed dock of the eastern prairies abounds 
here, and it is the only spot where I have observed it since leaving 
the Missouri. A species of geranium was very abundant, and also 
a great variety of asters. The surrounding hills are full of rolled 
stones or very coarse gravel, principally of sandstone, much 
stained by iron. 

After leaving the prairie, and crossing the Weber, several ver- 


229), WEBER FORD—RED FORK. 


tical trap dikes were observed on the hillsides; and one, in par- 
ticular, consisted of what resembled a gigantic stone fence, half a 
mile in length, regularly laid up, the stones composing it being 
entirely separate from each other, and from six to eight feet cube, 
the whole presenting the appearance of a fine specimen of Cyclo- 
pean masonry. 

The road pursued by the train crossed Silver Creek, and con- 
tinued down the right bank of the Weber for seventeen miles, 


where it was forded, and the camp made for the night. Grass — 


and fuel plentiful. The Weber bottom, as far as the mouth of 
Red Fork, five miles beyond, presents many beautiful little prairies 
on either side of the stream, fringed with belts of large cotton- 
woods, affording good locations for many small grain and stock 
farms. The rock observed from the road was principally a light- 
gray, fire-grit sandstone, with a dip of 30° to the north-west and west 
north-west, some pudding-stone, and earthy red sandstone. Day’s 
march, nineteen miles. Lat. 40° 53’ 41/7.3 Lon. 111° 36’ 26”. 

Sunday, September 1.—Engaged in drying up after last night’s 
rain, which had thoroughly soaked a portion of the baggage. In 
the afternoon the arms were cleaned and inspected, and the men 
engaged in firing at a mark. 

Monday, September 2.—Last night was quite cold, and at sun- 
rise the thermometer stood at 33°, with a heavy fret covering the 
ground. The horses took a icknapiede during the night, being 
frightened probably by a bear or a wolf, and this morning were 
very wild and difficult to catch. The cattle evince an obstinate 
determination to run back on the road, and require a constant and 
strick guard to prevent them. 

The road continued down the valley of the Weber, now reduced 
to about a mile in width, being confined on the right by high 
rounded hills, with gray sandstone cropping out, with a dip of 30° 
N.N. W. Crossing a small affluent, called Morin’s Creek, and a 
short distance farther on another, for which the mountaineers have 
no name, it not being deemed worthy of notice by them, as no 
beaver had ever been found in its waters, at five miles we 
crossed Red Fork, at its junction with the Weber, which is here a 
clear, rapid stream, one hundred and twenty feet wide, and two 
and a-half deep in the channel. At the junction, meridian obser- 


vations gave for the latitude 40° 57’ 41’... Distance from Salt 


Lake City, fifty-four and a-half miles. 


At this point the old road-turns off to the right in nearly a site 


RED FORK—MORMON EMIGRANTS. 293 


-easterly course, and follows up the valley of Red Fork: the ascent, 
except for the first three miles, in which the rise is tolerably rapid, 
is moderate, enough so for a railroad, and the valley sufficiently 
wide and direct for such a purpose, without the necessity of resort- 
ing to curves of a small radius. 

The valley of this stream seems to have been the result of some 
convulsion forming an anticlinal axis, the strata on either side of 
the caiion dipping in opposite directions; that on the left to the 
north-west, and the other to the south-east, with an elevation gradu- 
ally increasing as we ascend until it reaches 45°. High, perpen- 
dicular cliffs of red sandstone conglomerate, intermingled with lime 
and quartzose pebbles, and occasionally fine-grit sandstone, were 
continuous on the north side of the cafion the remainder of the 
day’s travel. ‘These strata varied from one to thirty feet in thick- 
ness, and near the mouth of the fork the cliffs were from one hun- 
dred and fifty to two hundred feet in height, with lofty peaks of 
the more indurated portions of the rock towering toward the sky 
in a wild and picturesque manner. 

As we advanced up the valley, several unconformable strata of 
horizontal sand and pudding stone were observed overlying the 
_ dipping rock, the red sandstone intermingled occasionally with 
the yellowish-gray, yellowish, and white varieties. To the right 
the hills are rounded at their summits, and their sides for the most 
part are covered with soil, through which the rock occasionally 
appears. Water, grass, and wood are sufficiently abundant for 
camping purposes. Toward the lower end of the valley, wild .cher- 
ries and wild hops of the best quality abound in great profusion. 

Ninety-five wagons were met to-day, containing the advance of 
the Mormon emigration to the valley of the Salt Lake. Two large 
flocks of sheep were driven before the train, and geese and turkeys 
had been conveyed in coops, the whole distance, without apparent 
damage. One old gander poked his head out of his box and hissed 
most energetically at every passer-by, as if to show that his spirit 
was still unbroken, notwithstanding his long and uncomfortable 
confinement. The appearance of this train was good, most of the 
wagons having from three to five yoke of cattle, and all in fine con- 
dition. The wagons swarmed with women and children, and I 
estimated the train at one thousand head of cattle, one hundred 
head of sheep, and five hundred human souls. 

Our day’s march was only sixteen and a-half miles. Several 


994 THE COTTAGE—ECHO CREEK. 


circum-meridian observations of Altair, gave for latitude of the 
camp 41° 2/ 27/.26; long. 111° 30’ 34”. 

Tuesday, September 3.—Ther. at sunrise, 31°. The road con- 
tinues to follow up the valley of Red Fork. In about six miles we 
came in view of a remarkable little eminence in a bluff of red sand- 
stone, which almost perfectly resembled a rustic cottage, with a 
- deep-arched doorway and gently sloping roof, covered with scat- 
tering cedars. The illusion was very strong, and became more and 
more perfect as we approached, until we almost expected to see 
some one issuing from the portal to gaze upon the passing train. 
The valley of Red Fork to this point is very beautiful, beginning 
to widen considerably, and becoming more level. 

For about a mile from camp, gray sandstone takes the place of 
the red, after which the latter predominates, but not in such con- 
tinuous ledges as heretofore. Opposite «the Cottage,” a broad, 
level ravine comes in from the 8S. 8. E., which apparently heads 
somewhere near the sources of Morin’s Creek, and, from the trend- 
ing of the hills, may connect with those of the Weber itself. If 
this should prove to be the case, the circuitous route by the mouth 
of Red Fork may be entirely avoided, as well as the descent to the 
Weber, and the rise from it, both which are unavoidable by the 
route now pursued. A short distance beyond the Cottage, a 
broad ravine comes in from the N.N. E., which is, in fact, the 
main Red Fork—the eastern branch, along which the road passes, 
being called Echo Creek. 

In the forks of the two streams, at the foot of a bluff of horizon- 
tal red and gray sandstone, observed for time and latitude. We 
called the bluff «« Chicken-cock Bluff,” from the strong resemblance 
to that bird of a large cedar on its summit. Ascending this ele- 
vation, I obtained a view of the valley of Red Fork, of Echo Creek, 
and also of the ravines coming into the former from the S. S. E. 
It is highly probable that a pass may be obtained by means of one 
of these valleys over to some of the head branches of the Weber, 
or to the river itself, before it enters the Camass Prairie. It is 
worth a careful examination, as success would insure an almost 
level and very direct route through the Timpanogas Valley to that 
of the Great Salt Lake at Utah. 

On the right bank of Red Fork, just above the junction of © 
Kcho Creek, in a cliff of horizontal red sandstone, the escarpments 
were much worn and rounded, as if by the action of water, and 
wrought into strange and fantastic shapes, many of them resem- 


~ 


CACHE CAVE—SUMMIT—YELLOW CREEK. 295 


bling turrets, bastions, &e. Yesterday numerous places were ob- 
served in the high cliffs below, where a black and apparently viscid 
substance, resembling mineral tar, had oozed out between the 
strata and trickled down the face of the cliffs. This would seem to 
indicate the vicinity of coal, although no other iw of its pre- 
sence was discovered. 

From the mouth of Echo Creek to Cache Cave, a distance of five 
and a-half miles, the cliffs begin to disappear, the rocks cropping 
out only occasionally through the soil, and the hills gradually di- 
- minishing in height. From Cache Cave, (which is merely a small 
hole or grotto in a large gray-sandstone rock appearing in a low 
bluff on the left,) the route winds up the valley of Echo Creek, 
and ascends a pretty steep hill to the dividing ridge between the 
waters of the Weber and Yellow Creek, a tributary of Bear River, 
into which it discharges its waters about six miles below. 

Here it was evident that the road should not have followed the 
valley of Echo Creek at all, but should have continued up the 
valley of Red Fork, from ««Chicken-cock Bluff” to a depression in 
the hills to the northward, leading over into Yellow Creek, by 
which the route would have been shortened as well as much iin- 
proved. From this ridge the Weber Mountains can be seen 
through the Red Fork Kanyon, distant about twenty miles, and 
also the mountains beyond Camass Prairie. 

From the observations taken from this elevation, there is every 
indication that by following up the valley of Yellow Creek to its 
head, a good route may be obtained over to the waters of the Weber 
before it enters Camass Prairie, by which the whole descent of Red 
Fork can be avoided, and also the consequent necessity of the 
ascent of the Weber for twenty miles to that beautiful meadow, 
whence, as before remarked, the route to the Timpanogas is as 
level as a floor. Here the roads might fork, one leading to Utah 
Lake, by the Timpanogas, and the other, by Silver Creek, Bauch- 
min’s and Golden Pass Creeks, to Great Salt Lake City. Any ex- 
ploration, for either a railroad or a permanent mail route through 
this region, should embrace a careful examination of the country 
in this vicinity. 

Encamped on the banks of Yellow Creek, which is about three 
feet wide, with steep banks, and fringed with willows, but no timber. 
Day’s march, seventeen miles. Distance from Salt Lake City 
eighty-two miles. Lat. 41° 09’ 00.2; long. 111° 14’ 18”. 

Wednesday, September 4.—Morning quite cool. Ther. at sun- 

15 


2296 THE NEEDLES—BEAR RIVER—COAL BED—TAR SPRINGS. 


rise, 31°. In a quarter of a mile the road turns abruptly to the 
right and follows up Needle Creek, (a branch of Yellow Creek,) 
with a gentle ascent, passing a ridge crowned with broken pointed 
crags of conglomerate, which, from the acicular form of the pro- 
jections, has gained for it the name of «The Needles.”’ Ascending 
to the head of this stream, we descend into the valley of Bear 
River, the low bottom of which is here about a mile and a-half 
wide, and thickly covered with clumps of cotton-wood. ‘The river 
is about four hundred feet in width, two and a-half in depth, flowing 
with a strong current over a bed of large pebbles. Crossing this 
stream the road keeps the valley of Sulphur Creek for about two 
miles, where we halted to noon, near an excellent spring of good 
cold water. A meridian observation gave for latitude 41° 08’ 08/’.28. 

A short distance north of the road, and on the north bank of 
the creek, a bed of bituminous coal was discovered, between two 
nearly vertical dikes of light-gray coarse-grit sandstone, one hun- 
dred and fifty. feet apart, the course of which is north, 30° east. 
The outcrop was about eight feet wide by four feet thick, and was 
only visible against the south side of the north dike. It appeared 
to be quite an extensive deposite, but its depth and width can only 
be ascertained by further examination. Specimens of it, although 
much weathered, burned in the camp-fire with a clear, bright flame. 
The seam had a direction apparently the same as that of the dikes, 
with a dip of 70° 8., 70° HE. From the base of the bluff is- 
sued several sulphur springs, and south about a mile and a-half, 
a spring of Petrolium, or mineral tar, oozes from the low bank of a 
little rivulet flowing into the valley of Sulphur Creek from the 
south-west. The emigrants collect it for medicinal purposes and 
for greasing their wagon-wheels. The bank from which it issues is 
full of rolled pebbles, but no ledge is visible. 

From this point to the southward lay a broad expanse of: 
country, considerably lower than our level, stretching away toward 
the heads of Bear River. To the south-east a little park of timber 
grew near the bluffs which form what is termed the «: Rim of the 
Basin,” and from which flow the heads of Sulphur Creek: to the 
eastward a low table extended toward the heads of the Muddy, a 
tributary of the Colorado of the Gulf of California. 

Returning to our place of nooning, we again struck into the road, 
and, passing over a level country for two or three miles, we at length 
ascended the ridge dividing the waters which discharge themselves 
within the Great Basin, from those which flow into the Pacific. 


RIM OF THE BASIN—-ROUTE FROM FORT BRIDGER. DOT 


The ascent is gentle and winding. Numerous springs burst out on 
either side, near the summit of the ridge, amid groves of aspen, 
which cover the sides of the surrounding hills. 

It was with no little exultation that we reached this eastern 
barrier of the Great Basin, in which we had been floundering amid 
dreary deserts and barren mud-plains for the last thirteen months, 
and to which we now bade adieu with feelings of unfeigned satis- 
faction. Directly upon the summit, by the side of the road, was 
the fresh grave of some poor fellow who had come thus far on his 
journey to the land of promise—a land he was destined never to 
behold. 

From the top of the pass, it was evident that a much better 
location for the road could have been made from the head of Muddy 
to that of Sulphur Oreek, by which the route would have been 
rendered much more direct, and at least two formidable elevations 
avoided. 

From conversations with several individuals well acquainted with 
the country in this vicinity, especially with Major Bridger, who 
has traversed this whole region for the last thirty years, as well as 
from my own observations, which go ina great measure to confirm 
their statements, I feel convinced that the best route for a road into 
the Salt Lake Valley would be obtained by ascending Black’s Fork 
to Fort Bridger ; passing thence to Bear River, distant twenty- 
five miles; then crossing Bear River and ascending its valley until 
we strike upon the heads of a branch of the Weber; following 
down which for fifteen miles, the main stream is reached about two 
miles north of Camass Prairie. Into the latter, access through the 
upper cafion is not difficult. This, so far as can be judged from 
the information obtained, and from the partial observations we had 
ourselves the opportunity of making, would afford a route entirely 
practicable for a railroad, the chief obstacle presenting itself being 
the liability to obstruction from the snows that would be likely to 
accumulate in the cafion of the Weber. From the Camass Prairie, 
as before observed, the road might fork, the branch which leads to 
the city descending the Wahsatch range by the Golden Pass as at 
present, while the other, following down the Timpanogas, would 
enter the Salt Lake basin near to Lake Utah. 

Following a ridge for about eight miles from «the Rim of the 
Basin,’ we encamped at Red or Copperas Spring, a tributary of 
the Muddy, (an affluent of Green River,) after a march of twenty- 
six and a-half miles. The road to-day has been hilly, but good; 


* 


228 FORT BRIDGER—BLACK’S FORK—MILITARY POST. 


fuel, grass, and water abundant, and at convenient points. Lat. 
41° 13’ 46’; long. 110° 48’ 00”. 

Thursday, September 5.—Morning cool and slightly cloudy. 
Ther. at sunrise, 87°. A march of sixteen miles brought us to 
Fort Bridger, on Black’s Fork of Green River. This is a trading- 
post much frequented by the Shoshonees, Utahs, and Uintah In- 
dians, and is owned and conducted by Messrs. Vasquez and 
Bridger, from both of whem we received the kindest attention and 
every assistance which it was in their power to render. 

Black’s Fork, upon which the fort is situated, is a considerable 
stream of excellent, clear, sweet water, which rises in the Bear 
River mountains, and discharges its waters into Green River, or 
the Rio Colorado of the Gulf of California. A mile and a-half 
above the fort, it divides into four streams, which reunite two miles 
below, forming several islands, upon the westernmost of which the 
fort is beautifully located, in the midst of a level, fertile plain, 
covered with a luxuriant growth of excellent grass. Numerous 
groves of willows and cotton-wood, with thickets of hawthorn, 
fringe the margins of the streams, and afford fuel and timber for 
the necessities of man, and shelter for cattle from the inclemency 
of the winter. Black and white currants are tolerably abun- 
dant, and are now ripening upon the banks of the rivulets. ‘The 
emigrant road forks here, one branch leading to Fort Hall, by the 
Soda Springs, and the other, pursuing a more southerly course, 
leads to the City of the Salt Lake, the distance to which by the 
travelled road is one hundred and twenty-four and a-half miles: 
this, may be materially shortened by a judicious location of the 
route. ; 

From its position with regard to several powerful Indian tribes 
which imhabit this region, Fort Bridger offers many advantages 
for the establishment in its vicinity of a military post. It occu- 
pies the neutral ground between the Shoshonees and the Crows on , 
the north; the Ogallalahs and Sioux on the east; the Cheyennes 
on the south-east; and the warlike tribe of the Utahs on the south. 
A competent force established at this point would have great in- 
fluence in preventing the bloody collisions which frequently occur 
between these hostile tribes, and would afford protection and aid 
‘to the great tide of emigration which, for years to come, must con- 
tinue to flow in one ceaseless current to Oregon and California. 

The party remained here several days, to readjust the packs, and 
to complete the final arrangements for crossing the plains. The 


NEW ROUTE THROUGH THE ROCKY MOUNTAINS. 229 


trunks and heavy baggage were left in charge of Major Bridger, 
to be forwarded by a Mormon train from the city; Governor 
Young having kindly engaged to see that they were safely trans- 
ported to St. Louis. Carrying with us, therefore, only such arti- 
cles as were absolutely indispensable, we prepared for our de- 
parture. During our stay, daily observations were taken, the re- 
sults of which gave for the lat. 41° 18’12/.8; andlong. 110° 82/ 23/”. 

On the 7th the mail arrived, bringing the unlooked-for intelli- 
gence of the death of the President of the United States. 

Before leaving Salt Lake Valley, it had been determined not to 
return by the beaten track, but to endeavour to ascertain the prac- 
ticability of some more direct route than that now travelled to the 
waters of the-Atlantic. If it should prove to be practicable to 
earry a road across the north fork of the Platte, near the Medi- 
cine Bow Butte, and, skirting the southern limit of the Laramie 
Plains, to cross the Black Hills in the vicinity of the heads of 
Lodge-pole Creek, and to descend that stream to its junction with 
the South Fork of the Platte, nearly a straight line would thus be 
accomplished from Fort Bridger, and the detour through the South 
Pass and the valley of the Sweetwater, as well as all the rugged- 
ness of the Black Hills, upon that line, be entirely avoided. The 
country through which the proposed line would pass was repre- 
sented as entirely practicable and as affording every probability 
of success. y 

Major Bridger, although at considerable sacrifice of his own in- 
terest, with great spirit offered his services as guide, he being well 
acquainted with the ground over which it was my desire to pass. 
The offer was most cheerfully accepted; and as our route would lay 
directly through the war-ground of several powerful Indian tribes, 
care was had fully to equip the party with arms and ammunition 
necessary for our defence. 

Tuesday, September 10.—We left Fort Bridger in the afternoon, 
and proceeding about five miles down Black’s Fork, encamped in 
a small meadow upon the right bank, with good grass for the ani- 
mals. A merchant-train for Salt Lake passed us during the day, 
from which I procured some sugar and coffee, of which articles we 
were nearly destitute, the supply of Salt Lake City having been 
exhausted long before our departure. 

Wednesday, September 11.—Ther. at sunrise, 23°. Our route 
to-day follows the emigration road down Black’s Fork, which is 
twice crossed. In eight miles from camp we crossed Smith’s Fork, 


230 VALLEY OF GREEN RIVER. 


a pretty little stream of good water, which also takes its rise in the 
Bear-river Mountains, near the heads of Black’s Fork, into which 
it falls about twelve miles below Fort Bridger. Continuing in the 
valley of Black’s Fork, after a march of nearly eighteen miles en- 
camped on its right bank, about half a mile above the mouth of the 
Muddy, another affluent coming in from the left. Lat. 41° 287 
56/’.88; long. 110° 18’ 50”. 

Since leaving the basin and entering the valley of Green River, 
a remarkable change in the face of the country is apparent. In- 
stead of the disturbed and upheaved rocks which characterize that 
region, flat tables or terraces of horizontal strata of green and 
blue sand and clay, and sandy conglomerate, or agglutinated sand, 
now form the principal feature of the country, standing alone, like 
island buttes, amid the barren plains, or forming escarpments 
which alternately impinge upon the banks of the winding streams. 
These tables, which extend from the rim of the basin to the South 
Pass, and thence to Brown’s Hole on Green River, are apparently 
the result of a deposite in still water. The layers are of various 
thicknesses, from one foot to that of a knife-blade, and the hills are 
fast wearing away under the influence of the wind and rain. 

The whole country looks as if it had, at one time, been the bot- 
tom of a vast lake, which, bursting its barrier at Brown’s Hole, had 
been suddenly drained of a portion of its waters, leaving well- 
defined marks of the extent of the recession upon the sides of these 
isolated buttes. As the channels became worn by the passage of 
the water through the outlet into Green River, another sudden de- 
pression followed, and the same operation was repeated at still a 
lower level. There are three well-defined levels, and the same ap- 
pearances of horizontal water-lines occur here as were noticed upon 
the hillsides of the islands in the Great Salt Lake; save that in the 
latter case they are more numerous and closer together, and the 
subsidence of the waters appears to have been more gradual. The 
surface of the ground was strewn with fragments of obsidian, black, 
shiny pebbles, flints, and white, yellow, and smoky quartz. 

A high wind from the W. 8. W. blew up clouds of dust, 
and, at every turn of the road, announced the approach of crowds 
of emigrant-wagons, wending their way to the Mormon valley, with 
droves of cattle and sheep, whose fat and thriving condition, after 
so long a journey, was the subject of general remark, and excited 
universal admiration. 

Thursday, September 12.—Last night was very cold, and at sun- 


‘<THE CHURCH —PILOT BUTTE. 931 


rise the thermometer stood at 22°. Shortly after leaving camp we 
ascended a hill opposite the mouth of the Muddy, and reached a 
remarkable isolated butte, to which the traders have given the name 
of «The Church.” It is composed of green and brown indurated 
clay and brown sandstone, seamed and furrowed by the elements 
into fantastic and picturesque forms, resembling somewhat the 
ruins of a huge antiquated castle. Following the old road for a 
couple of miles farther, over a level country of sand and clay, al- 
most denuded of vegetation, except occasional patches of artemisia, 
which seems to claim as its peculiar property soil where nothing 
else will grow, we came in sight of a high butte, situated on the 
eastern side of the Green River Valley, some forty miles distant : 
a landmark well known to the traders, and called by them Pilot 
Butte. 

Leaving the emigrant road here, we struck for this prominent 
elevation, passing, on our way to it, over a barren plain formed by 
the washings from the hills, which had covered up every vestige of 
vegetation, and presented an aspect of dreary desolation saddening 
to the heart. Passing a small brackish spring, which issues from 
the base of some coarse brown sandstone rocks, rising abruptly 
from the level plain to the height of twenty feet, we halted to take 
a meridian observation, which gave for latitude 41° 31’ 08/7.5. 
The mouth of Ham’s Fork bears north 20° west, distant four 
miles, the mouth of Black’s Fork being about thirty miles below. 
From this point the land descends gently for twelve miles to Black's 
Fork, which we crossed at an excellent ford, and encamped on its 
left bank, amid thickets of willows, and fine grass for our ani- 
mals. The stream was about a hundred and forty feet wide, 
and easily forded. In times of high water it is probably from two 
hundred to two hundred and fifty feet wide, with a depth of from 
six to eight feet. 

The vegetation, to-day, has been very sparse, and consists princi- 
pally of dwarf sage and grease-wood bushes, with low bunch-grass. 
Black currants abound on the banks of the stream. In many 
places the ground has been thickly strewn with gravel and pebbles, 
mostly siliceous, with occasional appearance of argillaceous hme- 
stone. Day’s travel, twenty-five and a-half miles; total from Fort 
Bridger, forty-seven and a-half miles. 

Friday, September 13.—Morning overcast, and threatening rain. 
Ther. 47°. As we were in the act of leaving the camp-ground, 
an alarm was given that a body of armed men on horseback was 


232 SHOSHONEE WARRIORS—RABBIT HOLLOW. 


charging down upon us at full speed. The pack-mules and loose 
animals were immediately driven back into the bushes, where they 
could be more easily defended, while, accompanied by Major Brid- 
ger, I advanced to the bank of the stream to reconnoitre. We 
soon ascertained that the party consisted of a band of some twenty 
Shoshonees, who were out upon a hostile expedition against the 
Utahs, and that, mistaking, as they said, the smoke of our fires for 
that of their enemies, they had charged down upon us, in full ex- 
pectation of effecting a surprise. As soon as they discovered their 
mistake, they crossed the creek in compliance with our invitation, 
and greeted us in the most friendly manner. The party was armed 
in a most heterogeneous way, some having rifles, others old bayonets 
fastened upon the extremities of long poles, and the rest bearing 
only bows and arrows, with a little round shield suspended from 
their necks. ‘They were, for the most part, well-mounted upon 
small, but apparently Sy cie horses. 

There being no longer any occasion for alarm, the acne were 
driven from hee cover, and leaving our red fidelte to pursue 
their own course to the Capes we resumed our march, although 
it had commenced to rain quite heavily, with every prospect of a 
stormy day. Recrossing Black’s Fork, we followed up a small 
depression, and in about a mile reached by a gentle ascent the 
summit of a long ridge, stretching eastwardly toward the valley 
of Green River. Crossing this ridge we struck upon the broad 
valley of another affluent of Black’s Fork, which we pursued for 
about six miles to its head, when we reached the ««divide’’ between 
that stream and the waters of Green River. From the “divide” 
we descended a long and winding ravine, called Rabbit Hollow, 
which joins the valley of Green River two miles above the mouth 
of Bitter Creek, a considerable stream flowing into that river from 
the eastward. The lower part of Rabbit Hollow will require to 
be partially worked, to avoid the bends of the dry bed of a stream 
which winds from side to side of the narrow bottom; but, with this 
trifling exception, an excellent wagon-road can be traced from 
Green River at this point to Fort Bridger, and by a very direct 
route. The north side of this ravine, near its mouth, is flanked 
by lofty vertical cliffs of indurated green clay and shales, overlaid 
by horizontal strata of a soft, yellowish sandstone. The same 
formation occurred on the eastern side of Green River; and the © 
turreted appearance of the crumbling sandstone cliffs, four hun- 
dred feet in height, was in a high degree imposing and picturesque. 


GREEN RIVER—BITTER CREEK VALLEY. 233 


At this point we crossed the river, between banks some thirty 
feet high and eight hundred feet apart. The deepest water found 
in crossing was three feet, but in high stages it is a formidable 
stream, and will require to be ferried. The bottom is about a 
mile in width, and is covered with thickets of willows and abun- 
dance of grass. Following down the left bank of the river, we 
encamped in a clump of bitter cotton-woods, where, owing to the 
suspicious vicinity of our Indian acquaintances of the morning, 
the animals were securely picketed and the guard doubled for the 
night. Two or three Indians had followed us all day, and, after 
partaking of a hearty supper, left us just before dark—as they said, 
to rejoin their companions. We did not, however, feel called upon 
to place the most implicit confidence in their statements, and the 
necessity of increased vigilance during the night was impressed 
upon the guard, lest, upon waking in the morning, we should find 
that our animals had disappeared. The weather began to clear 
before night, and enabled us to obtain an observation of Altair, 
which gave for latitude 41° 80’ 44/’.5, longitude 109° 38’ 40”. 
Day’s march, thirteen and a‘half miles; and sixty-one miles from 
Fort Bridger, by the route we have pursued, which, with a little 
care, can be much improved and shortened. 

Saturday, September 14.—Ther. at sunrise, 46°. <A dense 
fog concealed every thing from view. The night had passed 
without disturbance, which led me to suppose that the Indians 
yesterday had really no design upon our horses, as I had suspected. 
As it was impossible to move while the fog continued, the men 
were occupied in cleaning their arms, wetted by yesterday’s rain. 
Black currants and buffalo-berries abound upon the banks of the 
river. The latter growin great profusion upon a shrub about 
fifteen feet high. The berry is about the size of a pea, of a bright 
scarlet colour, and:contains but one seed. It has a pleasant acid 
flavour, and would make an excellent jelly. The Indians are said 
to be extravagantly fond of it. | 

Before noon the fog lifted, with the promise of a fair day; 
we saddled up and started on our journey, directing our course up 
the valley of Bitter Creek, which enters Green River about a mile 
below. Crossing the bottom of Green River Valley, we passed the 
mouth of Bitter Creek, and ascended it the whole day’s march. 
For the first half-dozen miles, the valley is much cut up by deep 
gullies, worn by the water from the rapidly decomposing hills on 
each side of the stream. ‘These, however, present no serious ob- 


934 BITTER CREEK—BITUMINOUS COAL BEDS. 


stacle to the easy construction of a good road: the soil being 
porous, and mixed with a large proportion of gravel, can be readily 
graded. So rapid is the disintegration of the sandstone bluffs, 
and so constant the wash of the soil, that the valley, so far, is 
almost entirely destitute of even a spear of grass, and presents a 
most desolate appearance. 

After passing this mauvaise terre the route is quite level, pre- 
senting no obstruction of consequence, except the necessity of occa- 
sionally crossing the creek, which will be somewhat difficult, as the 
bed of the stream is not unfrequently at a depth of twenty feet below 
the level of the bottom, with perpendicular banks on each side. But 
the earth is easily removed, and very little labour will render these 
crossings by no means formidable. Ata point thirteen miles from 
the mouth of Bitter Creek, we found a bed of bituminous coal 
cropping out of the north bluff of the valley, with every indication 
of its being quite abundant. 

For the first seven or eight miles after entering the valley, the 
formation was similar to that of Rabbit Hollow and Green River, 
and the strata horizontal: they then began to dip gradually to the west 
and north-west, until, at this point, the elevation had reached 20°. 
The coal was of the same character as that found on Sulphur 
Creek, near the crossing of Bear River, alternated in beds of 
various thickness, from a few inches to several feet, with yellowish 
and light-gray sandstones. Major Bridger informed me that, 
about a mile from the mouth of the creek, a large bed existed, 
which, from his description, resembled lignite, but which, owing 
to other occupations, I had passed without observing. One of the 
men reported to me that he had noticed it, and had seen a piece 
of coal lying in the bed of the creek as long and as thick as a 
man’s body. This had apparently fallen from an outcrop in the 
south bank, which was about four feet in thickness. Major 
Bridger also stated that a similar bed is to be found to the south 
of the mouth of Black’s Fork, that he had used it for years, and 
that it burned freely, with a clear, white blaze, leaving little re- 
siduum, except a small white ash. From this outcrop, as we 
ascended the valley, the strata became more nearly horizontal ; 
and at the termination of the day’s march, some four miles beyond, 
had nearly recovered its level position. 


The valley of Bitter Creek generally has but little snow in it . 


during the winter, and was formerly a rendezvous for the trappers 
and traders, on account of the abundance of buffalo. None of 


VALLEY OF BITTER CREEK. O35 


these animals are now to be seen in the region. The only vegeta- 
tion, to-day, has been a little dwarf artemisia, grease-bush, rabbit- 
bush, salt-grass in the narrow strips of bottom-land, and an occa- 
sional dwarf cedar on the bluffs. Appearances indicate that the 
bed of the stream, which has no banks above the general level of 
the bottom, has been completely filled by the spring rains, over- 
flowing the low grounds and carrying down immense quantities of 
soil, which has been deposited below, upon the broad flats of Green 
River. The ridges on each side of us to-day have been principally 
composed of horizontal gray limestone and disintegrating aggluti- 
nated sand; the escarpments, rounded into fantastic forms of bas- 
tions, buttresses, and turrets, by the action of the winds and the 
rains, were in many cases quite beautiful. Day’s march, seven- 
teen miles. Lat. 40° 34’ 41/7.8; long. 109° 23’ 9”. 

Sunday, September 15.—Morning slightly cloudy. Ther. at 
sunrise, 44°. The water being very indifferent, and wood difficult 
to be obtained, it was determined to make a march to-day to a 
spring supposed to be about ten miles up the valley. Bitter 
Creek, at this camp, flows in a bed twelve feet below the plain of 
the bottom, and the water in its present stage is about sixty feet 
wide by six inches in depth. But the accumulation of large piles 
of flood-wood shows that during the spring freshets, the whole 
valley, here about one thousand feet wide, is completely covered 
with water. The general course of the valley for ten miles was 
north, 74° east, over a very level flat. About halfway of this dis- 
tance the sand-cliffs disappear, and the valley opens wide, with 
rounded hills on each side, much less elevated than the bluffs below. 
At this point the stream forks, one branch coming in from the 
south, with a wide level valley. About thirteen miles above, it 
forks again, both branches heading within two miles of each other, 
in a small mountain, in which Red Fork, another affluent of 
Green River, flowing into it below Brown’s Hole, also has its 
sources. One of the guides says that there is a good route by Red 
Fork to Green River, and thence to the heads of Bear River. 

A short distance on our left lay a small, shallow lake, some six 
miles in circumference, formed by the drainage from the hills on 
the north and east, and very probably by some springs in the 
vicinity, as the waters appeared clear and fresh. 

After passing that point, the hills began to approach each other 
more closely, and the ascent became rather more abrupt, with 
numerous gullies, or drains, crossing it at right angles, rendering 


236 BITTER CREEK—EXTENSIVE COAL BEDS. 


the trace, heretofore unexceptionable, rather more unfavourable. 
Frcm the lake we followed the right bank of the east fork of the 
creek for six miles, when we again crossed it, and one mile beyond 
we broke out, from a bed twelve feet thick, some beautiful speci- 
mens of excellent bituminous coal, which burned in our camp-fires 
with a clear yellow flame. In the sides of the ravines putting into 
the valley are abundant beds of this mineral. 

The rock is very much of the same character as that observed 
below—red, white, and yellow sandstone, with a slight dip to the 
east and north-east. In some places, nodules of rock, deeply tinged 
with oxide of iron, were frequent. The rock is very porous, and 
rapidly disintegrating under the action of snow and frost, and is 
washed down into the valley by the spring rains in great quanti- 
ties, so that vegetation, except grease-wood and salt-grass, is very 
sparse. | 

Continuing up the left bank seven miles farther, we encamped 
in the forks, after a journey of twenty-five miles, near two cold 
sulphur springs which issue from the base of a cliff of light-gray 
coarse-grit sandstone, having fragments of coal interspersed, so 
as to form something like a pudding-stone of coal and cemented 
sand. 

Near the northern spring some beautiful white lichens were 
growing on the rocks over which it flows; and under the project- 
ing cliff, some highly coloured ones, with variegated hues, like the 
neck of a mallard. Currant-bushes were quite plentiful, and the 
plain near the springs has a dense growth of broad-leafed blue- 
joimt and wire-grass, among which I observed some purple asters. 
A mile and a-half north of the camp, on the North, or Evans’s 
Fork, is an outcrop of coal, ten feet thick; but most of that in 
sight appeared to be but imperfectly mineralized. Indications 
were also observed of the presence of 4ron-ore near the camp. 
Lat. of the camp, 41° 30’ 51/.4. Long. 108. 50’ 34”, 

Monday, September 16.—Our route to-day continues up the 
east or left fork of Bitter Creek. Sandstone cliffs bound the val- 
ley on the north side, in which I observed a stratum of coal, which 
was exposed for a hundred yards, and was at least ten feet in 
thickness. During the whole day’s march this mineral was met 
with in every favourable locality, and in quantities apparently 
without limit. The formation was the same as lower down the 
stream, except that the dip was north-easterly, at an angle of from 
6° to 10°. 


: HEADS OF BITTER CREEK—-BRIDGER’S FORK. oot 


Several bands of antelope were seen, and one of the hunters 
brought in a noble buck. Grouse, also, were abundant, and very 
fat. A bear, too, was seen taking a good look at us, but, not liking 
our appearance, he made off with all haste before we could get a 
shot at him. 

The trace for a road has been good, and the distance travelled 
in nearly an eastern direction was twenty-two miles. Our observa- 
tions gave for latitude 41° 28’ 9’’; longitude 108° 41’ 9”. 

Tuesday, September 17.—Our course to-day was up the east fork 
of Bitter Creek, about south 70° east, for eighteen miles, when 
we reached its head, ascending very gently to the dividing ground 
between it and the waters of Muddy Creek, an affluent of Little 
Snake River, which flows into the Yampah, about twenty miles 
above the entrance of the latter into Green River. Here we leave 
the valley of Bitter Creek altogether, having followed it from its 
mouth for seventy miles. On the level table forming the « divide” 
is a butte composed of sand and clay, a hundred and fifty or two 
hundred feet high, standing isolated and detached from the range 
of bluffs to the south and east of it, and to which, from its shape, 
we gave the name of “The Haystack.” From this landmark we 
travelled in nearly an eastern direction, gradually descending, for 
six miles, to the valley of a small branch of the Muddy, to which 
we gave the name of “ Bridger’s Fork of Muddy,”’ and encamped in 
its valley, although the water was so strongly impregnated with ~ 
alkali that the animals drank it with evident reluctance and disgust. 

The valley is here much cut off by abrupt gullies and ravines, 
formed by the wash from the hills, and in many places the ground 
is covered with a crust of impure soda to the depth of half an 
_ inch. The grass, since our-noon halt, has been very scarce, and 
our poor mules have fared rather badly. Several buffalo were seen 
to-day, and one antelope killed. Our hunters are calculating 
largely upon the sport before us as we approach the buffalo range, 
and are much excited at the prospect of once more revelling in 
their favourite fare. Day’s travel, twenty-six miles. Lat. 41° 
28’ 39/’; long. 108° 14’ 24’. 

Wednesday, September 18.—Ther. at sunrise, 44°. A slight 
<‘stampede’’ occurred among the horses last night, owing to the 
blundering of an old-buffalo bull into the camp. Several of the 
animals broke from their fastenings and fled in dismay, but were 
all ultimately recovered. 


238 RED GATE—PRAIRIE—ELK MOUNTAINS. 


Our course lay down the valley of Bridger’s Fork for three and 
a-half miles, when it opens suddenly between two high cliffs of red 
and green indurated clay, and strata of coarse-grit brown sand- 
stone, upon a vast rolling prairie, extending from the Platte to 
Snake River. To this opening we gave the name of « Red Gate.” 
The general direction of the low range bounding the western side 
of this prairie is about north-east and south-west. Upon entering 
the plain, a magnificent view opened before us. On our right, to 
the south and the east, extended the Elk Mountains, with their 
blue peaks, in which the Elk River takes its rise; to the left of 
these rose a high square butte, marking where the three branches of 
St. Vrain’s Fork enter the Snake River. On our left stretched far 
off to the northward the ascending ridge of prairie forming the «di- 
vide’ between us and the waters of the Sweetwater and the North 
Fork of the Platte, while before us, and at right angles with our 
course, ran four rolling prairie ridges, the farthest of which, about 
forty miles distant, formed the western boundary of the Park 
Mountains. Bridger’s Fork, the little stream whose valley we had 
followed to the Gate, pursued a wandering course to the south- 
east through the prairie, its existence marked only by an occasional 
clump of willows. A few buffalo bulls were quietly grazing upon 
the plain, and now and then a small herd of antelope, bounding 
away over the hills, gave life and spirit to the picture. 

The soil from this point to Muddy Creek is for the most part 
of an excellent quality, but, from want of moisture, can never be 
appropriated to any other purpose than grazing. ‘The grass, 
though thin, is very nutritious. Small sage, salt grass, grease- 
wood, a purple aster, together with bunch-grass, and, in the more 
sandy portions, small cacti, were the principal plants. The tops 
of the ridges were strewn with a flat, black gravel, with pebbles 
of yellow and white quartz. An occasional drain was crossed, 
which gave indications of having contained water quite recently ; 
but all of these were now dry. As long as the water lasted, the 
whole plain must have been covered with buffaloe and antelope, as 
the profusion of « sign” abundantly proved; but as this indispen- 
sable article was absorbed by the sandy soil, they seemed, from 
the direction of their trails, to have struck a course for the Ver- 
milion. Many large bear-tracks were also seen, making in the 
same direction. | 

After travelling six miles across this undulating prairie, we 
reached the right bank of the valley of Muddy Creek, twenty-five 


MUDDY CREEK—TRAPPERS’ TALE. 939 


miles above its confluence with Little Snake River. This stream, 


which rises in the Park Mountains, here makes a valley of four 
miles wide, and the descent to the bottom of the creek is from a 
hundred to a hundred and fifty feet down a washed and broken 
bluff of sand and clay, much worn into gullies and ravines. The 
descent is too steep, where we struck the bluff, for a good wagon- 
road; but, by adetour from the « Gate’’ of two miles to the south, 
the descent will be very much more gradual, and the greater part 
of the high, broad ridge over which we passed will be avoided. 

Upon the top of this ridge I found, scattered over the surface, 
a large number of silicefied petrifications of shells. Passing two 
remarkable little sandstone buttes, on our right, one of which was 
covered with cedars, (the first trees we had seen since leaving 
Green River,) and on our left two flat-topped whitish clay or 
marly mounds, connected by an escarpment, we encamped in a 
deep bend of the Muddy, which was fringed with willows, having 
selected the spot with the view of more securely guarding our 
animals from the nocturnal attacks of any wandering bands of 
Indians. 

We are now upon the war-ground of several hostile tribes, who 
make this region the field of mutual encounter, and increased vigi- 
lance is consequently necessary to guard against a surprise—an 
occurrence which, as one of its least unpleasant consequences, 
might leave us on foot in the midst of the wilderness. All firing 
of guns, without express permission, except in case of the most 
urgent necessity, has been strictly forbidden, and every man slept 
with his arms by his side. 

As we were reposing our weary limbs before the camp-fire, re- 
galing ourselves with a pipe, now our only luxury, Major Bridger 
entertained us with one of those trappers’ legends which abound 
as much among these adventurous men as the “yarns’’ so long 
famous among their counterpart, the sailors, on a rival element. 
A partner of his, Mr. Henry Frappe, had a party of what, in the 
language of the country, are called «free men,” that is, inde- 
pendent traders, who, some nine years before, were encamped 
about two miles from where we then were, with their squaw part- 
ners and a party of Indians. Most of the men being absent 
hunting buffalo, a band of five hundred Sioux, Cheyennes, and 
Arapahoes suddenly charged upon their camp, killed a white man, 
an Indian, and two women, drove off a hundred and sixty head of 
cattle, and, chasing the hunters, killed several of them in their 


940 VALLEY OF THE MUDDY. 


flight, the residue escaping only by abandoning their horses and 
hiding in the bushes. Intelligence of this onslaught reached 
Major Bridger, then occupied in erecting a trading-post on Green 
River; he sent Frappe advice to abandon his post at once, for fear 
of worse consequences. ‘The advice, however, was neglected, when, 
about ten days after, as his party was on their way to join his 
partner, they were again suddenly attacked by another large party 
of the savage allies. He had but forty men; but they instantly 
‘«‘forted”’ in the corral attached to the trading-post, and stood on 
their defence. The assault lasted from noon until sundown, the 
Indians charging the pickets several times with great bravery; but 
they were finally repulsed with the loss of forty men. Frappe 
himself was killed, with seven or eight of his people. 

I give this as a sample of the perilous adventures in which these 
rude and daring men, almost as wild as their savage foes, were 
engaged, as things of course, and which they related around their 
camp-fires with a relish quite professional. 

The only vegetation at this camp was a few scattering clumps 
of small willows and some black currant-bushes: the supply of 
» grass was scanty. Muddy Creek runs between perpendicular cut 
clay-banks, forty feet apart; the water at the present stage being 
only four feet wide and four inches deep. Day’s travel, very di- 
rect as to course, twenty and a-half miles. Lat. 41° 27’ 067.1; 
long. 107° 52’ 41”. 

Thursday, September 19.—Slight frost in the night. Ther. at 
sunrise, 30°. ‘The night passed without alarm; and, crossing the 
creek, we continued up its left bank, and soon reached a point 
where it made a long cafion through the hills. The ground was 
rough and filled with gullies made by the rush of the spring 
freshets. The soil was loose and sandy, and the waters had cut 
numerous deep and narrow channels across the valley, whose per- 
pendicular banks obliged us to pass along the base of the bluffs, in 
order to head, and thus avoid them. The creek had to be crossed 
some six or eight times, and, upon the whole, this has been the 
roughest and most difficult part of the route. Before noon we 
passed a spot where a party of fourteen fur-traders, under Mr. 
Vasquez, had «forted” and fought forty Ogallalah Sioux for four 
hours, successfully defending themselves and repulsing the In- 
dians. One of our men, a half-breed hunter, had himself been in 
the fight, and pointed out to me the localities with the most minute 
particularity of bloody detail. 


BEAVER DAMS—ATLANTIC WATERS. 941 


A meridian observation gave for latitude 41° 28’ 28’. A few 
aspens occur in the bottom, with abundance of artemisia, some of 
which were six and eight feet in height. An occasional outcrop 
_ of coal was also observed; the argillaceous shale, some three hun- 
dred feet in height, through which the creek cuts a channel, dip- 
ping north-westerly at an angle of 20°. 

Beyond this point the creek makes another cafion, which, re- 
quiring some reconnoissance, we turned down into a pretty little 
bottom, fringed with willows, currant-bushes, and birch, and en- 
camped, having made only fourteen miles. We found the creek 
filled, at short intervals, with beaver-dams, some of which had been 
but recently constructed, the chips made by cutting down the 
bushes, and the paths made through the grass and brush by drag- 
ging them into the water, being still plainly discernible. The 
stream furnishes some small fish, among which were speckled 
trout. 

Friday, September 20.—Morning clear and bright. Ther. at 
sunrise, 31°. Clouds however soon began to gather, and finally 
covered the whole sky. It had been determined to go on until ten 
or eleven o’clock, and then to make a halt of part of two days to 
rate our chronometers, and to obtain, if practicable, a series of 
satisfactory observations for longitude. But the sun being en- 
tirely obscured, and it coming on to rain, the march was con- 
tinued during the day. It unexpectedly cleared in time to obtain 
a meridian observation for the latitude. 

Leaving the camp-ground early, we continued up the right bank 
of the Muddy, over rather rough ground, covered with sage, for a 
couple of miles, to within one mile of the point where the main fork 
comes in from the Park Mountains on the south-east, where it 
heads. Here we turned to the left up a beautiful pass, about a 
mile and a-half in length, with a uniform gentle ascent to its sum- 
mit. From the top of this pass we continued for four miles over 
a gently undulating country, sloping to the right into the drain- 
age of the Muddy. Here we reached the dividing height between 
the waters of the Pacific and those of the Atlantic. 

One universal shout arose at the announcement of this fact; and 
visions of home and all its joys floated before the imagination in 
vivid brightness. That to which we had so long been looking 
forward, as a thing that might one day be, now seemed almost 
within our grasp; for we knew that the waters which we had at 


length reached, flowed, in one unbroken stream, almost to the very 
16 


249 VALLEY OF THE NORTH FORK OF THE PLATTE. 


feet of those who were dearest to our hearts. Often have I ex- 
perienced the same feeling, when, in the dark and solemn forest, 
I have sat by my solitary camp-fire on the bank of some murmur- 
ing stream. The waters seemed a connecting link with human 
beings, however distant; and as I thought whither they would flow, 
I felt myself not so utterly alone. None but those who have ex- 
perienced it, know how much companionship there is in the gentle 
murmur of a flowing stream. Such were now our sensations as 
with light hearts and buoyants spirits we galloped down the grassy 
slope. 

Before us lay an undulating country, descending gently toward 
the east; beyond, in the distance, frowned the Medicine-bow 
Butte, at the foot of which flowed the waters of the Platte, while 
to the southward of this famous headland stretched far away the 
Park Mountains, whence issue so many tributaries to the Pacific. 
From this important summit we commenced a scarcely perceptible 
descent into a wide grassy hollow, forming the valley of a now dry 
rivulet, which, in the spring, discharges its waters into Sage Creek, 
an affluent of the North Fork of the Platte. Two miles east of 
the «‘divide,’’ we halted to take a noon observation for latitude, 
which placed us in 41° 83/ 22/.5, the computed longitude being 
107° 30’ 48’".. Grass and water were scarce, and the growth of 
artemisia very thick, making it somewhat difficult to pass our 
little wagon over it. Near the mouth of this little stream we 
crossed over to another, heading near to it, and running parallel 
with it, upon the banks of which we encamped, after a deeply in- 
teresting march of nearly twenty-two miles. 

From what has been seen since crossing the summit, I am satis- 
fied that it would have been better had we kept more to the south- | 
ward before crossing it. If, leaving the ridge forming the south- 
ern boundary of the valley, we had followed either the stream 
upon which we are now encamped, or even the valley of another, 
some miles still farther south, we would not only have somewhat 
shortened our route, but obtained a greater abundance of grass, 
wood, and water, from a high ridge which bounds all these little 
streams on the south. Observations gave for latitude of the 
camp 41° 35’ 41’; longitude 107° 21’ 52’. Distance from Fort 
Bridger, two homaeed and seven miles. 

With the exception of the rough ground near the head of 
the Muddy, which offers no obstruction of consequence, a per- 
fectly feasible, and indeed a most excellent route, whether for a 


SAGE CREEK—NORTH FORK OF THE PLATTE. 943 


wagon or railroad, has thus been traced, presenting fewer obsta- 
cles to the construction of either than almost any tract of the 
same length in the country. The grades will be easy, the bridg- 
ing comparatively light; and, with the exception of the crossing 
of the valley of the Muddy, where a long and heavy embankment 
may be required, the cuttings and fillings will be entirely within 
moderate limits. In no case will an inclined plane be required ; and 
the route is more than usually free from the objection of high and 
narrow caiions, liable to be filled up or obstructed by snow during 
the winter. 

Saturday, September 21.—Morning clear and bright. Ther. 
at sunrise, 35°. Ice formed in the buckets during the night. 
Passing down the right bank of the little drain upon which we had 
encamped, we encountered the usual impediments from thick arte- 
misia, and numerous little gullies, many of which were deep and 
difficult to cross. To avoid them, we turned more to the south, and 
crossed Sage Creek, an affluent of the Platte, about four miles 
above its mouth. The water was eight feet wide, and three or 
four inches deep, with a free current, and vertical clay banks. 
This part of the route was over a sand and clay soil, denuded of 
vegetation, and strewn over with black schorl gravel, and an im- 
mense quantity of white quartz pebbles, in angular fragments, 
that did not seem to have been water-worn. 

After the crossing of Sage Creek, upon approaching the Platte, 
we encountered many ravines coming down from a ridge on our | 
right, the intervening ground being washed almost entirely bare of 
grass or vegetation of any kind. In many places the surface of 
the ground was covered with small broken fragments of crystallized 
’ sulphate of lime, of a rich brown colour, mostly as clear as mica, 
(for which, indeed, it was at first mistaken,) and many specimens 
were perfectly transparent. Large quantities of pure white quartz 
gravel, also, were brought down from the hills, and lay mingled 
with the gypsum. 

After a march of sixteen miles, we encamped on the left bank 
of the North Fork of the Platte, in a lovely bottom, amid picturesque 
groves and clumps of gigantic cotton-woods. The ground was 
covered with a luxuriant growth of nutritious grasses, among which 
buffalo-grass was quite abundant. 

In this region the bottom land is principally confined to the left 
bank, and is from a quarter to half a mile in width. On the right 
bank are escarpments of rock a hundred and fifty or two hundred 


— 244 NORTH FORK—INDIAN FORTIFICATIONS. 


feet in height—the cliffs rising abruptly from the water’s edge. 
The formation consists of horizontal strata of a soft, coarse, brown 
sandstone, overlaid by bituminous shales, above which are clay and 
earthy marl, the whole capped with a heavy stratum of sandstone. 
The river in its present stage is one hundred and sixty feet wide 
and two feet deep; the bed consisting of gravel, rolled pebbles, and 
boulders, among which a red feldspathic granite, gneiss, granite, 
quartz, and a very compact, firm-grained, ferruginous limestone are 
the most prevalent. 

The cotton-woods round our camp are the first trees, worthy of 
the name, that have greeted our eyes for more than a year. They 
seemed to us like old friends, and, as they waved in the fresh breeze 
over our heads, reminded us of those beloved woodlands from which 
we have been so long separated. Oh! with what longing desire 
had we looked forward to such a sight; while our souls, sick of 
rolling prairies, barren plains, bald and rocky ridges, muddy flats, 
and sandy wastes, sought in vain for the forest shade and those 
hills of living verdure which give the charm to every landscape. 
Day after day, week after week, had we journeyed over that deso- 
late basin, without a tree to be seen in the whole horizon. But now 
the rustling sound of embowering leaves assured us that we had 
once more reached a spot fitted by nature for the habitation of 
man. 

The place we now occupy has long been a favourite camp-ground 
for the numerous war-parties which annually meet in this region to 
hunt buffalo and one another. Remains of old Indian stockades are 
met with scattered about among the thickets; and the guide in- 
formed us, that four years since there were at one and the same 
time, upon this one bottom, fifteen or twenty of these forts, con- - 
structed by different tribes. Most of them have since been de- 
stroyed by fire. As this was the season of the year when we might 
expect to find them upon their expeditions, we were on the quz vive, . 
lest we should be surprised. Arms were inspected and put in 
order and a vigilant guard kept during the night. 

It had been intended to reach this camp by ten in the morning, 
so as to rate the chronometers and take lunar observations for the 
longitude. We arrived, however, barely in time to obtain a me- 
ridional observation, which gave our latitude 41° 82’ 49/.2. Com- 
puted longitude, 107° 6’ 11’. 

One of the pack-mules, having for his burden all the flour be- 
longing to the officers’ mess, had, by some oversight, been left 


COTTON-WOOD GROVES—INDIAN FORTS. 945 


behind at the morning’s encampment, nor was the incident dis- 
covered until near the end of the day’s march. ‘Two men were 
immediately sent back to recover him, but returned unsuccessful. 
They declared that the creature must have hidden himself pur- 
posely in the thicket near the camp—a trick to which it is said 
some of these animals are addicted. A detail of six men was 
ordered to start back for him early in the morning. 

Archambault killed a remarkably fat buffalo cow to-day, which 
gave us plenty of the finest beef. Side and hump ribs, and buffalo 
hump—that precious morsel of all mountain connoisseurs—graced 
our humble board, (or rather the top of our mess-chest,) and we. 
only needed the presence of dear but absent friends to render our. 
dainty repast all that an epicure could desire. Another cow was 
also killed, which contained a dead and thoroughly dried foetus, 
about one-third grown, which had apparently been enclosed in the 
uterus and preserved for a long time. The hunters say that such 
is not an unfrequent occurrence. 

Sunday, September 22.—We remained at this pleasant camp all 
the day, being Sunday. The detail sent after the missing pack- 
horse returned about one o’clock, having discovered him, with his 
pack undisturbed, hidden in the brush in the vicinity of the camp 
of yesterday morning. 

The day was warm and balmy, with fresh breezes ftom the north- 
west. The evening closed with heavy gusts of wind from the 
south-west, with appearance of rain. Several herds of buffalo 
were seen during the day. 

Monday, September 23.—Morning warm and cloudy. High 
wind during the night, with a slight sprinkle of rain. We followed 
up the left bank of the Platte, in an eastern direction, when we 
crossed it on a ripple, witha depth of eighteen inches. Swift 
current, clear, transparent water, rolled stones and _ pebbly 
bottom. 

A large portion of the way from camp was through a natural 
park of noble cotton-wood trees, sixty feet high, and two and three 
feet in diameter, and over a rich level bottom, covered with a luxu- 
riant growth of grass. Major Bridger informs me that, for twenty 
miles above, the river presents the same beautiful appearance, every 
little bottom formed by the windings of the stream being covered 
by a growth similar to that through which we had just passed. 

Immediately above where we crossed, were about twenty Indian 
forts, or lodges constructed of logs set up endwise, somewhat in 


946 MEDICINE-BOW BUTTE. 


the form of an ordinary skin lodge, which had been erected among 
the timber by different war-parties: they appeared to be wie 
strong, and were ball-proof. 

From the river we turned more to the northward, in order to 
pass around the northern end of Medicine-bow Butte, a small iso- 
lated mountain, detached from the main range immediately south 
of it. The ground rose gently for nine miles, when we reached 
the bank of a small stream issuing from the pass between the 
butte and the range just referred to. 

It was through this gorge that Fremont passed in 1842; but as 
he had represented it as very rough, and our guides declared it to 
.be wholly unfit for the passage of wagons, it was determined to 
look for a route farther to the northward. We halted to noon on 
this little stream, to which we gave the name of « Pass Oreek,”’ 
and to take a meridian observation, which gave the latitude 41° 
87’ 15/.8. Crossing this creek and passing over a pretty little 
prairie, covered with grass, now cropped close by herds of buffalo 
and antelope, we continued our journey near the northern base of 
the butte for about ten miles. At night, we encamped upon a 
branch of Rattlesnake Creek, at the foot of a spur, the sides of 
which were covered with a heavy growth of pine. The route tra- 
velled from the Platte has been too far to the right for a good 
wagon-road. By following up a wide, smooth valley running 
parallel with our route, and about a quarter of a. mile to the left, 
an excellent road may be obtained, with a regular ascent, until 
reaching the south branch of Rattlesnake Creek, where some un- 
even ground, but presenting no material obstruction, will have to 
be encountered. 

Numerous bands of buffalo and antelope were seen during the 
day, and in the afternoon, a monstrous bull was killed; but the 
hunters took only a piece of hide from the back, to bth: lash- 
ropes for the packs, the marrow-bones from the hindlegs, and the 
tongue, leaving at the very least six hundred weight of good beef 
on the ground for the wolves and the ravens. In several places 
along the route to-day, I noticed fragments of white crystallized 
sulphate of lime, bituminous shale, clay-slate, and marble; and 
after striking the valley of Rattlesnake Creek, indications of coal. 
Latitude of camp by observation, 41° 40’ 45’’.2. Longitude, 106° 
43/ 87’. Day’s march, twenty-three miles. | 

Tuesday, September 24.—Ther. at sunrise, 51°. Last night 
was cloudy, and we lost an observation of Altair. A good one of 


BUFFALO-BUTCHERING ‘‘ MOUNTAIN FASHION.”’ Q47 


Polaris was however obtained, about midnight, which gave the 
latitude. © 

Early in the morning, a large herd of buffalo was seen quietly 
feeding on the side of a high hill, about a mile to the southward. 
_ Archambault was soon in the saddle, and, approaching through a 
ravine, which concealed him from their sight, he reached the top 
of the hill immediately above them undiscovered. The whole herd 
was in full view of the camp, then busily engaged in packing the 
mules for the day’s march. Soon the crack of a rifle and the 
sudden fall of one of the dark objects on the hillside gave notice 
that the, work of destruction had commenced. Keeping himself 
concealed behind a large rock, the hunter very leisurely shot 
down four of these monsters, although one was much more than 
we could carry with us. When, satisfied with his morning’s success, 
he showed himself from behind his breastwork, the whole band 
scoured off as fast as they could “tumble ahead.” I rode up to 
the scene of this wanton butchery, and, for the first time, witnessed 
the operation of cutting up a buffalo. 

Contrary to the custom among us, the skinning process com- 
mences by making an incision along the top of the backbone, and 
separating the hide downward, so as to get the more quickly at 
what are considered the choice parts of the animal. ‘These are 
the “bass,” a hump projecting from the back of the neck just be- 
fore the shoulders, and which is generally removed with the skin 
attached: it is about the size of a man’s head, and, when boiled, 
resembles marrow, being exceedingly tender, rich, and nutritious. 
Next comes the “hump,” and the «hump ribs,” projections of the 
vertebre just behind the shoulders, some of which are a foot in 
length. These are generally broken off by a mallet made of the 
lower joint of one of the forelegs, cut off for the purpose. After 
these come the “fleece,” the portion of flesh covering the ribs; 
the “depuis,” a broad, fat part extending from the shoulders to 
the tail; the «belly fleece;” some of the ribs, called «side ribs,”’ 
to distinguish them from the hump ribs; the thigh or marrow- 
bones, and the tongue. Generally the animal is opened and the 
tenderloin and tallow secured. All the rest, including the hams 
_ and shoulders—indeed, by far the greater portion of the animal—is 
left on the ground. When buffalo are plenty, the hump, bass, and 
tongue—very frequently only the latter—are taken, and occa- 
sionally a marrow-bone for a tit-bit. 

This is called butchering “ mountain fashion,” and a most bar 


948 MEDICINE-BOW RIVER. 


barous fashion it is. The bulls are never killed for food except in 
case of necessity, their flesh being very inferior to that of the 
cows; but an old mountaineer cannot resist the temptation of a 
fair shot at one when it offers. 

It is vain to remonstrate against this wholesale destruction. 
The hunter, this morning, rather plumed himself on his great 
moderation in only killing four, when it was wholly within his 
power to kill as many as he pleased: at the same time he knew 
that one would have amply supplied all our wants. Indeed, of the 
four killed, but three were butchered, (that is, the choice parts 
only taken away,) and we left the ground, having two pack-mules 
and all the riding-horses loaded down with meat, the fourth 
animal being wholly untouched; thus abandoning to beasts of 
prey enough of the richest and sweetest beef.to supply a very 
respectable market for a week. All intercession in favour of the 
poor buffalo is looked upon by these old mountain-men with a 
strange mixture of wonder and contempt, which strongly reminded 
me of the expression of honest Dandie Dinmont, in Scott’s ad- 
mirable tale of Guy Mannering: «Weel, that’s queer aneugh! 
Lord save us! to care about a brock!” 

The train, in the mean time, had moved forward, ascending a 
dry branch of Rattlesnake Creek, running HE. N. E., with a 
gradual rise. Reaching its head, in a low gap, we attained the 
summit, and struck upon a hollow or depression leading down to a 
small branch, which, rising near the northern end of Medicine- 
bow Butte, winds its way through a broad and lovely valley, and 
discharges its waters into the Medicine-bow River. 

The route led us over some swelling ridges making toward this 
branch from the mountains on our right, and, crossing three other 
little streams, tributary to it, we reached in ten and a-half miles 
the banks of the Medicine-bow River. Here we encamped in a 
thicket of tall timber and underbrush, on an old Indian cam 
ground; the remains of several old forts, now decayed and in 
ruins, being still visible. 

On its north-western, northern, and north-eastern sides, the 
Medicine-bow Butte is surrounded by a well-defined ridge, from 
which it is separated by a broad intervening valley, the ridge ap- 
pearing to be concentric with this part of the butte, and three or 
four miles distant from it. Through this the Medicine-bow River 
breaks, passing for twenty miles between vertical walls of rock 
with wide alternate bottoms on either side. 


GAME—RUNNING BUFFALO. 949 


The scenery from the «divide’’ was in beautiful contrast with 
that of the country left behind us. Broad and grassy valleys 
were spread out before us, bounded by low rounded hills covered 
with verdure, over which ranged bands of buffalo, while little 
flocks of antelope bounded gracefully around us. The low bottom 
of the Medicine-bow, upon which we are encamped, is thickly 
covered with excellent grass, and the stream has an extensive 
fringe of willows and rose-bushes, with occasional groves of cotton- 
wood and aspens. The night was too cloudy to admit of an ob- 
servation; but a meridian altitude of the sun gave for latitude 
41° 41! 50/9. - 

Wednesday, September 25.—The wind blew furiously all night; 
and as we had for so long a time been unaccustomed to the sound 
of the blast rushing through a forest, our slumbers were frequently 
broken by the apprehension lest the tall trees, beneath whose 
spreading branches we had encamped, should be blown down upon 
our heads. Immediately upon leaving camp, we crossed the Me- 
dicine-bow, and ascended a high bluff, whence Laramie’s Peak was 
distinctly visible, bearing north, 45° east. The route continued 
_ over a rolling country, crossing several small streams heading in 
the Medicine-bow Mountains on our right. Game was seen during 
the day, in greater abundance than we had yet met with it; and, 
from the fact of our being off the great line of emigration, the 
buffalo were quite tame, some of the old bulls allowing us to ap- 
proach very near to them and moving off quite lazily. The ante- 
lope, too, seemed to regard us with more curiosity than fear, and 
repeatedly stopped within shot to take a good look at us, and then 
trotted off entirely at their leisure. Being already supplied with 
meat from yesterday’s slaughter, we suffered them to frisk’ around 
us unmolested; scorning to touch venison when buffalo-meat was 
so abundant. 

An incident occurred in the course of the morning which came 
near proving fatal to my friend, Lieutenant Gunnison. Seeing a 
small band of buffalo near at hand, he started on his horse to run 
one of them down, as the creature’s hide was wanted for the repair 
of our little wagon. The train had passed on for about a mile, 
when one of the men galloped up and reported that Lieutenant 
Gunnison’s horse had thrown him, and that he was fearful some- 
thing serious had happened. I instantly rode rapidly to the 
point indicated; and found my excellent officer partially delirious, 
reclining on the ground, his face and hands covered with blood, and 


we 


250 ) HAPPY ESCAPE—FRAPPE’S CREEK. 


his horse, a fine roan, lying dead by his side. The scene was soon 
explained. When starting after the buffalo, Mr. G. had handed 
his gun to one of the party, and, drawing a revolver from his 
holster, set off in pursuit. In crossing a narrow ravine his horse 
had stumbled and nearly fallen: the nervous contraction of the fin- 
gers caused by the endeavour to save himself had occasioned the 
discharge of the pistol, the ball of which, passing directly through 
the neck of the horse, had killed him instantly; and his rider was 
hurled with great violence to the ground. I was much relieved to 
find that no bones were broken, and that, with the exception of 
some severe scratches, and a violent jar of the system, nothing 
very serious had happened. It was a narrow escape, however ; 
for a broken bone, so far from surgical aid, would have proved no 
light matter. After the detention of an hour, Lieutenant G. was 
mounted upon another horse, and accompanied the train as usual, 
his ambition for running buffaloes entirely satisfied. 

A meridian altitude of the sun gave for latitude 41° 38’ 38/7.6. 
Laramie Peak bearing north 29° 80’ east, mag. The afternoon’s 
march was over a beautiful rolling country, lying at the foot of 
the Medicine-bow Mountains, whence issued several small streams, 
emerging from narrow cafions, their sides clothed with cotton- 
wood, aspen, and cedars—their windings through the plains to the 
northward being distinctly traceable by the rich belts of green 
that clothed their banks. The soil was sandy, and profusely 
covered with small fragments of white, smoky, and rose quartz, 
very pure, and in many cases nearly translucent, which had been 
washed down from the mountains. We made but one march to- 
day, and, crossing the east fork of the Medicine Bow, encamped 
three miles below, upon the banks of Frappe’s Creek, one of its 
tributaries. The east fork, where we crossed it, is about forty 
feet wide and one foot deep, flowing with a rapid current and pure 
limpid water over a pebbly bed. The bottom of this pretty little 
stream is about a mile wide, well covered with grass, and tolerably 
wooded with cotton-woods and aspens. The mountain-sides on our 
right have been well clothed with fir and pine. Frappe’s Creek is 
so called from the fact of Mr. Frappe having been some years 
since robbed, at the mouth of this stream, of a band of sixty 
horses, by a party of Aricarrees. Day’s march, seventeen miles. 
Lat. 41° 33’ 6’... Long. 106° 15’ 58”. 

Thursday, September 26.—Morning clear. Ther. at sunrise, 
48°, A high wind from the south-west. ‘To-day we entered the 


, LARAMIE PLAINS—ALARM. 251 


Laramie Plains, and travelled over a beautiful rolling country, 
covered with grass, with here and there a small lake or pond, 
formed in the low grounds by the drainage from the neighbouring 
hills. A meridian observation gave the latitude 41° 28’ 16”. 
From this point we took a course a little to the south of east, for 
a prominent landmark which rises near the heads of Lodge-pole 
Creek, an affluent of the South Fork of the Platte, and in ten miles 
reached the western fork of the Laramie River, upon the left bank 
of which we encamped. ‘The river is twenty feet wide and eight or 
ten inches deep, flowing with a rapid current over a bed of peb- 
bles. The bottom is about four miles wide, with abundance of 
fuel and grass. The trace to-day has been rather undulating ; but 
an excellent road can be located without difficulty. Buffalo have 
been very numerous and tame. Day’s march, twenty-one miles. 
Latitude, by observation, 41° 19’ 43/4. Long. 105° 57’ 12”. 

Friday, September 27.—Clear and calm. Ther. at sunrise 
41°. Slight frost on the grass in the low grounds. Crossing 
the west fork of the Laramie River, our course was nearly due 
east, over a gently rolling prairie. The trace is smooth, and had 
we crossed the Laramie Fork about a mile to the northward, it 
would have been as level as could possibly be desired, with not a 
bush or ravine to obstruct the passage. The timber which clothes 
the hills on the south ceases at their base. Artemisia has entirely 
disappeared. 

About eleven o’clock, two of the scouts who had kept on the left 
flank of our little party were descried descending from the hills at 
full gallop, waving their hats, and giving the alarm of Indians. 
We were at the time in the midst of a broad prairie, toward which 
rolling ridges sloped gently on either hand, and at a considerable 
distance before us rose a bold prairie ridge: not a bush or a tree 
was to be seen which could be converted into a covert for defence. 
The train was immediately halted, the pack-mules and loose ani- 
mals caught up and led by their halters to prevent them straying 
from the band, and the men were formed into two lines behind our 
little wagon, between which the led animals were driven, the 

whole being closed up by a guard in the rear. In a few minutes 

our simple arrangements were completed, and we moved forward 
over the plain, prepared to make as stout a resistance as circum- 
stances would permit. 

In a hollow on our right lay two lakes, or ponds, and some three 
miles ahead ran the main fork of the Laramie. Herds of buffalo 


952 ANTICIPATED ATTACK OF INDIANS. : 


were seen rapidly emerging from the little hollows on our left and 
spreading in great confusion over the plain—a sure indication that 
they had been disturbed by some cause behind the hills. At length 
scattered bands of mounted Indians were discovered moving rapidly 
at a considerable distance before us; and occasionally a look-out 
could be seen motionless upon the summit of some elevated mound, 
apparently watching our movements. At noon a short halt was 
ordered near some ponds of water, and a meridian altitude of 
the sun taken, which gave the latitude 41° 15’ 41’.4. We then 
moved forward, and, in three miles, reached the main fork of the 
Laramie, a beautiful little stream of pure, cold water, about fifty 
feet wide, and eighteen inches deep, flowing with a free current to 
the northward, between low grassy banks, over a bed of pebbles 
and gravel. It had been intended simply to make a short nooning 
here, and then to push forward toward the heads of the Lodge- | 
pole Creek with all possible speed. This intention was frustrated, 
however, by the appearance of the Indians, who were discovered 
moving toward us from various points and in considerable numbers. 
Uncertain of their intentions, or of the amount of their force, I 
deemed it prudent to prepare for their reception. 

About a quarter of a mile above us, on the right bank of the 
stream, was an isolated little grove of cotton-woods, of which I de- 
termined to take possession, and there to await the approach of our 
red brethren: this was accordingly done; the animals were taken 
into the grove and picketed. In a short time, by felling a few 
trees and piling up such as were found lying upon the ground, an 
enclosure was constructed, which a strong force of Indians would 
have found it somewhat difficult to carry in the face of thirty rifles. 
In the mean time, Indian scouts made their appearance upon the 
surrounding hills, reconnoitring us, and seemed to be as uncertain 
of our character and intentions as we were of theirs. Having 
completed our little field-work, the United States flag was display- 
ed, and we sat down to lunch, having eaten nothing since sunrise. 
Finding the Indians only hovered around at a distance, Major 
Bridger, shouldering his rifle, walked out toward them, and made 
various signs to an advance party that came out to meet him. We 
soon perceived that they had recognised him; when, finding that 
we were white men, and not a hostile band of Indians as they had 
supposed, they commenced a perfect race for our camp, and in a 
few minutes a stream of Indians galloped up, holding out their 
hands to shake with any and every body they met. They proved 


OGALLALAH SIOUX—CHEYENNE OHIEF. 2595 


to be a large band of Ogallalahs, (one of the numerous bands 
of Sioux,) who had discovered us early in the morning, and 
had been anxiously watching our movements all the day, having 
mistaken us for a war-party of Crows. As soon as they saw 
the flag displayed, they knew at once that we were whites, but 
had hesitated to approach us, through fear of the small-pox, 
which they represented as raging below and in the neighbourhood 
of Fort Laramie. They had fled hither to avoid it, and were 
much alarmed lest we should have it among us. Being assured 
to the contrary, they poured in upon us from all quarters, and our 
camp was soon crowded with them. Several of the chiefs and 
head men had certificates from the commanding officer at Fort 
Laramie, and from different emigrant companies, as to their friendly 
character, which they handed to me with an empressement which 
showed the great importance they attached to them. Some coffee, 
flour, and sugar were served out to them, together with all the 
tobacco I could spare; and after a plentiful repast, they departed 
for their village on the Laramie, about two miles below, with every 
demonstration of good-will. The head chief, who rejoiced in the 
very original title of «« Buffalo Dung,” gave me a warm invitation 
to pay him a visit in the morning; which I promised to do. 

The band consisted of several hundred: they were, for the most 
part, fine-looking men, straight, tall, and athletic, and generally 
well mounted. I afterward learned that as soon as they recog- 
nised our flag, and became satisfied as to our character, they 
had sent word to the spot selected for their temporary camping- 
ground, whereupon the squaws, reassured, immediately commenced 
putting up the lodges; and before sundown the plain was white 
with them. 

Among the Sioux was one solitary, dignified old Cheyenne chief, 
who figured in the undress frock of a major of artillery, buttoned 
closely up to his throat, and of which he seemed not a little vain. 
To my surprise, I found that he did not understand the Sioux 
tongue at all, and communicated with those of that tribe wholly by 
signs. The Sioux chief with the unpronounceable name, the trans- 
lation of which has already been given, was a noble-looking old 
man, and very much disposed to be sociable. He explained to me 
that he was greatly afflicted with sore eyes, and begged for some- 
thing to cure them. I had nothing but an old pair of goggles, 
with very dark green glasses, which I gave him, and with which he 
was very much delighted, mounting them with great complacercy, 


254 INDIAN LANGUAGE OF SIGNS. 


although it was then very nearly dark. With a spy-glass, also, 
they were very much pleased, and through it watched the erection 
of their lodges with great wonder and interest. A Colt’s revolver, 
when explained to them, excited many remarks, and evidently in- 
creased their respect for the strength of our little party. | 

There was one circumstance, however, that attracted my attention 
in this interview with these untutored sons of the forest more than 
any other; and that was the perfection and precision to which they 
appear to have reduced a system of purely arbitrary and conven- 
tional signs, by which, all over this vast region, intercourse, though 
of a limited character, may be held between tribes who are perfect 
strangers to each other’s tongue. After partaking of such food as 
could be hastily prepared for them, the principal men seated them- 
selves on the ground, in a circle around the camp-fire in front of 
the tent, and the pipe of peace was filled and duly circulated in 
regular succession. Our esteemed friend and experienced moun- . 
taineer, Major Bridger, who was personally known to many of our 
visitors, and to all of them by the repute of his numerous exploits, 
was seated among us. Although intimately acquainted with the 
languages of the Crows, Blackfeet, and most of the tribes west and 
north-west of the Rocky Mountain chain, he was unable to speak 
to either the Sioux or Cheyennes in their own tongue, or that of 
any tribe which they could understand. Notwithstanding this, he 
held the whole circle, for more than an hour, perfectly enchained 
and evidently most deeply interested in a conversation and narra- 
tive, the whole of which was carried on without the utterance of a 
single word. The simultaneous exclamations of surprise or interest, 
and the occasional bursts of hearty laughter, showed that the whole 
party perfectly understood not only the theme, but the minutiz of 
the pantomime exhibited before them. I looked on with close at- 
tention, but the signs to me were for the most part altogether unin- 
telligible. Upon after inquiry, I found that this language of signs 
is universally understood by all the tribes. 

At sundown the whole band left for their village; previous to 
which, a venerable old Indian traversed the camp, haranguing the 
young men in an elevated and monotonous tone, the purport of his 
exhortation being, as I was told, a warning to them not to touch 
or meddle with any of our property. But the old man’s advice was 
of little avail; for we ascertained, soon after their departure, that 
a couple of axes, a blanket, and an excellent rifle had mysteriously 
disappeared. ; ? 


SIOUX VILLAGE—BUFFALO-DUNG—THE IRON HEART. 255 


Saturday, September 28.—Morning clear and bright. At an 
early hour several of our yesterday’s visitors were on the ground, 
in the hope of securing a good breakfast. We mentioned to them 
our loss of the previous evening, and they promised to do what 
they could to recover our property. The train was directed to 
move forward under the charge of Major Bridger; while, accom- 
panied by a whole troop of Indians, I rode over to the village to 
pay my respects to the chief, according to promise. ‘This village 
was the largest and by far the best-looking of any I had ever seen. 
It consisted of nearly one hundred lodges, most of which were en- 
tirely new, pitched upon the level prairie which borders on the 
verdant banks of the Laramie. No regular order seemed to be 
observed in their position, but each builder appeared to have se- 
lected the site for his habitation according to his own fancy. 

We rode at once to the lodge of the chief, which was painted in 
broad horizontal stripes of alternate black and white, and, on the 
side opposite to the entrance, was ornamented with large black 
crosses on a white ground. We found the old fellow sitting on the 
floor of his lodge, and his squaw busily engaged over a few coals, 
endeavouring to fry, or rather to boil, in a pan nearly filled with 
grease, some very suspicious-looking lumps of dough, made doubt- 
less from the flour they had received from us yesterday. The chief 
courteously invited us to take something to eat, which, having the 
fear of the very questionable lumps aforesaid before our eyes, we 
modestly declined. By the aid of one of my men who had been 
among these tribes for two years, I made out to inform him that 
some of his young men had, when on a visit to our camp yesterday, 
stolen some of our property, and requested him to take measures 
for its restoration. He at once said that there were in the compa- 
ny of visitors a number of Cheyenne Indians, and that they, and 
not the Siouxs, (or « Dahcotahs,” as they all called themselves,). must 
have been the depredators. He promised, however, to send out the 
erier, and try to ascertain whether the things could not be re- 
covered. I soon recognised the cry of the old Indian of yesterday, 
who went around the village making proclamation of the loss. 
After some further conversation, another chief, named the «Iron 
Heart,” rose up and invited us to a feast at his lodge: we accord- 
ingly accompanied him, and found him occupying the largest and 
most complete structure in the village, although I was assured that 
the Sioux frequently make them much larger. It was intended to 
be used whenever required, for the accommodation of any casual 


256 SIOUX LODGE—BUFFALO-FEAST—SUPERSTITION. 


trader that might come among them for the purpose of traffic, and 
was accordingly called «The Trader’s Lodge.” It was made of 
twenty-six buffalo-hides, perfectly new, and white as snow, which, 
being sewed together without a wrinkle, were stretched over twenty- 
four new poles, and formed a conical tent of thirty feet diameter 
upon the ground, and thirty-five feet in height. 

After we had seated ourselves upon the skins provided for our 
accommodation, the pipe was duly passed around, and the feast 
was introduced. It consisted of a tin pan containing a parcel of 
dried buffalo-meat, which had been boiled in simple water, with- 
out salt, and suffered to get cold. This was brought in by an old 
squaw, and placed upon the ground before us, with a basin of water. 
Although we had not long before breakfasted heartily upon roasted 
rib and tender-loin, we were in courtesy obliged to partake of this 
rather lean fare with apparent satisfaction, notwithstanding that 
the pan containing it looked as if a thorough cleansing would most 
materially have improved its appearance. We accordingly ate a 
morsel or two, when, most fortunately, an old Indian came in, 
who, after taking a few whiffs of the pipe which was passing round, 
reached out his hand for the pan,.and very soon discussed the 
whole of its contents without the slightest scruple. 

The feast being finished, we rose to take our leave; when the 
chief courteously motioned us to remain, saying that they were 
haranguing the village in hopes of recovering our lost property. 
Although I had not the least idea that we should ever see any of 
the stolen articles again, yet we continued to sit, listenmg to the 
bellman-like proclamation going on outside, until at length I 
explained to the chief that I could wait no longer; that my young 
men had gone forward, and that it was necessary I should overtake 
them; whereupon we shook hands and parted. He was a remark- 
ably fine-looking man, of about forty-five, with a face denoting 
strong character, great firmness, and yet, as I thought, a kind 
heart. His influence with his people was said to be very great. 
One thing in his manner struck me with surprise: I observed, 
during our interview, that he always passed the pipe without smok- 
ing; and upon afterward inquiring the reason of an omission so - 
unusual, found that it was “against his medicine’ to smoke in the 
presence of others, and that whenever he indulged in this Indian 
luxury, it was when alone. Returning to the lodge of the fat old 
chief, whose merry laugh and cheerful physiognomy denoted a 
great love of fun, and not a little of good eating, we soon found that 


INDIAN LAWS OF HUNTING—BLACK HILLS. as 


all hopes of the lost rifle were vain, and immediately left the 
village. 

This band of Ogallalah Sioux was about a hundred lodges strong, 
and seemed to possess a large number of fine horses, as well as a 
good many excellent mules; the latter procured, no doubt, by 
trading with the emigrants along the road to California and 
Oregon. 

From the village we pursued a south-east course to overtake the 
train, which, after passing in an easterly direction over a level 
grassy prairie, with plenty of fine water, had ascended the western 
slope of the Black Hills. Having ridden a mile or two, we enjoyed 
an opportunity of witnessing what is technically called a «sur- 
round” of buffalo, by a band of about fifty Indians on horseback. 
The poor animals were in great confusion and terror, the Indians 
being in full pursuit. We did not halt to see the end of the hunt. 
During the chase a small band was driven near us, and a fine fat 
cow was secured by a shot from one of my revolvers. A mile or 
two farther on, we found a couple of our hunters very amicably 
engaged in dividing the carcass of another buffalo with half a 
dozen Indians, who laid claim to a share of the prey, on the 
ground, that although the buffalo was actually killed by the white 
men, one of their own number had first wounded it; in proof of 
which they pointed to an arrow deeply buried in its side. The 
claim was cheerfully admitted, and the game in consequence equally 
shared. The Indians told us that to the eastward of this point we 
would see no more buffalo; in this, however, they proved to be 
mistaken. At the western base of the ridge we passed through 
another village of fifty lodges of the same tribe, who were moving 
to the southward. 

For the last seven or eight miles, the prairie has been strewn 
with the carcasses of buffalo, from which the choice pieces only 
had been selected by these untutored epicures, leaving the remainder, 
from which they had not taken the trouble even to remove the skin. 
Carcasses thus left, on the open prairie are not unfrequently com- 
pletely cured, or rather « mummified,’ in the sun, so that they 
seldom exhibit any sign of decay. 

Ascending the western slope of the Black Hills by a very gentle 
rise, we followed the trail of our party, passing between low cliffs 
and detached masses of red and gray sandstone, worn into isolated 
pillars, hillocks, and other forms by the action of the elements. 


The beds appeared to be thick and extensive, but the strata were 
| 17 


258 HEADS OF LODGE POLE AND CROW CREEKS. 


thin, varying from half an inch to six inches, between which an 
occasional layer of brown and reddish argillaceous limestone was 
found interposed. 

Passing over an undulating and gradually rising country, for 
seven or eight miles, we at length overtook the trai, which had 
halted to noon on a small tributary of the Laramie River. Aspen, 
fir, pme, and cedar here occurred in scattering clumps, and the 
grass has been abundant. From this point we continued our 
course more to the north-east for four or five miles, over ground 
considerably cut up by ravines, when we reached the summit of 
the ridge, which gives rise to the head of Lodge-pole Creek, an 
affluent of the South Fork of the Platte, into which it discharges 
its waters nearly south of Ash Hollow, and about seventy miles- 
above the junction of the two great branches which form that well- 
known stream. lLodge-pole Creek here takes its rise in a high 
ridge, and falling with a rapid and sudden descent, forms a deep 
and precipitous cafion, at the bottom of which it continues to wind 
its way until it reaches the plain at the foot of the eastern slope 
of the Black Hills. It is represented as having a width between 
the cliffs which enclose its valley, sufficient for a road, by crossing 
the stream from side to side; but I was deterred from attempting 
the passage, not only by the rugged descent from the ridge, but 
by the quantity of timber growing in the cafion, through which it 
would have been necessary to cut our way the whole distance. In 
addition to this, the ridge appeared to be much lower to the south- 
ward, in the direction of the heads of Box-elder River and Fon- 
taine qui bouit, while, toward the northward, it appeared to be- 
come higher and still more rugged. This induced me to believe 
that we had crossed the ridge too far to the northward, and that a 
more feasible route could be traced south of our line of travel, by 
which much of the elevation we had attained (which si oe to 
about a thousand feet) might be avoided. 

We accordingly followed down the ridge in a 8.58. EH. direction 
for six miles, when we struck upon a little stream, which we sup- 
posed to be a branch of Lodge-pole, but which, as we afterward 
ascertained from some Cheyenne Indians, was a branch of Crow 
Creek, another affluent of the South Fork, and which flows into it 
from this point in a north-easterly direction. Here we encamped 
‘for the night, with good grass and water, after a very interesting, 
though somewhat fatiguing journey of twenty-two miles. Immense 
droves of buffalo were seen in every direction during the day. An 


ree Tl on ee eS ee ee ee 


i 


EASTERN SLOPE OF THE BLACK HILLS. 259 


observation of Altair gave for the latitude of the camp, 41° 8’ 2’; 
long. 105° 24’ 11.” 

Sunday, September 29.—Ther. at sunrise, 28°. After passing 
through about two miles of pine and aspen woods, the country be- 
came a rolling prairie, which obliged us to wind about considerably 
among the hills to avoid the undulations of the ground. Our gene- 
ral course was east for eleven miles, when, descending the eastern 
slope of the ridge, we,struck upon the heads of a little stream 
issuing from a rugged cafion of red feldspathic granite, at the base 
of the hills, and flowing into the plain below. We learned from a 
band of Cheyennes, who paid us a visit about sundown, that this 
was another branch of Crow Creek. Here the main ridge of the 
Black Hills suddenly falls off into a range of lower elevation, 
which again slopes to a plateau of clayey and earthy marl. The 
timber which had clothed the ridge ceases upon reaching the plain, 
but the stream is fringed with willows. Where the creek issued 
from the cafion, wild cherries, and yellow, red, and black currants 
occurred in great profusion and fully ripe. The yellow variety 
was particularly good, resembling in flavour a mellow sour apple. 

From our noon halt of yesterday, the formation has consisted 
chiefly of massive red feldspathic granite, with an occasional heavy 
out-crop of ferruginous quartz. Following down this branch for 
four miles, we encamped on its left bank, with good grass and water. 
A village of Cheyennes was encamped a short distance to the north 
of us, who, as soon as they descried our party, immediately paid 
us a visit. They hung around the cook-fires till the guard was set 
for the night, when I notified the chief of the fact, and desired 
him to send his people away, at the same time informing him that 
should any attempt be made during the night to disturb our ani- 
mals, the guard had positive directions to fire upon the marauders. 
He made them an harangue to this effect, and they immediately 
left us, some for a ride of ten miles back to the vicinity of our 
morning encampment, whence they had followed and accompanied 
us during the day, partly from curiosity, and partly from the hopes 
of a plentiful meal. Day’s march fourteen and a-half miles. Liati- 
tude by observation 41° 9/ 3/.5; longitude 105° 8’ 24”, 

Monday, September 30.—The camp was up long before daylight, 
and we were on the road by sunrise. Finding that the branch of 
Crow Creek, upon which we had encamped, passed too much to the 
southward for our purposes, we turned our faces to the north, and 
followed along the base of the Black Hills, about four miles distant, 


a 
260 CHEYENNE PASS. 


crossing the hills and hollows formed by the drains coming from 
them, the undulations, however, being quite gentle. On our right, 
about two miles distant, stretched a high table ridge, or plateau, 
rising one hundred and fifty or two hundred feet, its western escarp- 
ments abrupt, nearly vertical, and capped in this vicinity by argil- 
laceous limestone and sandstones, with occasional strata of pud- 
ding-stone. Between this plateau on our right and the Black 
Hills on our left, there is a marked depression or valley, averaging 
about four miles in width, and which appears to have been cut out 
by the violent action of an immense body of water flowing in a 
northern direction. The valley extends along the base of the 
Black Hills, from where we first descended their eastern slope, to 
the Chugwater ; the range of marly hills reaching, as our guides 
told us, to the Platte, in the vicinity of Scott’s Bluffs, and thence to 
«Chimney Rock” and «The Court-house.’” The formation ap- 
peared to be the same as that observed at those localities. The 
depression thus formed is called the «« Cheyenne Pass,” from the 
constant use made of it by that tribe in their migrations to and from 
the Platte. From the red caijion of Crow Creek to some distance 
down the Chugwater, a range of lower hills, apparently of lime and 
sandstone of different colours and qualities, occurs, flanking and 
following the general direction of the main back-bone of the Black 
Hills. Through these, the numerous streams which take their 
rise in the ridge beyond have forced a passage in deep, narrow, 
and rugged cafions, and, after crossing the Cheyenne Pass, have 
broken through the marly plateau on our right, in their passage 
through the plains to the eastward into the North and South Forks 
of the Platte. | 

Following the Cheyenne Pass nine miles from our morning’s 
camp, after crossing the north or main fork of Crow Creek, some 
two miles below its caiion, we struck upon the southern branch of 
Lodge-pole Creek, and, five miles beyond, halted to noon upon 
Bear Creek, one of its tributaries, where a meridian observation 
gave for the latitude 41° 21’ 45’’.7. 

We had now reached the heads of the stream, which I had pre- 
viously determined to follow to its confluence with the South Fork 
of the Platte. As we could expect to receive no addition to our 
supplies before reaching Fort Kearny, I despatched an express to 
Fort Laramie for such articles of food as were required, and occu- 
pied the interval until their return in making an examination of 
the eastern base of the Black Hills to the northward. 


RESULT OF THE RECONNOISSANCE OF A NEW ROUTE. 261 


Having now brought our reconnoissance for a new route from 
the waters of the Pacific to a point where its results can be at 
least approximately ascertained, it is very gratifying to be able to 
state that these results are, in a high degree, satisfactory ; more so, 
indeed, than I had anticipated. 

It has been ascertained that a practicable route exists through 
the chain of the Rocky Mountains, at a point sixty miles south of 
that now generally pursued, and in a course as much more direct 
as the chord of an arc is than the arc itself. A glance at the map, 
and a little attention to the table of latitudes, will show that from 
Great Salt Lake City to the head of Lodge-pole Creek, a distance 
of four hundred and eighty-four miles, the difference of latitude is 
but 35’ 42’; and that while the greatest northing made by the 
proposed line is but little more than 20’ north of Lodge-pole 
Creek, the greatest deviation to the south is but little more than 
three miles: so that the entire route through that long distance 
varies but a trifle froma straight line. When extended to the 
junction of Lodge-pole with the South Fork of the Platte, it will 
appear to be the chord of an arc formed by the present course of 
emigration. The distancefrom Fort Bridger to Fort Laramie, by the 
present route, is four hundred and eight miles; while, by the 
new route from Fort Bridger to the eastern base of the Black 
Hills, (a point equidistant with Laramie from the forks of the 
Platte,) it is but three hundred and forty-seven miles: so thata saving 
is effected, in the total distance, of just s¢xty-one miles. It must 
be kept in mind, too, that the distance thus ascertained was mea- 
sured by an odometer, following all the undulations of the natural ~ 
surface, in the course of a very rapid reconnoissance, without any 
minute knowledge of the localities, or any endeavour whatever to 
make even an approximate location for a road. When these locali- 
ties come to be minutely examined, and the comparative advan- 
tages of different courses ascertained and duly weighed, there can 
be no doubt that even this large saving in distance may be still 
further increased, by shortening the route wherever it shall be 
found practicable. The examination of the country proved it be 
more favourable than we had at first supposed. or even after so 
successfully crossing the summit dividing the Pacific from the At- 
lantic waters, serious fears were still entertained lest some formi- 
dable, if not insurmountable obstruction, should be encountered in 
the character of the ridge of the Black Hills, intervening as it 
does between the Laramie Plains on the west and the great slope 


262 ROUTE EASTWARD FROM HEAD OF CROW CREEK. 


to the Atlantic which commences at their eastern base. All appre- 
hensions on this head were, however, set entirely at rest by the 
reconnoissance, which fully demonstrated the existence of a route 
through these hills, not only practicable, but free from any ob- 
structions involving in their removal great or unusual expenditure. 

It was a subject of deep regret that our only remaining baro- 
meter (a cistern barometer) had been broken by the warping and 
cracking of its wooden frame in the dry and rarefied atmosphere 
of these elevated regions; as it would have been in the highest 
degree satisfactory to have made a precise comparison of the rela- 
tive elevations of the ground on the line of this reconnoissance 
with those already ascertained by previous explorations upon the 
old route by the South Pass, the Sweetwater, and North Fork of 
the Platte. Although this was unfortunately impracticable, yet a 
careful observation of the ground on both routes has enabled me 
to form a general comparison between them; and has led to the 
unhesitating conclusion that, in point of dimaniclin distance, easy 
grades, A from serious obstacles, and convenience and abun- 
dant supply of materials for construction, the line of this recon- 
noissance presents a trace for a road ie is not only perfectly 
feasible, but decidedly preferable to the other. 

From the head of Crow Creek, the way to the eastward lies 
open in various directions. By eee over to the Lodge-pole, 
and pursuing the valley of that stream to ity junction with the South 
Fork of the Platte, an almost straight line will be secured from 
Fort Bridger to the Forks of the latter. Thence the Missouri can 
be reached either by the north bank of the river, on the route at 
present travelled by the Mormons, or by following its south bank, 
and crossing over to the Blue, below Fort Kearny: the valley of 
the latter stream can thence be pursued to its junction with the 
KKansas, and thence along that river to its confluence with the Mis- 
sourl. Should a route still farther south be deemed desirable, the 
features of the country show, that- by adopting the valley of a 
stream flowing between the Lodge-pole and the Crow, it would not 
be difficult to strike the South Fork of the Platte in the direction of 
the heads of the Republican Fork of the Kansas, and by means of. 
this latter stream either to reach the Missouri at the mouth of 
the former, or, by a slight divergence, at some lower point, as 
might be Pinaehe most expedient. 

It had been my intention to continue the reconnoissance from the 
head of the Lodge-pole to its junction with the Platte, and thence 


ROUTE FROM FORKS OF THE PLATTE TO THE GREAT BASIN. 263 


either on the dividing ground between that river and the Repub- 
lican, or (had that proved unfavourable or impracticable) by the 
valley of the latter stream; either of which lines would have led 
us over ground as yet unexplored. Circumstances unfortunately 
prevented this design from being carried into execution. 

As any examination of the country over which we passed, other 
than a very general one, was foreign to the objects contemplated 
by my instructions, I have contented myself with simply adverting 
to such prominent geographical features of the country as came 
under my observation; being satisfied that so weighty an enter- 
prise as the selection and construction of a great line of communi- 
cation with the Pacific waters will not be entered upon without 
previously well-considered and connected explorations. These, of 
course, will be of a character much more minute and elaborate than 
could be possibly made in the progress of a rapid reconnoissance, 
the results of which must necessarily be of a character too general 
to form the basis of other than a conjectural estimate as to com- 
parative merits of different and conflicting routes. 

The valley of the Salt Lake being the only point between the 
Missouri and the Pacific whence supplies of provisions can be pro- 
cured, it must become an object of no little importance to embrace 
it in any scheme for a road across the continent. 

I have already, in a previous portion of this report, suggested 
the route which I considered the best between Fort Bridger and 
that point, and which, taken in connection with the line to the head 
of Crow Creek, will give an excellent trace for a wagon-road all 
the way to the city. As to a railroad, the route is good as far as 
Camass Prairie; but the trace thence by either of the caifions at 
present travelled through the Wahsatch range to the city, will, I 
_ think, be impracticable, or at least enormously expensive. ‘ From 
the Camass Prairie, however, it is proposed to descend the Timpa- 
nogas, which reaches the upper level of the Salt Lake Valley at. 
Lake Utah, some fifty miles south of the city. This part of the 
route I have not personally examined, but from descriptions given 
of it by the guides and others, I have little doubt that it will prove 
entirely practicable. 

After reaching the Salt Lake Valley, the road, as I have here- 
tofore suggested, might fork, one branch leading to Oregon, and 
the other to the Pacific within the limits of California. The former 
would descend the valley of the Jordan, to Salt Lake City, whence 
it would traverse a perfectly level country, along the eastern shore 


264. ROUTES FROM SALT LAKE CITY TO THE PACIFIC. 


of the Salt Lake, to the ford of Bear River; crossing which, it 
would proceed in a north-west course, following the present emi- 
gration road until it would intersect that from the Soda Springs to 
the Humboldt, or Mary’s River. The route to California would pur- 
sue a south course at the western base of the Wahsatch range of 
mountains, on the line now occupied by the Mormon settlements, and 
would either strike the Pacific at San Diego, or, by doubling the 
southern extremity of the Sierra Nevada, should that be found 
practicable, reach San Francisco by the Tulare Lakes and the 
valley of the San Joachin. As to the character of this latter route 
I have no precise information; but it has been frequently traversed — 
by various companies of Mormon explorers, who declare it to be 
perfectly practicable; and the Mormons themselves are seriously 
contemplating the construction of a railroad over it, by which to 
secure an outlet to the ocean for the products of their territory. 

Most of the projects for a railroad across the continent, as far 
north as 40° and 41°, look to the valley of the Humboldt as a point 
whence, by the branching of the road, the Pacific coast both of 
Oregon and California may be reached: the former by the valley of 
the Wallamutte, and the latter by the Salmon-trout Pass, or some 
other, through the Sierra Nevada. The mode that has been pro- 
posed for reaching this valley, is from the South Pass, by Sublette’s 
Cut-off, to the Soda Springs, and thence in a south-western di- 
rection to the valley of the Humboldt. This part of the route, 
(from the Soda Springs to the Humboldt,) I apprehend, from the 
formation of the country over which it must necessarily pass, will 
be extremely difficult and expensive. 

The northern rim of the Great Basin, or the elevated ground 
which divides it from the valley of the Columbia, does not consist, 
as has been supposed, of one continuous mountain range which may 
be flanked, but of a number of long, abrupt, detached, parallel 
ridges, extending in a north and south direction, and separated by 
intervening valleys, which constitute, as it were, so many summit 
levels, whence the waters flow north on the one side into the 
Columbia, and south on the other into the Great Basin. Any line, 
therefore, from the Soda Springs to the valley of the Humboldt, 
will necessarily be obliged to encounter these ridges at nearly a 
right angle, and will subject any trace for a road across them to 
variations of level, which cannot but prove obstacles of a most 
serious character. The route by the Salt Lake City, and thence 
around the northern shore of the lake, would intersect that from 


ROUTE AROUND THE SOUTH END OF SALT LAKE. 265 


the Soda Springs before reaching Goose Creek, and would, from 
all the information I have’ been able to collect, pass over much 
more eligible ground. From the city to the crossing of Bear 
River and the Malade, (a distance of eighty miles,) I know the 
ground, from personal cbaeriddion, to be unexceptionable. Thence, 
since the trace pursues a course as far south of the breaks of the 
northern rim of the basin as it is possible, on account of the lake, 
it is fair to presume that the inequalities of the ground will be 
much less than by the proposed route from the Soda Springs. 

Any line from the Wahsatch Mountains to the valley of the 
Humboldt, north of the Salt Lake, cannot but prove exceedingly ex- 
pensive, for the reasons just given. 

But by passing south of it, a line can, I think, be found which 
would be comparatively free from this objection. After reaching 
the Utah Valley by the Timpanogas cafion, the road might either 
be carried to Salt Lake City on the eastern side of the Jordan 
Valley, and thence to the south shore of the lake at Black Rock; 
or it might cross the Jordan at the traverse range near its 
cajion, follow down the western side of the same valley, and 
doubling the south extremity of the Oquirrh Mountain, reach 
the south shore of the lake at the same point, viz. Black Rock. 
rom Black Rock the route would follow near to the shore of 
the lake as far as Strong’s Knob, unless further examination 
should discover a practicable passage through the range of which 
it is the northern extremity, and which forms the western boundary 
of Spring Valley. The route thus far from Salt Lake City would be 
ovér an absolutely horizontal plain. From Strong’s Knob, the same 
level desert plain extends westward for seventy or eighty miles, to 
the Pilot Peak range of hills, which, following the general law of the 
great mountain ranges in this region, extends from north to south. 
Having myself traversed this desert from the northern end of the 
Lake to Pilot Peak, and thence to Black Rock on its extreme 
southern shore, I can speak with confidence as to its character. 
It is one uniform, level plain, without verdure, and presents ground 
for a road that is absolutely faultless. . 

Westward of the range referred to I have not penetrated; but, 
reasoning from the structure of similar ridges in this part of the 
basin,—which are generally short, abrupt, and disconnected pro- 
trusions above the general level of the country, having broad level 
plains between them,—little doubt is entertained that a passage 
can, without much difficulty, be traced through to the heads of the 


\ 


266 VALLEY OF THE CHUGWATER—SEVERE INJURY. 


Humboldt. The distance will not exceed one hundred miles; and 
the object to be attained renders it certainly well worthy of a care- 
ful examination. Should the result prove favourable, we have 
then a perfectly practicable trace from the forks of the Platte to 
the valley of the Humboldt, preferable in many respects to that 
presented by the Sweetwater, the South Pass, and the Soda 
Springs. 

With these general observations upon a route hitherto untra- 
versed, I leave the subject, and return once more to our cuca 
ment at the head of the Lodge-pole. 

The interval until the 6th of October was occupied in making 
an examination of the country to the northward of the heads of 
Lodge-pole; in the course of which we crossed the several sources 
of Horse Creek, and entered the picturesque valley of the Chug- 
water. The character of the country did not vary materially from 
that farther south. In several localities the ground was strewn 
with fragments of white quartz, and jasper of a blood-red colour. 
In the bed of the Chugwater, and on the sides of the adjacent hills 
were found immense numbers of rounded blagk nodules of magnetic 
iron-ore, which seemed of unusual richness. The Chugwater 
winds from side to side of a level, well-sheltered valley, clothed 
with abundance of grass, and is handsomely timbered with box- 
elder and willow thickets, affording covert for great numbers of 
deer, which were more plentiful here than anywhere upon the 
route. The valley is a favourite wintering spot for the Cheyenne 
Indians. | 

While encamped on the Chugwater, I sustained a severe injury 
by a fall, which not only incapacitated me from mounting my 
horse, but confined me altogether to my bed until our arrival at 
Fort Leavenworth. This unfortunate accident obliged me, although 
with the greatest reluctance, to forego the projected reconnoissance 
of the valley of the Lodge-pole and of the Republican Fork, to 
which I had looked forward with the most sanguine anticipations. 
It was a source of much satisfaction, under this severe disappoint- 
ment, that the great object with which we left Fort Bridger had 
been successfully attained. 

An express was sent to Fort Laramie for surgical aid, and for 
an ambulance, which arrived on the 9th, and on the 12th we 
reached the fort. Here every kindness was extended to us by the 
officers of the post ; and on the 16th we left our hospitable friends, 
the train being in charge of Lieutenant Gunnison. Taking the 


ARRIVAL IN WASHINGTON. | 267 


‘usual emigration road, we arrived at Fort Leavenworth on the 
6th of November. On the 6th of December, I arrived in Washing- 
ton, and had the honour to report to yourself in person. 
I am, Sir, 
Very respectfully, 
Your obedient servant, 
HowarD STANSBURY, 
Capt. Corps Topographical Engineers, 

U. S. Arny. 

Cot. JoHN JAMES ABERT, 
Chief of the Corps of Topographical Engineers, 
Washington. 


‘% 
7 


APPENDIX A. 


TABLE OF DISTANCES MEASURED ALONG THE ROUTE 
TRAVELLED BY THE EXPEDITION IN 1849. 


OUTWARD JOURNEY. 


From Fort Leavenworth, on the Missourt River, to the City of the Great Salt 
Lake: forming a Travelier’s Guide to the several Watering and 
Encamping Places on the Route across the Plains, by which 
each Day’s March may be regulated. 


HOMEWARD JOURNEY. 


A similar Table from Great Salt Lake City, showing the distances along the 
New Route explored in 1850, from Fort Bridger, across the Laramie 
Plains, to the heads of Lodge-Pole Creek, and thence, by Fort 
Laramie, to the Missourt River, at Fort Leavenworth. 


ALSO, 


The Measured Distances on a Route from Great Salt Lake City to Fort Hall 
in Oregon. 


APPRENDI X,A; 


TABLE OF DISTANCES MEASURED ALONG THE ROUTE 


TRAVELLED BY THE EXPEDITION IN 1849. 


OutTwaRD JourNEY from Fort Leavenworth, on the Missourt 
River, to the City of the Great Salt Lake. 


Inter- 
mediate 
Distance. 


DATE. 
PROMINENT PoINTS AND REMARKS. 


Fort LEAVENwoRTH—Latitude, N. 39° 
21’ 14”, according to Lt. Col. Emory; 
Longitude, W. 94° 44’, according to 
Nicollet. 


Spring Camp—Good cool water, and| 6.480 
fine grass for the cattle. 
June 1| Noon Hatt—Near this point the road| 7.443 


has since been abandoned for the| \, 
ridge between the Missouri and Kan- 
sas Rivers; the directions are there- 
fore made succinct until reaching 
the Big Blue. 
Brancu—Fuel and water plenty........ 
SMALL CreEk—Road difficult on ac- 
count of deep gullies. 
InpIaN CRrEEK—Cross_ three 
creeks in the first 12 miles. 
INDEPENDENCE Roap—Crossing of the 
St. Joseph’s and Independence Road. 
Kansas Brancu—Travelling over the 
ridge; rising ground 6 miles, then 
descending, to branch creek. 
COTS so adnsobeehagacs Sadbooneaacosss 


small 


See eeeesse Beeeee DEF ESS CEES EHLES OEEEEH SEES OO 


CREEK Camp—Here we defiled to the 
left, leaving the road almost half a 
mile, for fuel and water. 

Tue Groves—Fuel and water; small 
creek, running north. 

NEEMAHA—Tributary of the Missouri.. 

NEEMAHA Brancuo—Road has wound 
along the ridge between two branches 
of this creek. Noon halt. 


270 


Day’s 
March. 


everes 


eeecee 


coves 


eveese 


Total from 


Fort 


Leavenworth. 


6.480 
18.923 


18.665 
25.048 


43.485 
46.485 
58.985 
62.8438 


68.178 
83.078 


93.868 


100.852 
109.722 


APPENDIX A.—TABLE OF DISTANCES. 274 


OUTWARD JOURNEY—Continued. 


Inter- | : Total from 
eee PROMINENT POINTS AND REMARKS. mediate ae Fort 
1849. Distance. * | Leavenworth. 


June 8| Camp CreEK—Fine timber and grass.| 7.692 | 16.562 117.414 
Creek runs south. 
‘¢ 9} Vermition—This creek heads about} 12.680] ...... 180.094 
1 mile to the north-east; the ford is 
miry. It enters the Big Blue 20 miles 
above its mouth. 
Bia Buve River—Bank steep, bottom! 13.863 | 26.548 143.957 
hard, and water 24 feet deep. River 
120 feet wide, with a brisk current. 
‘¢ 11) Kercnam’s Creexk—This is sometimes| 12.031 | ...... 155.988 
called Ten-mile Branch. Enters Big 
Blue below. Good grass, and some 
wood. Road excellent, on rolling 
prairie after ascending from the 
Blue River. The Independence Road 
joins it 8 miles from the Blue. 
TuRKEY CrREEK—Plenty of wood, and| 5.205 | 17.236 161.193 
good grazing. 


‘¢ 12| West Turkey CrEEK—A branch of Big | 10.588 | ...... 171.781 
Blue River. 
Wverrn’s Creexk—Tributary of Big| 6.3855 | ...... 178.136 
Blue. Good grazing, and fuel in 
abundance. 


EMIGRANTS’, or WALNUT CREEK—En- 9.915 | 26.858 188.051 
campment to south of the road, in 
order to obtain water, wood, &c. 
This is a tributary of the Little Blue 
River. 

«¢ 13] Sanpy CremekK—Camp. InaboutIlmile; 4.901 4.901 192.952 
from the last creek is a small branch; 
then rise considerably to a ridge; 
then descend to Sandy Creek, where 
we find feed for the cattle. 

‘¢ 14/ Bia Sanpy Cremxk—Bed is 200 feet} 12.918 | ...... 205.870 
wide, and heavy for teams. 

AFFLUENT—Stream small, and flows| 8.280 | 21.148 214.100 
into Little Blue. The road lies on 
the ridge between the Sandy and 
Little Blue, and touches ravines in 
which water may be had in several 
places. Cross some small creeks, 
branches of Little Blue. 

“15|Lirtix Broz River—lLeft bank, up| 8.003 | ...... 222.108 | 
which the road lies. The road passes 
over some spurs from the plains, into 
the valley-bed, to avoid gullies. The 
valley is two miles wide, well wooded, 
and grass excellent. 

GAD area en lank seeaccens oo sbwaniegcn alee 17.500 | 25.508 239.603 

‘¢ 16 | Lirrte Buuz Vattey—The road crosses | 24.540 | 24.540 264.1438 
some sharp gullies, and the wheels re- 

. quire to be locked frequently. Camp. 

“ 18) Litrnzs Brur—Point of leaving the} 12.165 | ...... 276.308 

stream. 


2 


June 18 


72 


DATE. 
1849. 


19 


APPENDIX A.—TABLE OF DISTANCES. 


OUTWARD JOURNEY— Continued. 


PROMINENT POINTS AND REMARKS. 


THiIrRTY-TWwo Mite Creek—A_ small, 
winding stream, with scattering trees. 

PuatTeE River—Strike the Platte in a 
broad valley. This road has since 
been abandoned for one on the left, 
more direct to Fort Kearny. 

Fort Kearny—Passing up the valley 
of the Platte, fuel difficult to be had 
in high water, and mostly-on the 
islands. The fort consists of several 
adobe buildings for offices, stables, 
and two wooden structures for quar- 
ters. Emigrants and travellers ob- 
tain various supplies at the Sutler’s. 

Camp 18—On the Platte............ssecee0s 

Camp 19—On the Platte. Fuel along 
the river. 

Camp 20—On the Platte.......... ....0se00 

Camp 21—On right bank of Platte Ri- 
ver; the road level and good, but 
fuel scarce. 

Camp 22—On South Fork of Platte, 6 
miles above the junction of the Forks. 

Camp 23—On South Fork. Water, tim- 
ber, and grass plenty. 

Camp 24—On South Fork..............000. 

Camp 25—On South Fork...............00. 

Forp oF SourH Forx—This is the up- 
per ford, and easily crossed in low 
stages of the river. Width 700 yards. 

Asn Hortow—At the outlet of small 
creek into the valley of North Fork. 
The ascent for one mile on a hard 
road, to the rolling prairie; continue 
on this to within one half mile of 
North Fork; then descend pretty 
steep hills, and wind down Ash Hol- 
low to the North Fork. Fuel in the 
hollow. 

Camp 28—Along the North Fork; road, 
most of the way heavy sand. Cedars 
in the gullies on the left. 

Camp 29—On North Fork. Sandy, and 
over spur of the table land on the 
left or south side. Court-house 12 
miles ahead. 

Camp 30—At a cool spring, 10 miles 
from Chimney Rock. Water cool and 
abundant, but fuel must be brought 
from the bluff, two miles. 

Scort’s Buurr—These bluffs are about 
5 miles south of the river. The road 
up the bluffs steep, but on good, hard, 
gravelly ground. A small spring at 


Inter- 
mediate 


Distance. 


7.430 
12.355 


14.920 


10.892 
22.832 


23.469 
24,222 


20.818 


22.895 


19.824 


25.040 


31.017 


-Day’s 


March. 


14.920 


22.895 


19.324 


25.040 


31.017 


Total from 
Fort 


Leavenworth. 


283.738 
296.098 


311.0138 


821.405 
044.237 


867.706 
891.928 


-412.746 
424.084 


437.990 
464.777 
479.879 


497.930 


520.825 


540.149 


565.189 


596.206 


APPENDIX A.—TABLE OF DISTANCES. 


OUTWARD JoURNEY— Continued. 


PROMINENT POINTS AND REMARKS. 


July 9| the top of the first hill. Robideau 


66 
66 


66 


6é 


has a trading post and blacksmith’s- 
shop here, but the post is to be re- 
moved to a creek south, and over the 
bluffs. 

10 | Crrrxk—Affluent to Horse Creek. No 
timber, good grass. 

11 | Horsr CreEK—No wood on this creek 
below the hills. The road now passes 
over rolling ground, leaving the Platte 
some distance to the right. Much 
of the way but few scattering trees 
to be seen. 

11 | Camp—On the Platte ...............seeee eee 

12 | Fort Lanamie—Camp half a mile above 
the fort. Cross the Laramie Fork 
below the fort about one mile. 

18 | Warm Sprinc—The road taken leads 
over the Black Hills some distance 
south of the river; it avoids the 
Kanyon passes, and usually has bet- 
ter feed for cattle, except when it is 
consumed by the multitude of tra- 
vellers. 

19| Brrrer Creek Brancn; Cop SPRING 
—Leaving the Warm Spring Valley, 
ascend a pretty steep hill, and pass 
over ridges to Bitter Creek. 

Bitter CareEK—Main branch, up which 
the road lies; fuel and grass abund- 
ant; stream 10 feet wide and 6 inches 
deep. 

21 | Horsz-sHor CreeK—The road follows 
up a dry branch; no water until ar- 
riving at Horse-shoe Creek. Timber, 
grass, &c. Cross at the forks, and 
rising, pass over two ridges, with 
dry beds of streams. 

Sprinc—aA spring and bed of creek, the 
water in pools. 

23 | La Bonré River—From the spring rise 
to a high, undulating ridge, the road 
very crooked. Descending for 14 
miles to dry bed of creek, follow this 
half a mile. 

BARIAIOEE A Jaga ceaeci eas sc shigube' sav olsinevakvoweess 

Camp—aAt the head of a spring branch. 
Road over hills, the wheels require 
frequent locking. Cross branches 
now dry. But few good situations 
for camping. 

24|La Pritz River—Stream 16 ft. wide. 
We have crossed two dry beds of 
creeks, and a spring containing wa- 

\ 18 


Inter- 
mediate 
Distance. 


13.062 
1.504 


19.970 
16.264 


13.428 


10.469 


5.720 


138.490 


7.059 
10.085 


6.000 
8.284 


7.250 


Day’s 
March. 


21.474 
16.264 


13.423 


16.189 


eeecee 


Total from 
Fort 
Leavenworth. 


609.268 
610.772 


630.742 
647.006 


660.429 


670.898 


676.618 


690.108 


697.167 
707.252 


713.252 
721.536 


728.786 


274 APPENDIX A.—TABLE OF DISTANCES. 


OUTWARD JOURNEY— Continued. 


: Inter- ’ 
ae PROMINENT PoINTS AND REMARKS. mediate uae 
1849, Distance. ‘ 


July 24| ter from the melting of the snow. 
This occurs occasionally in pools, at 
this season. 
CHERRY CREEK—A small stream; not| 4.250 waail 
to be depended on for water. Grass 
very scarce. 


ELDER CrEEK—Small stream, good wa-| 1.020] ...... 
ter, and scattering timber. 
FourcuHeE Boisr River—Good grassand| 3.228 | ...... 


fuel; stream 380 feet wide. 
NortH Fork or THE PirattrE—Camp| 4.275 | 20.023 
on right bank. Grass scarce. 
‘«¢ 25| Derr Creex—Clear, good water, with| 5.000] ...... 
abundance of fuel, and coal found on 
the east side of the stream, a little 
above the crossing. Also a coal-mine 
in the hills. 
Pratte Frerry—Lower ferry, esta- «AGO: (lererencaien 
blished by emigrants. River rapid, 
muddy, and deep. 
Camp—After crossing, the road is} 9.820 | 15.580 
through heavy sand most of the way, 
and the sand-hills often touch the 
banks, and must be ascended. The 
south bank is preferable, as far as 
the upper ferry. 
‘¢ 26 | Spring Camp—Brackish water in some} 12.222 | 12.222 
ponds. The road has risen upon the 
undulating table-land above the river- 
level. Road heavy. 
«¢ 27| Upper Ferry—Opposite upper ferry. 5200 hes taesaes 
Rep Sprina—Near the Red Buttes.| 18.871 | 24.071 
Road to-day hilly, and through heavy 
sand most of the way. There are 
mineral springs and alkaline lakes 
along this part of the route, danger- 
ous for cattle. Water good in some 
places along the road. 
«¢ 28 | Serine AnD RivuLtetT—Grass notplenty. | 16.821 | 16.821 
Sage the only fuel. 


‘© 30] GrEASEWoop CREEK—Six feet wide; G00 T Ag wak 
road sandy. 
Satzratus Lake—Lake west of the GI204. 4h aes 


road. Water poisonous to cattle. 
The salt used for bread-making. 

SwEETWATER RiveR—River crooked,| 6.464 | 18.778 
150 feet wide, 2 feet deep. At this 
season, the current is gentle. Grass 
plentiful, fuel scarce. 

‘© 31 | Rock InpEPENDENCE—A granite rock, of 50 || ageies 

oval or egg-shaped. 

Devit’s Gate—A kanyon of steep| 5.270] ...... 
rocks, 400 feet high; the river runs 
through the chasm. 


Total from 


Fort 
Leavenworth 


733.0386 


734.056 
787.284 
741.559 
746.559 


747.319 


757.139 


769.361 


774.561 
798.432 


810.253 
816.353 
829.567. 


829.031 


829.781 


835.051 


DATE. 
1849, 
July 31 
Aug. 1 
3 9 
66 3 
(13 4 
66 6 
6é 7 


APPENDIX A.—TABLE OF DISTANCES. 


OUTWARD JoURNEY— Continued. 


PROMINENT POINTS AND REMARKS. 


CreeK—Bad crossing, from the steep 
banks. 

CreeK —Tributary of Sweetwater, 5 
feet wide. 

Camp—On Sweetwater, in a bend, on 
the south side. 

SwEETWATER—Road leaves the river, 
and passes over hills. 

ALKALI Lake—The efflorescence on the 
shores is like snow. 

Camp—On Sweetwater. Grass good... 

PiateEAu—Cross the river four times 
this day. Camp on a plateau, after 
leaving river, some miles back. The 
water with a sulphurous smell. Sage 
for fuel. 

IcE Sprine—lIce found by digging in 
the ground. 

Steep Hitn— Descending from the 
bluffs. 

River—The road joins the river, which 
is crossed frequently. 

Camp—On south bank, having forded 
four times. Good place to encamp. 

Roap—Leaves the river, and passes 
over hills to avoid kanyons. Water 
in small streams every two or three 
miles; greasewood and sage for fuel. 
Some of the hills are steep, and re- 
quire the wheels to be locked to de- 
scend safely. 

Camp—On the Sweetwater. There 
are alkaline springs in the vicinity, 
poisonous to cattle. 

Souta Pass—Summit of the ridge 
which divides the waters of the At- 
lantic from those of the Pacific. 

Paciric SpRinacs—Fine grass, good wa- 
ter, and sage plenty, for fuel. 

Paciric CrEEK—Crossing miry.......... 

Camp—On Dry Sandy. The water is 
brackish ; fuel scarce, and little grass. 

Junotion—The Oregon road over the 
‘‘Dry Drive,” or Sublettes’ Cut-off, 
branches here. Take the left hand. 

Lirttz Sanpy—20 feet wide. Water 
good, and fuel plenty. 

Bic Sanpy—Barren, sandy land. Road 
good. Stream 110 feet wide by 2 feet 
deep. 


‘“ 8/ Bia Sanpy—Cutting off bends of the 


river, no grass or water on this drive, 
after leaving the river two miles from 
the last camp. 


Inter- 
mediate 


Distance. 


480 
.500 
2.361 
7.639 
.900 


13.500 
19.977 


.250 
10.000 
3.113 
6.830 
0.750 


22.587 
9.000 
4,531 
1.500 

10.2389 


5.500 


8.205 
6.201 


19.045 


eeoees 


eceeco 


275 


Total from 
Fort 
Leavenworth. 


835.531 
836.031 
838.392 
846.031 
846.531 


860.031 
880.008 


880.258 
890.258 
893.371 
900.201 
900.951 


923.488 
932.488 
937.019 
938.519 
948.758 


954.258 


962.463 
968.664 


987.709 


276 


DATE. 
1849. 


Aug. 9 


6c 10 


6 11 


ss 618 
(73 90 


APPENDIX A.—TABLE OF DISTANCES. 


OUTWARD JOURNEY—Continued. 


PROMINENT POINTS AND REMARKS. 


Forp or Green River—Good camp 
grounds on this river. Ford crosses 
diagonally. A ferry is kept near the 
ford at high water. 

Camp—On right bank of Green River.. 

GREEN River—Road leaves river, and 
crosses to Black’s Fork. No water 
for 15 miles. 

Buack’s Forx—100 feet wide, 2 feet 
deep. Grass and fuel. 

Camp—On Ham’s Fork..........sseeeseeees 

Ham’s Forx—Cross the stream, which is 
50 feet wide near camp; good bottom. 

Buacr’s Forx—Strike Black’s Fork 
again; cross the stream three times. 
Grass and water, for camping pur- 
poses, along its entire bed. 

SmirH’s Forx—Tributary of Black’s 
Fork. 

Fort Bripg—ER—Camp, one quarter of 
a mile east of fort. The road, for 


several miles, leaves the fork to the | 


right, but it can easily be reached 
for making camps. There is a road 
from Ham’s Fork, on the west side, 
to be used when the creek is swollen. 

Muppy Creex—Empties into Black’s 
Fork below Fort Bridger. The road 
is the same as is described in the 
homeward journey of 1850, as far as 
the mouth of the Red Fork of Weber 
River. 

Bear River—The road winds up a 
ridge, and crosses a mountain. This 
can be avoided by striking up Muddy 
Creek a few miles, and crossing over 
to Sulphur Creek. 

Rep Fork—In Red Kanyon, below 
Cache Cave. 

Weser River—At junction of Red 
Fork. 

Camp—On right bank of Weber, at 
“¢The Obelisks.” 

Weser Forp—Ford good, but current 
swift, over pebbles. Good timber, 
and feed. Leave the river here, to 
avoid kanyon. 

Lone Hitx—Here you ascend a hill, 
winding along a rivulet of good water. 

Summir—The descent from this to 
Bauchmin’s Creek is on a side hill, 
and dangerous. 


BavUCcHMIN’s CREEK—18 feet wide, clear | 


water; fuel and grass plenty. The 


Inter- 
mediate 
Distance. 


11.778 


1.000 


2.250 


15.500 
3.875 


2.000 


12.125 
18.727 


14.830 


19.082 


18.754 


16.540 


2.000 
2.000 


1.500 
4.750 


4.500 


eeoves 


ercces 


14.830 


19.032 


tecces 


Total from 
Fort 


Leavenworth. 


999.487 


1000.487 
1002.737 


1018.237 
1022.112 


1024.112 


1036.237 
1054.964 


1069.794 


1088.826 


1107.580 
1124.120 
1126.120 
1128.120 


--1129.620. 


1184.870. 


/ 


{ 


1138.870 


i a li lg eT ELT Ne at IEE 


APPENDIX A.—TABLE OF DISTANCES. ZiT 


OUTWARD JOURNEY— Continued. 


Total from 


Inter- ’ 
— PROMINENT POINTS AND REMARKS. mediate ae Fort 
1849. Distance. * | Leavenworth. 


Aug. 21| measurements are taken, to-day, from 
Clayton’s guide-book—the Odometer 
being disarranged. 

‘¢ 22 | Bra Kanyon CreeK—Cross Bauchmin’s 
Creek thirteen times; then, at 8 miles, 
begin the ascent of the mountain 
along a small creek, well wooded, 
and grass good on the ascent. The 
summit, at 12 miles distance, gives a 
view of the Great Salt Lake Valley. 
Hill steep to descend. 

‘¢ 23) Bia Kanyon Creex—Leave this creek] 4.500] ...... 1156.870 
and turn to the right. You can pass 
down on the left through the Golden 
Pass, and avoid this mountain. 

Emigration CrrEK—Youascendalong}| 1.750] ...... 1158.620 
hill, and then descend a steep one, 
to this creek. 


1152.370 


Camp—On Emigration Creek............. 1161.620 
MovutH or Kanyon—Opening to the} 2.000] ...... 1163.620 
valley. 


1168.620 


City oF THE GREAT Sat LAKE........ 


G*. G9 OF ol IL u8V LPP OOP =e 
ww eeeevce 
aa 
oO 
a 
= 
H 
RM 
4 
A 
fy u¥L 89 oLIT | 9.0 SP OP ¢0'L 
oO 
Hy 
4 
A 
< 
= 69°8 
< 
‘a 
eH 
A 
a 
<=) 

Ay 
<q 180 90 oSIT «80 JOP o0P 
‘opnqts : ae “eee 
PuyLsao'T opnzryeT moan s seq 


8l% 
¥9°S 


i aed 


G0°% 


69°8 


‘9008 4ST, 
©} 8Ipe ur 
-10} UT 


VSS, ‘yaivg 8,AoptVg Jo opis 4som 4v ‘oAoqe Sy—HONVUgG 
‘ASSBVIS PUB 9]}U0S JUSOSep—snoTAdAd 4o[NATI 
% pessOLd SUIAV][—NAGTAD S.NINHOAVG OL HONVUG AUVLOAAIUY, 
"9019 
uofuvy JO Y10,7 YING 04} JO Solavpnqi} [[VUIS [BIOAOS SSO.L0B 
puv ‘ojjues used Sty JUSOSY—‘GYNVY HOLVSHVAA 40 LIWWag 
‘OPIM JOOJ Q Yoo) *107 Ad 
pue ‘jeny ‘ssvaod jo Aquatd yAIM ‘yo010 oY} avo ‘puR], 9[qv4 
94} UO SO] pvOY—AATUD NOANVY JO MUO HLAOG AHL IO auoq 
‘putas st Arteu00s ey, “toqvy, Aq poaordut Sureq 
SI pvorl oy, ‘poammber orvo yvors—sowl} [vaoAos wved}s 
OY} SSOLQ —“poyoRjop oq YSNUT O]}7VO PavMAOF OY} YVY} JdnIqQVB 
OS 91 SOAIND 94} JO 9ULOS pu ‘syx90L AV[NOIpusd.zed jo syutod 
punoae ‘soovld Auvu Ur ‘ynd SI pvor OY, “osnoy-][[0y 94} ssvd 
pus ‘uokuvy sojugq °‘Z ‘OoN duvwQ—uaauQ NOANVY 40 su¥40,7 
"TAY W99.M9G 
§}001}S PUB SOUBT YIM ‘S}OT O10B-9AG UL Wo privy : A410 o4y 
JO qos ‘oarvnbs soptur xis ,“pleLq Sig ,, oy} Jo avd vw ysno1yy 
sessed pvor oyy, ‘yooagQ uoduvy Sig 10 ‘saya-a0y-04-290 
uo UIvjUnNOU Jo oseg ‘T ‘ON dweQ—NOANVY SSVq NUaTOYH 
‘syurvg Avp-1044e7 Jo Aouoptsorg oy} Lq ‘QPRT Ul popuno,y 
‘SOSNOY-910}G OYIIT, puv ‘osnoy-jJano0g “uIP, v ‘szURZIQVyUT 
000G SUreyu0d pu ‘Ax041IIET, YVIQ Jo [eytdey :4ooy Foz Aq 
SSL SPOT : OPA JooT TET szooays : 0a3u00 04} UI yooTq-o[duay, 
oY} IIMA ‘sorenbs UL yno prve[—suUIvzUNOM Poesy AA 94} WoAT 
Inds @ jo osvq oY} Sucre ‘YINOUT S}I OAOG SoTIUL U9} ‘TOATI 
uvp1Or JO YUBq OY} Wor ySvo SuIpuoyxXe ‘orenbs sol[rut Ano 
‘OIV'T WUBI Yvory oY} Jo AOT[VA 94} U[—ALID AAV] Iv tvaUyH 


“SHUVNGY GNV SINIOd INGNIMOUT 


66 » 


83 i 


1Z ‘sny 


“OS8T 
“TLV 


278 


“aunossipT ay} UO ‘YyQLomuUaQnaT 20,7 0) hn) aynTT NUS! ywatH Wolf KANYNO CUVYMANOF, 


279 


TABLE OF DISTANCES. 


‘ APPENDIX A. 


«90 .2G ol IL | 1°19 90 olf 


wV& OE oLIL | GulG GO olP 
u9V OV ol LT |L0°ULP LG o0P 


u9G 9 oI LT | Sulh 9 OF 


8'.9F 88 ol LL «80 i8F o0F 


66°99 
09°79 
68°6P 


86'°8P 


&L°0P 
GLE 
09°SS 
68°08 


88°63 


66° LZ 


ge9OL 


eeeone 


OT'6r 


GFL 


“STTIY etoe) 
oy} WO VST1 oy} 4V puB ‘etey yno Suipvords Aozjva Assess 
VY ‘Opis W}40U oY} WO SYOOA OY} WoOA, SuInss[—soONIudg GAY, 

‘MOATBY POY UI SUOT}PV.10q 
-I9AOL OY} WOT ,..Y9IQ OYOT,, pol[Vo “yZL0U 94} TMMOTZ TOs 
-UBVY Pe] S10}U9 WBOI}S [[VUUS W—UOON—aaN Tg M000-NaMOIHO 
reeeeeeeesecseresees coveees Tagag syueq {uosves stq} 98 IQ—ANIAVY 
‘S}IUITINS popunod 4IIM ‘doo4s aae¥ ST[TY OY 
epIs YINOS 94} UG ‘onbsoanjzord Ajsurpoooxs o1e pur ‘squrod 
7}99}-MBS OFUL JNO ‘FSIY JOO OG] Iv SYNTQ o7vsLeM0[su00 puv 
eUoJSpUS OYJ, ‘potmMbor st Suratap pnjorvo pur ‘does o1v 
SYUVG OY} {SOUT} [VIOAOS YAO poy Sosso1o proy—g ‘ON dNVO 
‘y10,q poy jo AaT[va oy} dn uany puv ‘puofoq spaek Qog 
pBot pjo 94} urof—azoary 10q9 A Y}TM DoTjOUN! s71.1v0N—aU0g aay 
‘IOATY AVI JO SUISSOIO V OF YOOLO STYZ WOIJ YO SpvoyT 
[e411 W “qoogs = Aq Og Sa0qoM oy} Jo Areynqiuy—wAU0,7 S.NIMO][ 
‘OUT[[OAVI} JO UOSBES 
Q[OYM OY} SuLInp 9[qvpxoZ puB “SUISSOLO YUOT[ODKS UB ST SIU, 
‘ploy ssugq woploy azo ‘xoddn ‘pazoz roqoM—yF ‘ON dmepQ—auoy 
‘1vpdod YIM posoutsy ATOSUOp JOAIA !SO[IU F LOF TOATI 
IOG2M JO PUL] W0Z}0q 94} JO [OAST 9Y} UL U99G SY PBOL OT, 
‘IOATL LOGO 0} 90UBIZUO SII VOU ‘PALOZ OY} YY—UMATAD AMATIC 
‘JUBpUNQ’ ssvid pus 
suodsy ‘esprit oy} JO Opts ysvo oy} UO St Surads SIyJ—DHNIUdg 
‘O]JUIS YUOOSOp puB yUSDSY 
— O01 TOATIG puv TOATA JOQoM UooMyoq aPdIyY dO LINWAY 
‘pdoj pooyy “yaVg s,Ao[AVg IO “4GsoA UI Sv ‘SMOTIIM pus 
{possvie [[om “yooz 7 Aq QE Yoord { yavg ISVm U[—ATTAD UAATIC 
“AQ IBOU STII OY} UO BISTUL 
-011V ‘poo ou ynq ‘Az1UTOIA oY} UI sdutads [009 eu [vseAeS 
O18 0104} > yIVG jo yavd ysvo UO—g ‘ON amigos atop 
"asep T 
pus ‘Opra yooy OG WRITS ‘yIV_ OY} SUIGIP paOZ AOMOT BY} 
MO]EG SuILO}UE {TOATI LOQ9M JO YoUvIq Y—AAAUD S.NINHONVG 
‘QSUBL 913389 OU V—II Yono1y} 
SUIUUNA ‘MOTIIM TIA POSUITT STUVOTIS YIM ‘MOpBoTE [NjIjnvEq 


§ 9? 


Zz 4ydeg 


Se 


APPENDIX A.——-TABLE OF DISTANCES. 


280 


ors6 7 62> | oN} JO UN, OY} pee “TTY oy} Jo yoos oy} FY-—aSVG NIY 
‘ox ‘SesLOY po][Vs LOZ 
[VlOyousg oq OF pres st pus ‘suOSeM Sutsvois Loy porsyyes oq 
UBd JOTIRISGNS SIy} oT0yA ‘scutids iv} oY} O18 o10Y} ‘Y}NOS 
Q[IM 90 YNOGY ‘spIs yILOU OY} MO Fni[q oy} Ur yno sdowzo 
[v09 Jo peq Wy ‘ssutads anyd[ns Su014s [er9A0s “4Ysvo 90Uv4SIp 
qIoys @ pus fy MOS oY} UO 12yvM Pood Jo Surtads yusT]I0xe 
OL F6 whee €6°L | UV ‘oprd yooy QT Yooro { Sutsso10 oy} AvoN—UGHUD WAHATAY 
*POOM 109309 YT 
POIdA0D SYUVG TOATI OY} PU ‘APIA OTIM & J[VY 91ey ST pULT 
110}}0q Sys, ‘equiva ydop—oprIM yoo} QOP TOATA puv ‘4JTAS 
27°26 song FG quorind {setqqod ossae, uo paoy f yuRq YYSIYy—uAAIyY uvag 
65°16 Seance GOL | crete ceetesteseseeeeeneeentneeetcreeeecereeeerereee ATM MVE LO WOLLOT 
‘[OAVT IOMOT BV Yonory} ‘o[iut J[Vy vB YYNos 
94} 0} poltarvd oq plnoys pvor oy, “Asvo pus opjues yUOOSOp 
FP'68 ampe'ee 92°9 | puv yUDOSe OY —ABATY IvOg puv YooID MOT[AX WoeMJOq LIWWAG 
‘JUIOd SIU} VOU POssord oq 0} ST ‘oor MOTIOX surol yey 
JOTMALL [[VUIS B ‘YI0IQ eTPIeN ‘sotqqod poyueweo ‘ozvI1OMO[S 
80°28 etna 838° -009 JO SOssvmM OI}[VSVq :syOoL poyutod YSI_}—saIGGaN GHL| Fb » 
‘punois way pus ‘yzoouIs ynq ‘doeqs ‘durvo sty} 03 pueos 
-op 0} ‘T[IyY—syuvq davys pur ‘oprm yooy ge wIvod}S ¢ LequIT, 
OU {SMO][IM YJIM PoOSULIy ST WOM ‘yoo STY} UO MOpvouT 
02°68 L8°9T CPT |eug y ‘syLOJ Moyo oTIU v J[V—9 “ON duvpN—xAaHD MOTTA A 
‘YJOOUIS pvor 
GL°08 sh cg" ‘deeys Ay0ad soovyd pmos ul puv ‘“SUOT ST TITY oY, —LINWAg 
02°08 sacegs OL’ | tt -FIUIUINS oT} 03 4UEDSV aLOJOq [[IY oY} JO osvg—NNIUdg 
‘SPNTQ oy} uO 
savpood puv ‘AyuoTd oSus puv ssviy “pvor oy} JO 410 O[IUT 
POLL ee PSS «| BVT ‘SPOT Oy} ur st oavy oyy—Areynqiay B pue GAvO aHovO| ¢ “4deg 


uGG (LO oL TT | 6'u80 80 ol PB 


w8L PL ol LL | 6.00.60 of B 


“£400 ‘qoroyy, | OUI “OS8T 
"epnysuory “opnyryey OMT 18S 3 £0q oyvrpeur ‘SHUVMGY GNV SLNIOd LNININOUT : 
WOT [BIOT, y -10} UT avd 


*Paenu27}Uuo)j—AANUNOP CUVMAWO FL 


281 


TABLE OF DISTANCES. 


APPENDIX A. 


wOG BT OTT 


9°99 86 olD 


8G ZB OIL | B'uZL ST olf 


100 8F oO TT 


9F 21 olF 


SP LST 


96°8PT 


LY’OFT 


GG PPL 


OL 861 


GP PST 


0s FIT 
LLGIT 


08°80L 
G8°66 
66°66 


ecoose 


LLLT 


Gov 


g9'9T 


eooeee 


09°96 


seecee 


69'8 


6h'G 


96% 


69ST 


G6" 


STOL 


eo'T 
L6°S 


“fquepd oses { Ayuvos pur ‘ystyoVAq 973417 
e rojeM {yNTq euojspuvs posser Jo quriod & Jopug—PNTIUdg 

‘YINOS oY} O4 LOYJAVJ ‘OyNOT MOU B JO WOT} 

-G1O[Ax9 OY} OOUSTAULOD “ZYSIXI OY} 0} SUIUIN, puB ‘ssvq YNOg 
84} YSnoIG} pvot plo oy} OAVET OM OLOF[—AVOY ssvq HLAOG 

‘SSBId poOs puv ‘soysnq JuBIIND 

MoTTAk pus YOVTQ ‘pooM-m0}}09 Jo seaody “Appny 9q} Jo 
qmou oy} optsoddo AyrvoU ‘“GuVq yYsT4 Uo dmBO—wWAOT 8, MOVIg 

*JOnF TOF osvs pus 

SMO][IM {MOT[VYS SOpTM Joos OE SSUIvJUNO Ivog Jo Joos 4v 
‘QULUS OY} IVOM SOSTY—YIOT SOV JO PouvIq—wWuAoy 8 HLING 

“IOBUOT ST FI NG ‘YO SStaaVy JO YJNour oy} 07 ‘Yueq 

qJO[ OY} WO OS[V YOVA, VST OLE], “JOT OY} 0} AVF FI SOAVOT 
Wd}JO pus ‘yoII oY} JO sUANY oY} HO 8ynd pvoy—AAOF 8 WOVIG 

‘SUIVJUNOPT OATY v9 oy} JO 

JOO} OY} 4B AT}JUNOD OYV] OY} SUIVAP PUG ‘SPUVIST [VIOAVS SUILOJ 

puUe “410 OY} SAO’ SOPIAIP Woot oy], “Suizvas ouy pue 

‘any Joy oouBpUNge UT cequiIg, “ysod Axeyt[IM v Loz UoTyIsod 

poos B fepva} ALIOY} SpueMIM0d puG ‘SUvIPUT JO Soqtt} [Voss 

moeMjog peyenzis A,qvarupe ‘{4sod Surpea, & soz ‘osplag 
‘supe fq [Inq { 1OATY WoseTH Jo YAOT Sov[g uQ—wadalug 1x07 

‘ado[s o[}UIS pu SUOT 

90Udy} SUIpusdseq ‘[ONJ LOJ oses puv ssvais poos y}IM ‘yoo19 

SIY} WOT] SoTIU J[VYy v pue ouo ‘scutsds wos “pv oq 0} 107M 

Ayeusn st etey} ‘o[qQv} oy} WO pus ‘pusose nok YoIyM TINT 
dooqs Jo osvq ye :mojoq Appny Oy} sutof yoIyM—AAAUD TIIPT 

‘ploy pood f opt 
joy OZ Wet} = HIOT SWovig jo Arveynqizy—wAaAp zaaayy 

‘UIVIJUNOP, WIY oy} Woaz ands wv jo 

emogyoVq oy} uo AUIS popuedsep svy peoroyy, “woat Aq por 

pesury ey1sodep & OAIs ouLoS pue ‘atoy onsst sdutads [v10AS 
:outava Assvis pvouq &@ U[—y/ ‘ON duvgQ—sdNIuag svamddopD 
“ees SUIDUIM pus ‘epjues ‘Asvo ueeq sey yUI0Se OY .—LINWAg 
“=noom jo AyuoTd £1078M SNOTOI[Sp pus [00D—NYNIUdg Nadsy 

‘[OA0] AOMOT B UO ODPL OY} POSsodd pus ‘ToyJAIVJ OOUYSIP 410YS 

® YI2ID inyding dn poMoT[OF OAVY plNoys pvod oy, ,, UIsVg 


GL » 


1 


OT » 


APPENDIX A.—TABLE OF DISTANCES. 


282 


«60 186 o60L 


uOP .8E o6 OL 


ul 0 OIT 


‘opnyisuoy 


8°u0F VE ol P 


GuPP OE ol P 


6°80 TE ol fF 


‘opnqyey 


LI'LEG 


€8°81G 
16°G1G 


T&°G0G 


L8°861 
86°S8T 


I8°SsT 
G8°6LI 


T6°TLT 
06°6ST 


“£419 
exeyT 4189 
THOTT [BOT 


98°V% 


eenece 


60°LT 


LééL 


PP 'GS 


“TOrLeyy 


s,Avq 


vé'8 


669 
09°0T 


vP's 


69°T 
LY'T 


9v'P 
PPL 


TOSI 
CVG 


90TR IST 
oyeIpeuL 
-10jUT 


SSUvAG ‘oUojspuvS AVIS JO SyOOI FnIG YSty Jo osvq oy} woz 
ONSST pus “YULq WYSIL oY} UO oAv SoUlAds OY, ‘[oNF 1oJ UMOp 
pequn1o ATIsve e1v pUB ‘yoeL0 OY} SuOTe synTq ey} UT yno 
dous [woo jo speq [vtoAog—FT ‘ON dmvg—syNIuag unnatag 
‘dooys syuvq ‘deep yee g puv ‘apr 4ooj QQ yoor0 
'44SII OY} UO 9dVA, SSuUvAy YUL yey ey} 07 ssorg—auog 
Treeeeee see sopmod [BUS B ST 4SVO-YZIOU OY} 07 O]IM B J[VET—aNog 
‘snoanydins pue ysrpovaq APY SIs 
10yVA ‘Tony AjUO oy} oSvs pu ‘poyruty L10A ST oynOA OT} 
Jo yavd Sty} UO SSVI OY, "4410,f S.OSprlag 0} Yooug 10391g Jo — 
Y}LOU SONUIZUO YOY “GFRyT UL ‘SesULYyIy Wor syuvAsIMO JO 
Ayred B Aq opvul ova} B SI STYL—eT ‘ON duevg—aovay, s.sNVAT 
. “YINOU OY} UOIT SoTIUL OM 
‘BNC YSry B tepun ‘yuvq Jo, OY} WO TVOd st ouoyg, ‘dueg 
WOTF YIU OUO UL YoolH 10}}1g OYlIIG—augq-1v0O MATUO ATLLIG 
‘qUBpUNY’ Jong £ yoorg 10991gG 
JO 30]}NO OY} AOU OF YUV Yo] UMOP SUTMOTIOY—ZI “ON dNVO 
"1109}0q 9} UL YOoLO v JO 
jouuvyo Arp oy} SuoOTV SpUIM puv ‘MOTOFT IQqQUy Ozu spudos 
-op pvol oy, ‘A[[VUOSvIp ssoto f10yVM oY} Jo 28u3s STUY 
4B poos paoy f apr Joo (QQ TOAtE fyuvq yySiy—uTATY NaaUyH 
“MOTLOP] FIQQvy UMOp sduey} { ep}UEs yuVOSe pu ‘poos 
pvoy—aAeAly W9etH puv YI SyovlgG woemjzoq ospria Jo Liwwag 
‘JonjJ Loy soysnq 
-juvLind puv AI19q-o[vynq Sopra yooy QGT yoouo $Asvo paojz 
‘p00S pBOt { MOSVES STY} YL 1OJVM OU { PULT 21qB7 O43 JO Opis 
U19}SVO OY} UO ‘seTIM xIS JOJ ‘odors oYUZsS W—wUOT SMOVIG 
‘YIOT SOVT_T JO puog yvors oy} ssor0v sory 
9JNOL OGY, “OPNIT}V] 94} SOAIG WOT}VAIOSO [VUOIPIIO|{—NIVIG 


‘SHUVNGY AGNVY SINIOG INGNINOUd 


“Panunjuoj—KENUNOf GUVMAWOPT 


i) 


ig) oe) 


cL 


“OS8T 
“aLYC 


283 


TABLE OF DISTANCES. 


APPENDIX A. 


u8V eked oLOT 


u1G LP oLOL 


uL¥ GG oLOT 
uLL «G0 o80T 


u¥G WT o801 


«00 .TE o80T 


«60 «LP o801 
uV§& (OG o80T 


ul¥ LG ol ¥ 


G86 (86 olV 


1°90 (13 ol B 
u¥VI 86 olP 


u6E 86 olP 


uGL WG olP 


60 86 olf 
ul OE olf 


POLE 
90°606 


68°906 


00°96¢ 
E1186 
¥G'6L6 
G9'°GLG 
FS'696 
96°896 
TL°69¢ 


8P'60G 
GP 6&6 


L6°E&S 


amas 


GP'0G 


eeoeee 


10°96 


eeoeee 
seocse 


T&°GS 


80°T 
L9°S 


68°0L 


18°8 
68°L 
69°S 
T0°9 
8g" 

98°6 
69°6 


90°0T 
aya 


08°9 


‘soysnq puv 
80013 aeped uemdazvose oy} uo puv ‘Arroyo ‘Lateq-ootAaos 
‘uodsy ‘Atp-0} SSO.19 0} SoT[[NS [eIEAVG ‘SIOqtINU o[qevzEpIs 
-H09 UI d10Y We9eS SUILP Ysotg “popavns ATISve oq UBd 977}%0 
‘SSBId POOS YIM otttvrd T[VUgG—e] ‘oN dwvg—sNvg usAvag 
“UOSBES FINA OY} UI JUBPUNGe syUVIaMd puUv soTLIeYO 
‘erquorjovrd epeu Alisve proy—wuaGUQ AdaoyW dO NOANVY 
"‘SYOOL 
pus siz does Jo 4ooy oy} 4e ‘puR[-o[qv} MOTIvU UO ST peor 
oy} s1eyM ‘UokUvY oY} 109U0 OM MOOS ATOA ‘puNnos3-0]93%q 
SonbsvA «vou ‘yuvq JysI4 UQ—AATAD AGA 40 CNG HLUON 
"SHBPOD JIM PdLOA0D ‘peor oY} JO 
Y}1OU $999Nq SMOISpULS OMT, ‘OPIS ySvO dT} UO YUDDSep dood}s & 
SV WOM Vopll V puNore ‘YINOS OG} 07 ssed prnoys por ony, 
"LOATY OYVUYG 9[}4V] JO pouvrq W—/T{ ‘ON dwvg—uyauH aaan, 
‘SOSUBI O[VYNG puv sureid Assvas 10A0 ‘poos AIA St 
pvol OY], “YILOU OY} WoT ospIt MOT V UQ—I[vyY BOON—aDaIYy 
sre eereeeseerre agro APpNIA,.JO SuUIv{Ad OY} JO Opis JSOM JY—SNIVIG 
"U1y} AOA SSUIG PUL ‘TON 94} SoYsIuUIN Ysnaq-esvg 
‘qSBoq JO UVU LOZ O[QVYUIIp Ajoreg ST 1O}VM SINT, ‘“YIOT 
SJosplig JO pvoy ‘109UM oUT[VyTy—9T ‘oN duvg—spNiudg 
‘JSVO PUL YINOS Oy} UO snvozv{d TesAv] WoTT 
qaedu Surpueys ‘Avpo Jo ‘YoIy 400% QZ Wooveq W—HOVLG-AV] 
“VIUUENS 07 
edojs oyju0s v dn ssud ‘s103eM yo0I9 10}91q BuIAveT—LINWAg 
“ALUTOTA SIY} UL YOUBAG UIVU oY} 1o}Ue 4Sve-YINOS 
uo ssulids woIZ suveTys [[vUG—}[vy WOON—aadaTg DNIvag 
‘Ayuotd odojoque pu “uepunqge susy oseg 
"[ONy IOJ OSes OGAv] pure ‘poos ssvrg—eT ‘oN duvgQ—wu0 Isvy 
““**"9DNIT}C] UL WOTYISOd SOATS MOTZBVALESGO UVIPTIAOM OY ,—NOON 
‘syuvq-Ae[9 doojs Jo JuN0DDv UO 4NOUZIP AOA SI Sursso.10 
oY} yng ‘eduLAsIp oy} SouOpord “{[IY STYY MOA, YOOTD 94} MOTLOF 
OF, ‘punord [OAST OY} MOT[OZ UL pvor oy, “no Surtddo.z9 
[G00 YJIM ‘osprt B sso10e puv ‘yUeq yo, 04} dg—ayaIyY TVOO 
*IOATY 
2448[q 94} «vou ‘surof pus ‘qouerd 4svo-4}.10U oY} SoyVy OVA, 


61 


8T 


LT 


91 


= 


99 


2) 


9 


99 


APPENDIX A.—TABLE OF DISTANCES. 


284 


WALT 90 oLOL | G'u6P ZE ol 


u6G 1G oLOT 


u8P OE oLOT 


*opnyisuo'yT 


ulG GE olP 


$66 i&E olP 


‘opnyryery 


9T1Ss 


ST'8rs 
960FE 


6L°ESE 
68°1é 
VG LEE 


T0138 


6L°818 
Tv'318 


“£410 
oxy 41eg 
HOI [VOT 


62°9T 


@escee 


GLTS 


“yore yl 
s4eq 


86° 


G6" 
LYv'9 


06°T 
G9'P 
&6°9 


GGG 


889 
LOG 


dome SIT 
oyeTpoul 
-10}U] 


‘uretd oy} 03 A[[enpeas ostz ooUETY, 
‘Asvo syurq pue 00g 7 ‘oSvys quosord ye 103vM ysodoop 
ferddiz @ 1aA0 paxoy f possvas [fom AoT[VA pu s][IT {sosous 
POOM-U0}}09 UL YU FJeT OY} Suoje prot vyy—awoq ALIvIg 
*JUOUL 
-91,J ‘JOO Lq WATS Opnytsuo0y, oy} Lq poyooyo oav yutod sty 
0} S90ULISIP poInsvem oy, ‘a1oy YIM your A,yUenboaz ore 
sp1oy opeynq {Suizvis oug {seqta, Surmoqysiou jo punois 
-IVM :8}LOJ UVIPUT P[O Jo survmor ‘yavd puve osaorsd avypdod 
[HJINVIG BV :yURq IoT—uUGAIY WIVIg AHL 10 WUOT HLMON 
‘OSPII pol[ns B 19AO ssvd pus pusose uoy} ‘ uIReTYs 
AOT[VYS YStIq B@ { oprim yooy QP ‘ yJnos wory—uaGUD OVE 
“AI][VA YIN VSVY sosso.ro 
9L0Y YD J0OIG WoIy 9ovVAy, SUVAT ‘“SUTLJUNOT MOq-ouTO 
-Ip2J 94} pavaoy St durvo oy} WoT UOTJOOIIG—HOVU, .SNVA 
‘100d OYIVA Ssvas {yuLpUNgV osvg “YINOS oY} WOI, MIvIT}9 
[wus @ jo yurq WYSE UD—OZ ‘ON duvQ—HONVUg Wiad] 
"1989 OT} 0} 
[OMS MOT B IOAO Ssvd pu ‘qOUvAG SITY} OAVOTT—HONVUG ISH 
*10940q PUB IOZLOYS oq p[NoM ‘uIeyuUNOT Uedsy Jo 
QSVq oY} SuUo0TV ‘sIq} JO YINOS oY} 0 OYNOL OY, “YoorD osvg 
0} Youvaq v jo pvoy “surads ovoid uodsy—q[vy UO0ON—PONIUdG 
‘10JVM PUB [ONJ JO SUOTPVIOpIS 
-100 Aq pouIULoJep Suroq oo1oyo oy} ‘soovjd Auvum ur possoz90 
oq Uv puL ‘SUILIUNO] IOATY PUTA, OY} 07 SonuTyUOD AryUNO0D 
ousedmvyd oyy, ‘ssurads [[vus [eroaes ssvd ‘opjues AOA 
JUOOSY—SUTLV A OLINVILY ONV OIOVG THL NAWMIGE LIWWAg 
‘SUIVIUNOP, YAVY OY} Ul sosta 
YI ‘SYAOJ S}T MOTO O[IUL ONO YO0IO Siq} OAVOT—UATUD ACAA 


“SHUVNGY GNV SENIOd INANIWOUT 


‘PaNnuyuUuojg—KANUNOP CUVAAWOTT 


co” 38 


1G» 


02 “3deg 


285 


APPENDIX A.—TABLE OF DISTANCES. 


«8G ST o90T 


uL¥ VG o9OL 


ul 98 o90L 


uL& 8 o901 


ul& &G o90L 


«90 S$ ol PV 


9°88 8 ol 


6°09 TP ol B 


GuG¥ OP olP 


8ST LE olf 


96°868 


66°S6E 


G8"868 
LG°068 


66° L8E 
OF 986 


60'8E 
808s 


OL'8Lé 
O08 9LE 


€9°GLE 


68° TLE 


0F'998 


TS°S98 


TS 198 
G3'098 


r6°9T 


eececee 


99°OT 


16°&3 


seecee 
eeeecee 
@seeee 


eecoece 


L6G 


VIG 


89° 
GEG 


Gg'L 
887 


8'T 
80° 
08°T 
L9°§ 


vo'T 


66°97 


60°S 


00% 


90°T 
60°6 


‘moduey 

94} WOAJ o[u-j[vVy & qnoge ‘deep yooy & ‘aprm yoo g $ AOATYT 
MOQ-OULDIpey, JO Wouviq y—ez ‘ON duvg—wyaGUD 8 ,dddvuq 

"purely W107}0q [HJIyNVeq Fsono14} sunt 

pus ‘poq ATqqod yuortmmo prdea ‘deep zoo; [ pus ‘opt joo QZ 
yooro { uokuvy 94} AVON—UAAIY MOG-ANIOGT BUOY LSVy 


"SSUId poos 
 TOATY MOG-OUTNIPOT JO YLOT Yseq 07 Arvgnqiy—wAqwuDN waaty 
@eeeeereaece Cceoeeeeeee ceeeeeseesr eeseeesee et) LF MOON —NOILVAUHSAO NVICIUEL 

"poos 


‘soq0}Vd UL Ssvis {8907} 1OY}JO pu YOrG [[VUg—NuaTUD HONIG 
j sauynb quanbalf pup abvs absn) fo 2un0000 
uo ‘ojn {ng uns 02 snovabunp upd ay? fo jund siyf,—ARTUD NASY 
‘quero pider ‘doop SouoUl g PUR 
PIA Joos Gz ‘eBe4s MOT JUOserd 4e TOVVA\ ‘WAG pue ‘popoom 
‘MOT OG SyUBq 94T—ZZ ‘ON dwuvj—ugAIyY MOd-ENIDIGAY 
“SCUIN[O-MOTIIM £90048 4SOMOT 
‘doop sSoqour g puv opm yooy g ‘Surids 10j}vM av9[Q—xGGED 
tes eeesaacenseessnses seevausen seeeeesenseseraen seteesaee seeeeeees TOTIBATOSO 
woou fq opnyyey ‘107M oInd Jo Yoo1d [[VUgG—uGEND aLLaAg 
"yoorQ ong 07 ATVINqQI4} ‘syeor0 
Surads [[VUs [VIOAVS SUISSOLO ‘SUIV]G OIUIVIVT 0} JUGDSOp IYSITS 
% U9q} ‘pvor poos ‘oadojs eues v uo des 0} pusdsy—LINWAG 
‘9[}}VO LOJ SSBIG-TOUNG poos puv ‘fon 
IOJ GOI puv ‘SMOT[IM ‘MB £09ING MOQ-OTLOIPSTY pUv ySOAK 
Wo9M}Oq ‘Yooly oyvuselyyey JO Wouvaq yNos UO—TzZ ‘ON dNVO 
‘SO[IUL OM} IOJ SUIPUIM por puB ‘ozIsOdep snodsd¥T, 
-[Is1e@ 9Y} UL sol[jns AueU :4UeTIND YstAqg $90B4S YSOMOT 4B 
‘soyour 7 Aq ‘oprIm 4aoj Gg ST $41 Jo SoM OY} UO STII 94} pus 
‘OVJNG MOQ-oULIPET, UL Spvoy Yo010 SIU —AATUD AMVNSATLLVY 
079g 
MOQ-9TIOTPIT JO OPIS SOM OT} WO SO44Nq JO JOO} WO1q—HONVUg 
‘soysnq-o0es ydooxe ‘pooM jo oynz14Ss9p 
urejd {ssvq MOQ-OULDIPey, OU} WoL, Spusdsog—xAaaUD ssvd 
"HOL}VALOSGO UVIPIIeT + *o}}Nng MOg-euTO 
-IpeJl JO SOM S][IY OY} JO osvq OY} 04 SUIPUSJXqY—NIVIg NO NOON 


GG 


VG 


9) 


99 


TABLE OF DISTANCES. 


APPENDIX A. 


286 


all ¥G oGOL 


u6& GP oGOT 
w6L L9 oGOT 


vG& (90 o90T 


‘epnyisuo0'yT 


u60 80 olP 


P'ulP ST olf 


P'uSP 6T ol 


u9T 86 olf 


“opnyyey 


TG" LOP 


8G°9GP 
FO'SSP 
66 'OSP 


6P SPP 


GS GPP 


A a 
gO'TSP 


L8°61P 
STLIP 
SCIP 
09°60P 


oL'SOP 


“AAD 
oye'T 1189 
wIOIy [CJOT, 


G8°1Z 


@eeeee 


98°F 
16°06 


‘qoieyy 


s Acq 


66'0L 


bG's 
9GL°G 
08°T 


LI'9 


69°L 


919 


ou IST, 
oyerpour 
-104U]. 


‘a8 pia oY} UO St pwor oy} { Suruedo uokuvy oy} J Y—AAMAY MOU) 
“TNs OY} 07 Surdeoy Aq poploaw oq Wed o9VI} OY} UT YOOID 
SINT, “ot Mory 07 A[LOYNOS WAN} 07 “oor 9[Og osporT Jo 
SPVOY PeLoq UIT} OY} ULOTJ POBUVYO SVM OFNOL OY, [.—HONVUG MOU) 
sroeeeeeoeres TOUBIG MOTD PUB TOATY OTUVIV'T U9OMJOG—LINNAG 
Ce reece serecesoe seenesoes cone eee SOULE EU utr ouBvaq [jews pus ‘ONIUdg 
‘[wnpvis oyinb st ostz oy, “odors STITH 
YOU Oy} wo ,,“W0}oJ, ONT, », per[vo ywod ouoy vw 0718s0ddQ—No.uay, 
0}98T I OY} JO YLOT YING oy} YIM uorounl 971 
04 AdT[VA S}I UMOp odUDy} {yo0IQ O[Og Ospory 0} ‘ATT0N}10U 
qoopop ‘YOolQ MOI) SUISSOLO pUB ‘OSPIT TOMOT BV TOAO STITT 
YViq oy} Suisso1 ‘AjAoyynNos 910m dooy prnoys pvor oy} 
OLOF[ ‘[ONJ pUB 10JVM POOD { pVOL [OAD] OUI,J—HONVUG ONIUdG 
‘sropvay oy Aq posn 
‘IOATI SITY} SMOTV ‘OIMIVAC'T J10,J 0} pVOI YW ‘spunoss-SuyuNny 
suuefoyQ pus xnorg ‘oovid sty} 98 paoy poos puv ssvad 
quofeoxg ‘yO T «avpdog yw yuvq YSIyY—AUOT NIV AIWVUVT 
; ‘OPNITZR] LOZ SMOTIVA 
-I1OSGO UVIpPMop, ‘“10ywVA oTqvarostur ‘spuod [[wus OMT—sanog 
‘yIOT ISOM Squepunqe 44,01d ony pus ‘opqusney 
-XOUI Ssvisd {popzoj ATISvO {WAY pUv MOT SyUBq £90v4S MOT 
‘deep soyout g Aq ‘opr yooy G7—pz ‘ON duivQO—mu0g GINVUyT 


sereeececocceorre LOATY OLUICVAVT JO TOuUBIG ‘UIve.149 [[RUS—NVaULS 

COe COCO OHe OHHH EEE HEHEHE EH Ee SO[TUL MOT Le IOF Aaqyunoo Sul[[OyY—anog 

Geese Cees eeses seeeeeeeees OpN}1}V] LOJ SUOTJVALISQO UGIPTLO]\—SNIV Tq 
‘98a OU 


WOIJ O[IU [ WIvOTYS OYVT puoo0g ‘HI0,7 oY} WO S90.1}-1e[dod 
eMOg “peor [NFVYSIop puv ‘ssvad YIM por0aood ureid oy, 
‘OY [[VUS OJFUL YWOU SuruunI wees [[VUg—AUOT AUVT 


‘SHUVNGY ANV SENIOG INGNINOUT 


“PONUNUI)—AUNUNOP CUVAMAWOFT 


66» 


8G » 


Lo 


287 


TABLE OF DISTANCES. 


APPENDIX A. 


uG¥ (99 of OT 


u&G 6G oVOT 


ulG GT oGOL 


u¥G 80 o8OT 


v8 6& olf 


8°uSE SV olP 


L'uG¥ 1S ol¥ 


9°80 160 ol B 


OF Org 
88°STS 


02°909 


GP 10g 
OF 'G6P 
61°06P 
69°68P 
8o°S8P 


PP E8P 


SP 08h 


99°TLP 


66 ILD 


TL’LT 


66 08 


e@eoeoses 
@ceoeve 


aseece 
eeecce 


ecocce 


G9°PL 


86'S 
89°6 


Gly 


66'8 
L9'T 
06'T 
lV 
VLG 


00°S 


LLG 


&V's 


GLE 


*pozJAOSUL OLOY JOU ST “STII, OUT 
oy} Suome ‘yu10d siq} WoOIZ opeM sdULSSTIOUNODEI WY “OplMm 
QIU B J[VY VOY ST YOTTM “puB[-T10}}0q ST} UQ—GG “ON ANV) 

“OpN}I}L] LOJ 
WONCAIOSYQ ‘OptA VIM WJANOJ-ou0 pULl-W0O}j0Oq—LIVH NOON 
‘MIBITIS VY} JO OpIs Tove UO puL] YsIy ‘yuepuUNgE ssvad pu 
[ONT {991M} WVII}S UIVU OY} SSOLD ‘ SoTTUA YANOJ-9914} pus 
INOJ JV YOO oY} Sussox0 ‘yuvq UMOp ssvg—AGAIY AWLVMPIAAY 
' ‘ssvg ouuesoyy 
OY} JO APIs 4Svo Oy} UO o[qv9 OY} SuIOf yey} Nveqeid & YonoaAyy 
SYVOIG JOA OY} OAFT “OW ‘SMOT[IM ‘Aopyo-xoq jo AyueTd 
ST 910Y} OLOYA ‘S0}VS SND OY} 07 ‘STG} UMOp “ooUsY} * 10}BM 
-8nyp jo youerg Lap B 09 ,,“OptAIp,, MOT B Joao ssvd woos nos 
‘qouvaq YLOU oY} Wor1g—gz “ON dwep—uqAIyY UFLVMNNHHD 


tereeceeeevorsooee QQMALI YSAVUL @ PASSO) GUIAVJ[—HONVUG HLUON 

seeeeveceereoseeeressoeres “ONIM JOO F ¢ youvslq UIB[—AIaUD ASUOP 

sieeeeeueceencsseeseuseecoeees oTIOIBIG YINOS pesoddng—uaaug asuoy 
‘sjood ur 

I9ywM $Y9aIO sog sspor7yT Jo YOuvag v ‘yoorp Iweq—TIvY NOON 

"OPIM J0OF § _ 

fyoorg uokuvy MOTOG SO[IUL OM} JNOYY—AATAD wIOg AOAdOT 

‘yuep 


-unge ‘uofkuvy oy} 48 yng ‘ssed oO} UL [ony OU FopIM yoog 
‘ yoorQ e[og espoy Jo qouvaq posoddng—HONvug F1I0g TNA0T 

"YYSII oy} 

0} SOTIUL OMY ‘PLOT OY} JO [PAST OY} GAG’ Joos poIpuNy OM} 

‘puv]-9[qe} B pue 4saM ON} UO STITH YOV[{ oy}? JO OOF Oy} 
Suoje ‘ssvqg ouuasoyO Oy} OUI ‘YJLOU SuISSeq—HONVUg MOU) 

“TOATY 0}48Tq OY} JO YOY YING 

8} 0} ‘Y90IQ 9[Og OSpoT UMOp MoT[OF AT[VUY pu ‘soyULIq 

[[VUs [er0A0s Surssos9 ‘A[pIVMysve ORIIIS plnoys yurod siy} 

WOI, pvor oyy, ‘MOTEq ysnf pUBS 94} OJUL SYUIS Yoortd ot} 
{[ONf OF SMOT[IM ! papue seer}, ‘ YUVq Ajo] UMOG—JZ “ON dWVD 

‘soTLIeyO youTq ‘ysnq MOT pus ‘seysnq 

ULOYIMVY PUB s}URIIN f9QuUITy IOF ‘pooM-U09Z}09 “IZ ‘sould 

‘Y901Q MOoIQ Jo soqouvag-Sultds puv SOUIABA JO Speoy 94} 4B 


T “PO 


Of » 


— 


TABLE OF DISTANCES. 


APPENDIX A. 


288 


uv00 GP o$0T | L°.8S 8h olF | OZ°L19 


89°609 
u8G 8G o$0L | L.98 89 oLlF | 89809 
60°609 


OF'S89 


P9'PLY 
b9'GLG 


u9G LE of OL 11GG iGL OOP a aes 


u9G 18 oPOL | G'u8E.ZL oGP | $8°99G 
I8°Té¢ 


FO'LE 


eeeeee 


60'FE 
G9'SI 


“£qI9 “OIULVIV'T 


‘opngisuoy “opnjyye'y oye] 4[BS 4.10 


LOL [BIOT, | MOAT [BIOY, 


“ore 


§ Ag 


69'L 


06°9 


69°T 
69°9T 


£0°PS 


g9°ST 


*90UR4SIT 
oyRIpouUL 


-10}UT 


‘ured oY} UO ssBIs poos 
pus ‘synig Sy} U0 siBpey ‘adpta4 oy} Soessoad pvod oY} 
oLeyM ‘SpN[_ suoyspuRY Jo yoo; 04} 4% Sulids W ‘sesnoy 
poytosep ‘plo JO MOd “o[MATL [[VUIS B JW—sIIAIG s,LL00g 

"Y9OID OSIOP{ 4B TOATL 
OY} SOAVST YOIYM ‘YOVI, plO oy} 03 ATMOS pusose OM 
‘syNi[g Sjjoog Aq 4VY} ULY 1oZ10Ys 9q 07 pres {uO 
qOU SI 4f ynq ‘“AOATL OY} SUOTe sol] peor W—wOVU, aIO 
*y9o10 [[Vus B 0} ‘ATonbri[qo ‘e74e,Tq spxvMo} U9q} 
‘OOUVISIP JAOYS B YOO OSLO] UMOP SuluINT—ge ‘ON ANVO 
‘90 ST 919Y} 91ZY— * 
IOATL OY} SUOTV OOLVOS OUTODEG SLY IOQUII[—wNaaUD aASUOTT 
‘JHVpUNG’ LOGuUIT, puvs ssvis 
Survey ‘So[IU g WL0}}0G JO |IpPIM osesaAy “Luedmog 
IN UVOLIEMY oY} 0} Sursuojeq 4sod-Sutpray—s auvHony 
"UL0}}0Q 9978[q SuoTe u9yy ‘syueq Avo 94} 
TO SosuICMT YOY ‘IOATI 9} PIOAV 0} OSpla B TOAO ssvd 
‘sty? MOTOg  ‘szoxord ynoYIIM ‘osnoy SuIpes, Y—s AVaaVG 
‘sosodand Surduvo x0f ssvais pus requir, poos 
OH8[q YON oy} Jo Yuvq FYSIL oy} UQ—GeE ‘ON ANVO 
‘OIUILBIG'T 
WOIZ Sour oAY 4sod-Surpvs, s.o}JoUOSSIg SSUgq  “JLOJ 
oy} MOTEq ysnf WIvOI}S OY} PIO ™—UTATY TINVUVT IO au07 
"JLOZ OY} 0} OOUDY} puv ‘Suradg s,10j,uUNn;T 
07 , ‘AoumiyO,, oy} Surssed ‘1oyvasny9Q oy} uMop dooy 
0} 10939 ST 4] ‘SoTIu 0004} peor oy} sBuojoad ,,‘o[0F] 
S,UdYSOH ,, JO SOl[NS OY} PIOAV 0} JSaM OY} PAVAOY Sut 
-UIN} PUB ‘PUV]-9[q%} SUT[[OL oY} 0} SUISIY—AINVUVT LWT 
weseeeee"NBOT POO { PUB] U10}}0q OY} SUOTY—ee ‘ON ANVO 


'SHUVWAY ANY SENIOg ININIMOUT 


*“panuijuUo) —AUNYN Of CAVMANOT 


St » 
Lite. ca 
9T » 
Die 
OL “490 
‘OSST 
“ILVE 


289 


TABLE OF DISTANCES. 


APPENDIX A. 


u8G 160 oGOL 


«WOT SF oGOL 


uVV 1G o80L 


08°SSL 
LuBl LT ol¥ | 8o9TL 


FO'STL 
98°COL 


88 °00L 
88°G69 
- | SL'P89 
S'PL9 


UGG SE ol | 89°CLI 


Go 199 


GG 099 


u6& GP olP | €0°8E9 


96-291 
6§°0ST 


0¢'6FT 
GO'LEL 


PS VEL 
bG'9GI 
ve SIL 
resol 


¥8°90T 


88°S6 


88 'F6 


6L'GL 


66°08 


ST '0& 


GOVE 


GS'bs 


LOL. 
61'T 


SSL 
SVG 
00°8 
02°8 
00°0T 
0G'T 
OF TL 


00°T 


61°GS 


68°06 


‘qoos $e punoy om “Gpg] ‘Alp ut 
‘Q1OYM “SOYOUL QT TOUUVYD UI 107eM qsodeep {400% Gt0ZG 


pioy JO WIPIM—ALLVIG FHL 10 MYOT HLAOG sO auoq]| ZZ 5, . 


‘SUIPUIOS 


-Op [IW 4Sv] ‘pAOJ oY} 03 peor ou oud} {pusose 07 
ity dooys “Suoy] ev wey} f MojpfoY Url 4uODSe o[QUES solrur 
JIvy @ pue ouo tog ‘paoy aoddg—of ‘on dueg—auogq 
“"" TOSBOS SIT} 7B ‘107VA LOZ “urod ysvy oy} yV¥—NviIarIuapy 
“AAOT YINOG oy} Fo 
Suissor9 t9ddn oy} 0} sano ‘sy.10f OM oY) Jo UOOUNl 313 
0} Spvoy UO !syOvT} [VLOAOS AQ POSSO.1D SI OSpTA 9} pur 
*‘Y1O,F YINOY OY} LOF SOAS] pvOr OY} OLOPT ‘syatoy O77"Tg 
oY} W9IMJO ISPTL SUIPIAIP OY} WoT, yno-Surysem oy 
fq opvm ‘usvyo doop v St sIqg, “W0}}0q 94} UT mReA)sS 
[[vus & :poom rejdod pus yse jo sdurpunqe svy MoTOT 
qsy ‘sosodaind surdurvo «oj ‘s]pry oy} ur ‘savpeg ‘puts 
AAvoy | Syn{q JO 400} 4B sonuryu0o prot oy,—MOTIOR, HSV 
“pug OF FHOBTp [SAT 0748Tq JO YUVA UO—GE ‘ON ANVH 
“SO[IUL OMY TOF WOTPLADTO SIY? 4B 
deey pus ‘19AlL 94} OAOGL 4o0F QQ yHOqe pusdsy—saantg 
‘s[ood ul ‘YyNOS 07 9OULISIP JALOYS V 1OYVA, ‘MOUS 
SUI} WOly SMOG 193v4 ‘poq Apurs W—waGUQD aug 
‘yuRq ytou f ynig deo1g oy} proar 04 ‘ospta Apueg—apvary 
‘deep seyour g ‘opr yoo gz “G'g’*g Moy SuLMO0jO—saaAD 
‘901BOS TON 5 w10730q 
oy} Jo xed Sty} Suo[e ssvas poon—ege ‘on duvQ—aiiviTg 
‘OOpPIL puv syntq 
yoor-Loumiyg purgeq Ae][VA @ UL Sosy -4seM-YQNOS 
oy} WoT sMOo,g “gooy € AQ Qooy QZ pog—uaquQ WIT 
“yueq Ysa oy 
SOTON} OTOY JN ‘SOTIM 99.17} 03 OUO WOAT SYULG TOALL OY 
mou ‘AvM oy} JO Sou ‘poqavdop sey poy ‘STII, OY} UL 
« OSNOY 4AN0),, OY} MOLO SOTIt OM} YNOYWY—LIV]ET NOON 
‘YINOS 9Y} 0} ST[TYy oy} UO 
SIZPIO pUB ‘SPULIST OY} UO POOM-JjIIp OMG ‘[eNF ou ynq 
‘SSBId POOH ‘“SO[IUL JOIN 0} SUOPIA U10}}0q—/e ‘ON ANVO 


IG 39 
03 » 
6L » 


ST ‘PO 


19 


9P'S88 | Z9'6IE at IP'ST | ore poos peor fyuvg ey} Suopy—INIOg GIIW-NanLHDIG | 97s, 
“IOATI OY} 
WoIy poyeSizat ATISve pus ‘uvoT ouLOS YIM ‘Spurs ATIVUI 
SI [IOS OY], “PAVISI oY} WOAT FYSnOAQ [ony puv ‘pesso.10 

GO'0L8 | I3'F08 | F8'bs 68'S OG S1eY UVO AIATI OY} JO YOuUBIG [[VUS Y—PPp ‘ON AAV) 

"919Y POUIvyqo oq UBD JONJ { possBAs [TOM ST ‘soTTUE 

66998 | 68°00E sara CG'OS | W9IFG JNoGs 0} Pouopra BEY W10}}0q SUT — ATAU AWATT! Go » 
*SLOT[OAVIY AG PoT[OF 


wa eq ATqvqoad [IA ynq ‘suooveq sv oAcoS YOIUM ‘spoomM 
oO TL’G§8 | L8°69G | 82's POST | -009700 osrey OMG ,,‘SeeTT, OMT,,, OF} IV—Eh ‘ON ANVD 
S “yuvq TINOoS oy} uO 
BA LL9T8 | 86°0SZ sete CL'G | AQUI], ‘O}VIPOUIAJUI SUIvOT}S [[VUIS OMT —NaGUD HLANOT 
4 "sIBpoo 
A 4 [VWOISvd00 YIIA ‘STITY pUG SPNTq OFUL ‘Y}NOS 944 WO U10}}0q 
Fy oy} seqovoidde 41 sv ‘moHOIG SI 91GB OY, “UIvIqGo 0% 
© qNOWIP poo ynq ‘soo1} YIM POTOAOD TOATI OY} UI SpUvIST 
a S9TIS | SL'ShS | ov | OBIT | £9998 Id Ou} Jo LoT[VA OG} UMOP SuMUyUOD—AAMED Lsaly | $3 » 
AQ ‘9A0Q’ Ssulids 
a WOIF SOSTA YOIYM ‘319910 [[VUIS B UO duIvOUD puB ‘o[IU Ou 
, 2998[q 94} pavao}y uMOp ssvd ‘puvy-o1q"%} oY} JO JOO; ON} 
: GP008 | 89° FES | 8ags 89°SZ | 48 AOC} pvor oy} Sulavoy—zp ‘ON duvj—xawuD ONTUdg 
<q “U10}}0 OY} 0} Spuddsop AT[VUOISBDD0 pu 
bd ‘goujd SIq} 3B pPUVT-9[qv, OY} 0} SOSTA pBVOL OY], “ssvis 
a euy pus poom jo Ayuotd sr o10FT *0998Tq OY} JO SHAOZ 
A VB 9LL | OO'TIS Sepa OLTL | 94} JO wonounf oy} oAoge ojIm & YNOqY—aNAOUH ANVN| EZ » 
a ‘SpUVISI OY} WO POOMUO}}OO [[VUIS pu 
Ay SMOT[IM  ‘PoYVSIIII 9q p[nod ‘opt sorIu oory} ‘ured oy} 
= PI'GOL | OSGI | FSIS | FELL | OH “peor poos ‘u103,0q Joao] oY} SUCTY—T[P ‘ON aNVO 
OC LEL | 98°IST aren OP'SL |" sutads [ooo ‘osavy, ‘oug Be YUBq OAT OY} IVON—DNIUdG | ZZ “290 
“Ay ‘a : ; I 098 
‘opnyzisu0'T ‘epngyryey Samet itaa Bey he bree mene ‘SHUVWAY GNV SLNIOg INININOL : d 
mory eyo, | rosy pejog,| 384°C -10} UT TLV 
> ' ; 
al *PANUHUID —AANUNOLP CAUVMAWOFT 


208 


APPENDIX A.—-TABLE OF DISTANCES. 


160 6E oL6 
u98 GS oL6 


wT 8 086 


u9T LT OF 
ul¥ J&L oOP 


uP 88 oOF 


g¢090T 
TS"8P0L 
89°8d01 


16101 


669001 


66166 


G6°886 
G6°916 


GP'S96 
LV L¥G 
Gv 16 


60 'FE6 


y9'TT6 


T7106 


“ILF6F 


L9°G8P 
v8 COP 


L0°6FP 


80° LPP 


80°96P 


80°86P 
80°TLP 


89°L6E 
69° 188 
89'GLE 


81898 


08°StE 


LG°S88 


09°S& 


eecoese 


86°9% 


eoovece 


09°93 


@eoeee 


OF'66 


T9°GE 


98°T? 


vO'GL 
68°61 
LL’ST 


66°L 


00°9T 


00°¢ 


00°61 
0g'sT 


6ST 
¢0'9 
OP'L 


8E°SS 


86 OT 


G6'ST 


‘qU90SB Suo] & aoUNY, ‘“doop yoox 47 Aq opt 4ooF 
O9L £08738 MOT YB TUveTZG ‘AIYUNOD Sulljoy—anIg HIG 
‘SSVI PUG ‘109BM 
‘ong ‘ong sIq oy} jo AIv{nqII[—HONVUG ATI, NAY, 
‘punois way m0 
qnq ‘AT[IY pue SuUIpUIA Teg Sey peor OY] —AGaUH ATMUAY, 
"yoo1 
AdYIN, JO YOuvIG YW “poo YIIA ‘SoTLUT OATOM4 phe 
“STITT OY} WO SSVI “Pood 109VM PUB JONT—AGNVG HIG 
‘ravATINS ATIVO Ul SuTduIVd TOJ 19zBA 
pues poom {UOSBoOS SIT} 9B TOyVM 9[}}TT—NAAUD MOA-XO 
*yo010 Jo peq Arp wo ‘poom 
0} SOTIM Jey VB pu ANOZ 99U0q} ‘soTTUL OM} FY—UATAD 
‘SSBIS Poos 
pue ‘s901, Sur104v0g ‘“yooIQ UIny, Jo youriq—wagup 
‘punoj 018 yse pu ‘suI[s ‘syvoO YOIFM 
Suo0je ‘sorjns £q pozdnait09zUl W99J0 VAOLH YVO WoOAT oyNo.I 
ey, ‘“oovd siyy ye ont of}9I]T save] OMM—AATUD Nay, 
“Apuvg Aq wor sojiu U9} 10y puvydn oy} wo 
199q BUIAVY ‘IOATI 94} SOYLIYS LOY pvVoOt OY T—AAOUH AVO 
‘syuBq 
94} wodn sosuidul T0AII oY} OTOYM ‘puNnoIs «AIYSIT 
8Y} 0} sostx A][VUOISVOD0 pvod YT —AITAD AGNVG AUG 
"-surof a0 JSOM-YINOS OY} WOIZ YOURIG W—LIVE] NOON 
sreseessooee TQS pUDOSep pus ‘AT[NG BV SSOIQ—HATG WILL] 
"S901} POOM 
-109}09 Sur10}}800 ‘sfood ur vy} pus ‘109BM 97991] £19 A 
‘nT PMT Jo Areqnqiy YW—AaTUO AIL OML-ALUINY, 
‘do048 JOU STII] ‘olaread Suyejnpun uv uo 
OPIAIp OY} SSOIQ ‘oMOJSpuBs puUv pUS Jo sHNTq w39My 
-0q ‘pury-9[qv} oY} 0} Suissed ‘pue ‘u10}30q oY} ssor0e 
‘A][10}SBO-YINOS SI OSANOD OY, *0998[q IY} OAVO'T—ALLVIG 
‘sopoIyae [njosn AUvu pus ‘sq}0]9 
‘Se11900I5 YIM Sr90T[oAvI7 ATddns [[TM 01038 s.AeTINS oT, 
‘IOATY 9}}8[q OY} WOay TIM & Jo s1djzrenb ser1yy ynoqge 
‘ur1vid [oA] OY} UI ‘puvs]T puvaryH oyisoddQ—aANuvay Laog 


T 


lg 


0€ 


66 


8G 


9G 


"AON 


7? 


2? 


99 


9? 


‘~O 


"TANOSSTI[ 
ey} jo syuonye ‘swmvT}S |[VUIS [VIOADS Sosso10 puV 
Spusdosop UY} pUB “YAOF OY} WOT, SOTTUL U9S4ZINOJ UTTTYTAL 
100 VP ob6 wVI 1G 068 | 60°T6GL | Go°969 | PL°Ss PL°SS | TIE ¢ OPTATD », 94} WO SPU POL OU] HLYOMNAAVEAT CHO) 9 99 
‘SIOATY LINOSSTP PUB SYSURY JO SJUONPFe MooAyoq oSpra 
Ge°S9GI | [9°c09 19°GE 68 FI | SUrprArp oy} UO ‘snonuls pvoy—AAaH) TONMANTdaaN] 
’ ‘ISpPL OT} 
OF'SSZL | Z9°L89 agrees 9F'S woi1y Surpusosep uosy} puv ‘odojs Suoy wv Suristy—uaGUO 
O00'8PSL | 9L°G8g ee 99'P ‘ALULOIA UL SSVIS PUY IOQUIT} OUY SVT ‘ YJNOS suny—wAaAUD 
FS'ShZL | OGLLG eat 09'L ceeeus tesneeeee seeeeuaee caeeserse coeeeeereermereeeTINOg SUNY—MTIUO | @ 99 
‘pvor poos $ punois 
PL'GEZL | 06'699 | LGGL | LG'6L | Wer oF ‘Arvgnqra, oy} WoTZ “pusosY OM\—HONVUG VHVWEN | F >» 
‘SOTOUBIG U9IMIOG . 
LE'9IZL | g8'0g9 | ST'9S | 80ST | Surssed por oy} ‘ VyBUION OY} OFUT “YJMOS sUNY—xAAAUD 
‘deep soyout 7 Aq opin 
CO'FOSL | Ga'se9 | “Tt | GOS | 9003 FOT ‘eyRMON Stgq oy} 07 Areynqry pesoddug—nonvug 
"‘pVOt OY} WOA, OOULISTP JAOYS B 4B 
‘puvy LOY} WO TOQUIT, YIM “Quonbow suvedjs Jo spvoy 
ey} Sutmvy ‘esptr oy} uo sdooy yous}, MOU OT} | UOT[IUT 
-I9A OY} IVOM ST “YJLOM OY} 0} “pvVoL Pl[O OY T—NOMWONOL] § >, 
“One 
FO'OSIT | OZ FIG | sgTs 6F'6L | Siq WoTZ O[TU v J[VY YoOuvs [[VUMS G SSO1)—NOTTINUTA | G “AON 


TABLE OF DISTANCES. 


APPENDIX A. 


‘opnyts : fi Pb ona cleat eae ‘SUUVNAY INV SUNIOG INAININON 
PuzLowoT Opngye'T oye'T 189 410.0 8 £6q 9yerpour ( 
WLOL [BIO Y, | WLOAT [VIOT, ‘ -10} UT, 


292 


‘PINnUuyuUuojy—AENUNOL CUVMAWOTT 


APPENDIX A.—DISTANCES TO FORT HALL. 293 


DISTANCES measured by the Odometer, from the North Line of the 
City of the Great Salt Lake to Cantonment Loring, in Oregon. 


Wee Terman NOM TOWN. Sasecavise siscnuans caeneeecs eaiecwecninawelnccem ante 37.05 
Weden’s Pass, opposrte the Pass imtorsesass stirrer ane crates 43.68 
Maple Creek, tributary to Bright Creek. ......5.ccscdcdecesccese acces 53.29 
Mater Aer CeO est siacac ted cdlcacddaaesbdaeal eeeeade la ctees aneaee 61.19 
Sol LSI OIL Aa ene OD ROLE PORE Ce Cer RE OR RE ee RRM ap TAmR: pat 64.97 
Hot Springs—Fresh, Cold Springs 25 feet off...........c.cesceeeeees 71.96 
Ree NEM AT EVO athe atthe vs ccc cla gees Seed seneesgnetal eee sesens tees 79.10 
Bhs She aR, BR OES ied oy Sit eee he ORE PRE ORE eee oe 91.96 
Second Spring—4 fine springs, 4 mile further..................eseees ied 
PePRMMEPUEHTeACIL.  gees ease aetet less bh ansddad/ueeennashs ceqarene a atedaes sesek teas 103.39 
EEE EIU acc etttens castes vc klys Ube ese teeten sue cet me sec ee ce ee toe neem 107.43 
Peete Gee en io abet ete an Anesaes shen ban oes sista tc ars ass See ee R 111.40 
Pape CMia ett Otenaccatsevencdraddsdsedendaceeacndtace savtodhtebaescecueacende 114.33 
MERIT Gr CaN: 5 ,oratest tenance dassse att cares seth hase a Tah Es Meese ake T1822 
Hedspeth’s Cut-off—road from Bear River ...........cscesee coeceeees 125.46 
imac of Pass, in Hedspeth’s Ont-offrssciccoscdisssstdsnesi dha eee 129.84 
eran or ce cer Vist lexciercets conse dese Wssacad vase tagcat eit caese ee ee Re 132.07 
PeMIB Ae OU OM OL ASAI See toot ite Shab Rae daatlads onee tds deekte attest 134.78 
Foot of divide of Malade and Pannack RiversS...............000seee0 140.42 
ere OME ANTUMCIS 12 Jassrsnceshstsacelscees ahsoks casdee sasek dese ase eats 145.07 
Branch of Pannack—after 3 small branches have been crossed. 155.59 
Mtr ACLU LP INITAU CIEL SS scree salseckcatwcsee encase ctoate scents as cndenesee cea tes 159.12 © 
“0. UDACITENT IILI2S: yO) a SE Se oe ee Re ALA Se SS SARE BALL 164.09 
Erie Me Ge AMD AK Aegis Hiss hay Ldacomencatnenssesassee cesta bs aces eh L797 
PEE CEMNGCUIE IN TVOT sasese same sleds celeacet eee Corey utes Were ane cee oe deta 177.00 
UN GISTET GLE S800) 20 ete ce Baa aan Ce ned eS SceE elec core iaacic ssneee SCBdeione 179.81 


Cantonment Loring, 5 miles beyond Fort Hall............s:.ssseee 187.75 
ee EE NE eR UR Eee ee ee ee eT 


a a ee ee 


994 APPENDIX A.—DISTANCES TO SAN PETE. 


DISTANCES FROM THE STATE-HovseE, City of the Great Salt Lake, 
to the Settlement in the San Pete Valley, on a branch of Sevier 
River. Communicated from the San Pete Company’s Journal. 


PROMINENT POINTS AND CAMPING DISTANCES. 


To Kanyon Creek Bridge 
‘¢ Mill Creek 


bmp) 


Bishop Crosly’s 
Second Cotton-wood Creek 
Forks of Road 
Dry Creek 
Willow Creek 
PAGS PUI oon cde sa suesiticchi pv asses unecealesieeeeciis «'sauemsuneees 
Summit of Hill at Jordan Kanyon 
Dry (Utah) Creek 
American Spring 
American Creek 
Marsh Creek 
Cedar Grove 
Timpanogas River 
Spring Base of the Mountains 
Spring Creek 
Pimquan Creek 
Pequinnetty (Spanish Fork) Creek...........scssceee ceseeseee 
Clear Creek 
Petenete Creek 
Aph “é 
Wa-ka-tiky (Summit Creek) 
Pungun—Ghost Spring 
Warm Spring Creek 
Watage «¢ At 33 miles take left-hand road 
Ona-pah (Salt) Creek at kanyon 
Kanyon Forks 
Third Crossing of Ona-pah 
Summit leaving Ona-pah Valley 
Pleasant Creek, in San Pete Valley 
Springs in Slough 
San Pete Creek 
San Pete Crossing 
Timpa Creek 
Mouna Creek 
“* City Creek, 12 feet wide 


AN OIweH DDN NHWY wee cwbhhd 


BORON BIR BOB) BoR)] Bos obojeibp 


bS GO Co “Inpmww Qe 


HB[CO HEB [02 BB [00 BJ BJ 


HR HS|COHS|COHS|COHS|00 Ont 


bo CO ee & O17 HO 


en 
re) 


Notr.—The San Pete Settlement was began in 1849, and in 1850 a city was laid 
out by the Presidency. The distances measured by an Odometer, attached to 
a wagon. This is on the road to Iron City, in the Little Salt Lake Valley, and 
the southern road to California. 


APPENDIX B. 


LATITUDES AND LONGITUDES OF THE PRINCIPAL 
TRIANGULATION STATIONS; 


TABULATION OF THE TRIANGLES DEVELOPED IN THE SURVEY 
OF THE GREAT SALT LAKE; 


AND TABLE OF GEOGRAPHICAL POSITIONS. 


NOTE. 


Tue Angles of the Triangulation were measured with a seven-inch theodolite, 
by Draper. The instrument was scarcely competent to the work, from the low 
power of its telescope, the great distance between most of the stations, the mirage, 
and the almost constant haze that pervaded the atmosphere. Many repetitions 
of the readings were consequently necessary to secure the requisite accuracy in 
the results. This was peculiarly the case in obtaining the Azimuth of the Base 
Line, which was fixed by observations of the Polar Star. The work is believed 
to be sufficiently accurate to correct the detailed measurements, as well as to form 
a basis upon which a triangulation may hereafter be extended over this great in- 
ternal basin, should such a work ever be contemplated by the Government. The 
natural features of this desolate region, abounding as it does in lofty eminences, 
widely separated by intervening level plains, is admirably adapted to such an 
operation; although its execution could not but be attended with great labour and 
privation: Many of the deserts would furnish extended plains, absolutely level, 
upon which a degree of the Meridian could be measured to great advantage. 


APPENDIX 


B. 


GREAT SALT LAKE VALLEY. 


Latitudes and Longitudes of the Principal Triangulation Points 


— 


co CO “I lor) Or He © bo 


fax ek est 
bp -§ oOo 


tt 
Or pe CO 


Lape 
© CO ~I1 > 


and Stations. 


LOCALITIES. 


Near Adobe Hall, in the city. Long. 


from Frémont. 
East end of base-line.........seceseees coceee 
West end of base lime...........sceccseseeeee 
Constitution Hill, north of city........... 
Rose Spring, west side of Great Salt 
Lake valley. 
On mound near mouth of Dry Cotton- 
wood. 
On the table east side of Jordan Kanyon. 
North-east shore of Utah Lake............ 
Near Timpanogos Kanyon, east side of 
Utah Lake. 
West side of mountain, at Jordan Kan- 


yon. 

Young’s Peak, highest Point on Ante- 
lope Island. 

Black Rock; south shore Great Salt 
Lake: salt-works. 

Pacrinaton Uslamd 24.5.0 .ihe.dectesesenteeses 

Bremont Ueland 22. jhe iceoenctde adhere sssoandees 

Strong’s Knob, west side of Great Salt 
Lake. 

ECOMOMLORY LP OLMb, awexnccnastxcaeadie vaste gaa 

Gunnison Esland. vst... sssaessecceenadeated 

Holphine slaw. 35. seta. .cdichedeades os Gevwed 

Horned Frog, on east side of north- 
west branch of Lake. 

Saturday, on mound, north-east side of 
Spring Bay. 

Lighthouse, north end of Lake, west 
shore. 

Mud Island, near mouth of Bear River. 

Dome on Stansbury’s Island......... ...+.. 

Head Peak, north of Bear River Bay... 


Latitude. 


40° 46’ 04” 


40° 53’ 15.6 
40° 51’ 28” 

40° 48’ 03”.6 
40° 38’ 30”.9 


40° 34’ 24’.3 
40° 26’ 51”.8 
40° 21’ 24”.9 
40° 19' 47.9 
40° 26’ 49”.7 
40° 57’ 20".5 
40° 43’ 08” 

41° 00’ 06".6 
41° 10’ 04”.6 
41° 14’ 28”.2 
41° 11’ 54.3 
41° 20’ 23.9 
41° 27’ 38".7 
41° 27’ 55".5 
41° 33’ 06”.2 
41° 42’ 22.6 
41° 14’ 538”.8 


40° 51’ 18” 
41° 35’ 46.6 


Longitude. 


112° 06’ 08” 


TE2° TL g”.8 
112° 17’ 45”.8 
112° 06’ 51” 
112° 16’ 00” 


112° 00? Fa" .7 
112° 06’ 40.5 
112° 04’ 38”.1 
111° 53’ 87”.9 
112° 09’ 05”.2 
112° 25’ 20”.3 
112° 26’ 02”.4) - 
112° 46’ 26” 

112° 32’ 47” 

113° 05’ 08’.8 
112° 88’ 02” 

118° 03’ 23.8 


118° 11’ 52’.5 
112° 54’ 44".7 


.112° 59’ 42”.4 


118° 02’ 29”.7 


112° 26’ 29.8 
112° 42’ 81.8 
142° 27' 077.8 


297 


APPENDIX B.—TABULATION OF TRIANGLES. 


298 


G6G0ET8'0 
GISPBPL'T 
0€8¢166'0 


€999228'0 
66908180 
L096998'0 


66616860 
GP9I9LE6'O 
69996650 


LIST¥96'°0 
GGPSG6L'T 
66616860 


OL9PZET'T 
SPPO96L'T 
LIETPhS6'0 


ZOSTZIO'L 
OS9ZSLI'T 
OL9PEST'L 


EISI8LL°0 


EL9T8LL°0 
GF6LES8'0 
GOSTSIO'T 


*‘SUIYy Ie dO'T 


81Z09'9 
9PLOF'T 
P9086 


‘6601'S 


81609°9 
PGES 'L 


GPhE69'8 
IL¥98°8 
GGGOL'G 


01166'8 
OF9L9°ST 
GhE69'8 


€99G°ET 
GSOL'ST 
L166°8 


0F83'0T 
SIL6'FL 
6999°ET * 


0000°9 


0000°9 
TSS0'L 
O0F83 OL 


“SOTTAL 


*HLONDT 


G'ISh ‘Fs 
Sle'rL 
9°Z9L‘T9 


1°660‘LL 
q’1sse ‘hs 
P'SIL‘88 


P'106 ‘CF 
9°S08 ‘OF 
1°660‘LL 


8°L09°LF 
S'TLL‘Z8 
F106 ‘SP 


089‘TL 
F'8Z6‘Z8 
8°L09°LP 
C662 FS 
F'820‘6L 
089‘TL 
089‘T¢ 
089‘Ts 
G'0Z9‘L8 
G'662 FG 


“7907 


Hit 
Oy Fy Fy 


ee 


ole 
ro 


BAR 


AAO AA 


AAA 


one raAgca 


OOM 
ads 


‘aytsoddo optg 


‘SU'IDNVIU, LO 


ela) 
CO Fy ey 


P9G80L9'6 


—P88S9F6'6 


GO8L682L°6 


GI8&ZPP'6 
PE8LEE6'G 
6916786'6 


FE0G166'6 
1919666'6 
GLOOPLE’6 


GELPELI'S 
LO8GPI6'6 
LEEG899'6 


PISTSE6'6 
66928666 
TL9899L°6 


SLISPG8'6 
IZ9L86'°6 
I8G8PhP6'6 


8610G91'6 
16996686 
1810686°6 


Priolg | 


wl iGO ofS Cone ee eeeees eee oe sesees *reveeeeeeceees sooppSouvdulLy, ‘T 
182 99 oft Tl Pe eeee tenses eeeece eee" T90ID kiq jo 4900 IBON H 7) 
00 69 oL8 Oe eee ee tee eeseee Ceres sees seers uofusyy qe70 qsunT “7 
68 0 o9L oeeees covcccces Dials Ket hig jo 497900 IVON i 
W8L/99 083 000000 Ce oe ee Ser seeeee ceseee "+ TOLUV YY qviy 480M 1) 
180 69 oF OL eee ee eee coceeescs seeeeeees eT OAUV YT yvin q4SUny a 
VARIN o8L ee ee eeecceece seeeoeore TO KUB YT yin 489 MM ") 
HOLL oL8 eeeeocves want ae Cem atone << SOE RT, yqvin qSBny 7 
ude 10V ofl -PPCHCHHES CHF H EEE FOE E Ee SEE EEEE ES POOM-0}40F) kag OL 
WAS /L0 o8G #0000 Coe ores coeseener veveeceee nokuey yin qSUnT 7 | 
$40 9F of ZI see ee oees eeeseeese veeeeeoes seeeeeeeeNOOM-109900) hig aC f 
9',,00/90 oL% oie cpt none gyraensyne gages ese es reo a OT TC QSO% ‘ad 
HET BE COS [hehe epee ce HON TOMO, kag 
uSGuLP oG8 #08 Pee Coe oem Coe ees oee eeeeee evens prevesees SUTIOG as0% 
uo 109 ov$ eee seeeee seeeeeesee eeeeeeses eee ee LLEEL MOIJNITISMO/D 
9G WAS olP #00008 Serre eress ceeoeeoes ee eorrescescs sutidg 9804] 
VLOG o9L SCOPE CHEE EEEE CHEE EEE SHEE EOE “TILE MOT}NZLYSUOD 
uwLh8 8h ol9 "OOOOH COE OOHE OOH EHEEHE DOO HO OHS OOP BOL EEE Oe OSV 489M 
eeecee eececccee hit asi O00 e eres eoevecoes (SEPT oseg 
P61 OF off oe eee eee coven esee vesseenes le) (|!) | M01} N}1}Su0N 
BT 16S oGP 0000s erro Ooes BOO EEE oes COO e EE EES BHO HEOEes es esVg 489M. 
P'u6P iP oZOL OOOO Poe core eeees Cera eees Cees sees Coeeed "OSU {SAT 
‘suy *NOILVILG 
NOILVTOGV, 


209 


APPENDIX B.—TABULATION OF TRIANGLES. 


8FP0906'T 
6660168°0 
PLLESOE 'T 


686999E'T 
8LO6LIS'T 
SLPOLSP'T 


OZPOPLE T 
SLO6LIS'T 
8PF090G'T 


LIOGEZP'T 
O6ILETS'T 
OZFOFLZ'L 


OL9PGEL'T 
P8889FG'T 
O6ILELS'L 


OL9FGET'T 
LEGT98S'L 
SS8lPIE'T 


9TPETOL'T 
GLOLPSL'T 
0&68691'0 


T0LG616'0 
9TP6LOLT 
$809996'°0 


O€8E166'0 
6891668'0 
T0LG6L6°0 


OITLO'9T 
LLOSL*L 
01666 06 


LG0G'°&3 
T9TG9T 
6198S 


0G6L°8T 
T9TS'9L 
TLL0°9T 


SL6P 9G 
GLO OT 
0S6L1°81 


G9GE°6T 
9Sg9'LT 
GLSS "OL 


6999°ST 
G9GE'61 
GOSL'ES 


OI61S FL 
0€6L6 FL 
8PPS9'S 


SPI8E'6 
OI619°FT 
80L70°6 


F9E08'6 
816F9'9 
SP7188"°6 


F'S8‘F8 
F'Z80 ‘IF 
9°608‘90T 


929 ‘C21 
G00‘L8 
GPS LST 


9°L83 66 
0°S02‘L8 
F998 $8 


1°906‘68T 
9°198‘98 
9°83 ‘66 


6°8h0‘ZOT 
GIGS 86 
9°198‘98 


089‘TL 
6'8F0‘ZOL 
6GL ‘SSL 


8°099‘9L 
L'$68‘SL 
9°S48 6% 


$89 6P 
8°099‘9L 
G'89L°LF 
9°Z9L‘TS 
9° LOL‘Gs 
re9‘6P 


‘kW 
TPUW TM 
‘T POW A 


BUOIYO'T IVD 
‘TT “IBD WA 
‘Sug TW 


"PI “280 “A 
‘PIAL “280 
‘PI ‘34 “A 


TWAT 180 
Wa A “A 
PI “80 °A 

OF 

PD YOU “AA 

“A YT AA 


HHA 
ASO 


Sodds =S4 
BHR AHH 


Be 
eo 


9€99698°6 
LIPOPSS'6 
69696966 


OSTS806°6 
6868092'6 
¥896666'6 


GO9TPL6'6 
6960816°6 
68919066 


GTL1966'6 
0889982'6 
8LIOLP8'6 


GGCS0L6°6 
GO9GLE6'6 
8060868"6 


LEL8LOL’S 
¥99GTC6'6 
$869666'6 


OF0LP66'6 
6697186'6 
vS9TS8o"6 


LOGSI8L°6 
886986'6 
SG1EL6L'°6 


$966086'6 
LI8&G18"6 
T¥88196°6 


68 LV oL¥ 
uLG 00 olG 


«VG iLL of LT 


uG0 90 ofS 
uLG GL oG& 
186 LP 006 


u1G G3 o0L 
uG& 1&9 oGG 
uVS OV o6G 


u8¥ GE oL6 
wOV SP oL€ 
uG& OV ob V 


uVV AW 069 
uVG YE o8G 
u6G GT 064 


wT GG oG& 


G’,0§ GS o9G 
GGL GE oL8 


u&L V0 o18 
WOL BI o9h 
WL iL 066 


«G0 OF o8& 


186 /LG o$0T 


uO08 GL oL& 


uw OT ofL 
uVG 8G oOF 
uL¥ 0G 099 


COO CCE e CHEE FEDS ES HEH OE HEE OH EEEH TEE ES puvist pry 
eee nesses eeeese eeeeee See eeeeee ee MOL §,oun0 
eee eee ee C8 0068 FoR 8 ee HOF O8 rocereee METS | JUOUII AY 


000000 000008 050088 088008 SHOE 88 seeree* NUBIST JUOWOL 
000 coccce cocces Sees enseisna tees sees 5G Otay §,8U01}9 
ee UC pa G DUD oa Dou 15 1) (i §,W0}.SULLIVD 


000008 0000080 C000 F8 090000008 080080808 *pueysy JUOUIA 
Pee eee cee cence les cecces cece eeeessccss yVog §,0Un0 X 
eG a SC ORLI O afs| 8 ,W0JSULLIVD 


BOSOOOIOOUIOOOISOCIOUOIOOOOOOROOOOOOOOCIOOOOIOT yeog §,.oun0X 
00 coccece Teen's eis See eae > TE TST §,WOJSULIIV) 
COC le Pee eee COSH OS OOOH HEROS BETH OOEOe SHO8E8 yoy yovlg 


00000 000000 cree ee Deve e eels secnne ge aveiekss <OONT YoVvlg 
oo cae cee eee Cees ee ees serene ees weeeeeeee yeog §,.oUnN0 X 
0 eee e odes Coo Ceo eee COs eeeeeD RC he) | MOTJNYI}SMO/ 


000000 2800S CEEe 000808088 009890000 posse SC Od §,0Uno X 
odes ce ceceee coccce cosceees Baicelnesihlvo eT asoy 
COC eo0es coeeeseee CHR0C8008 ee eee UWOTJNFIYSUO/) 


©0000 090000000 06900,0000 699008 80880008 *qUlog ensuoy, 
00008 OOO F2L2Ee BOOTS HEe COELHO 2EOS coe VaLOg: oy OL 
000000 Cees 0eees Fosse cees vesoerere Bence ress sesouvduty, 


000008 090080808 Coe 20808 COOL LEOe See ee PULO 3) LON Bi 
Cdeees cvevcncen scsencee vvcgosciee ssvis'cianes'* SUELOET su0'yT 
©0008 F808 F8 000408 Ce28S8 COOH CHEE 8e **** su DOUvAUIL], 


eee cose coer eeee ee OT UT qin uo quTog Suo'Ty 
000 veces ore es Fences eee es Dee eee eee “'spsouvduly, 
eeecee seeeee vee ts steer TOO a kag jo Wi) RA @) IBIN 


‘OL 


Ob 


OR 


ae 


300 


1850. 


TABLE OF GEOGRAPHICAL PoSsITIONS. 


LocALitIEs. 


Adobe Hall, Great Salt Lake City. The 
longitude from Col. Frémont. 

Junction of Forks of Golden Pass Creek; 
near point of union of old and new 
roads. 

Dividing ridge, Wahsatch chain.......... 

Parley’s Park; spring east of Bauch- 
min’s Creek. 

Weher. Rivers apper ford, ....ssacnsnnansians 

Junction of Red Fork and Weber River. 

Nprme in Bed Kanyors 2....... deophes. os 

Red Kanyon at Chicken-cock Bluff; 
junction of Echo Creek. 

At Yellow Creek, tributary of Bear 
River. 

Sulphur Creek: one mile east of Bear 
River Ford, and north of Tar Springs. 

Copperas Spring, in Great Coal Basin... 

Hr BRU Br ss ss olan see adeelasi one ieengeineutistn tio wee 

Black’s Fork, near Muddy Creek........ 

Near Spring on the plain, in the bend 
of Black’s Fork. 

East bank of Green River........0..0+-s00 

Bitter! Creche an 2. WENGE Re ons Bekah 8 

Bitter Creek, at the Sandstone Bluffs... 

Bitter Creek Prairies 45. s0se» addopis Boon sis 

Bitter Creek; head spring branches.... 

Bridger’s Fork of Little Snake River... 

On the plains of Muddy.........0¢ j0.s.0006 

On the west bank of Muddy Fork, a 
branch of Little Snake River. 

North bank of the Muddy, near Vas- 
ques’ battle-ground. 

In the Kanyon of Muddy Fork............ 

On a rivulet near Bridger’s Pass......... 

Branch of Sage Creek... 4...1.< 0s sidens 

On North Platte. The longitude taken 
from Col. Frémont’s maps, 1848. 

On the plain west of Medicine-bow 
Mountain. , 

Rattlesnake Creek, near Medicine-bow 
Butte. 

Laramie Plaing,....escess tvs. o0st SP Ra RE 

Laramie Plamgs. ..adisgge 253 «xs. sitePejae a oinnite 

Feappe’s ‘Greele,. 2. acs cevdee tosddvesotussocewe 

On route across Laramie Plains.......... 

West Fork of Laramie River.............. 

2.72 miles west of Laramie River; Main 
Fork. 

In Black Hills; on a branch of Crow 
Creek. 


Latitude. 


40° 46’ 08”.3 


40° 45’ 40”.5 


40° 44’ 48” 
40° 43’ 04.8 


40° 53’ 41.3 
40° 57’ 41”.07 
41° 02’ 27".26 
41° 06’ 26”.1 


41° 09’ 00”.2 
41° 08’ 08.18 


41° 13’ 45”.7 
41° 18' 12”.8 
41° 28’ 56.38 
41° 31’ 08”.5 


41° 80’ 44.5 
41° 34’ 40.8 
41° 30'51".4 
41° 28’ 09’.2 
41° 24’ 12” 
41° 28’ 39” 
41° 28’ 14”.1 
41° 27’ 06”.18 


41° 28’ 28”.49 


41° 27’ 41” 
41° 33’ 227.3 
41° 35’ 21” 
41° 32’ 49”.2 


41° 37'15”.8 
41° 40’ 45”.2 


41° 41’ 50”.9 
41° 38’ 38.6 
41° 33’ 06” 
41° 28’ 16” 
41° 19! 43.4 
41° 15’ 41.4 


41° 08’ 02” 


APPENDIX B.—TABLE OF GEOGRAPHICAL POSITIONS. 


Longitude 
west of 
Greenwich. 


112° 06’ 08” 
111° 53’ 14” 


111° 46’ 55”.5 
111° 38' 46”.2 


111° 36) 267 
111° 40’ 46” 
111° 30’ 34” 
111° 22’ 06” 


111° 14’ 138” 
111° OF 22" 


110° 48’ 00” 
110° 32’ 23” 
110° 18’ 50” 
110° 04’ 15” 


109° 88’ 40” 
109° 23/09” 
108° 50’ 34” 
108° 41’ 09” 
108° 31’ 00” 
108° 14’ 24” 
108° 02’ 11” 
107° 52’ 41” 


107° 41’ 21” 


107° 38’ 48” 
107° 30’ 48” 
107° 21’ 52” 
107° 06’ 11” 


106° 53’ 31” 
106° 43’ 87” 


106° 36’ 57” 
106° 24’ 47” 
106° 15’ 58” 
106° 06’ 35” 
105° 57’ 12” 
105° 45’ 39” 


105° 24’ 11” 


GEOGRAPHICAL Positions—OContinued. 


APPENDIX B.—TABLE OF GEOGRAPHICAL POSITIONS. 


301 


Longitude 
1850. LOocALITIES. Latitude. west of 
Greenwich. 
No. 37.| Crow Creek, one and a half miles below | 41° 09’08”.5 | 105° 08’ 24” 
the kanyon. 
«¢ 38 Branch of Lodge Pole Creek, in Che-| 41° 21’45”.7 | 105° 12’ 21” 
yenne Pass. 
«¢ 39 | Branch of Chug-water River.............. “41° 45’ 35”.8 | 104° 59’ 25” 
«¢ 40 | Chug-water River; Bridger’s ravine.... | 41° 89’ 58”.4 | 104° 56’ 45” 
Soe | Ort, LATAIMIIC, .a...00000.0ceeied nade ¥. chants 42°012!.38".2..1.. 104° 314.26" 
s¢ 42 | North Platte River; left bank, below | 41° 55’ 36”.7 | 103° 58’ 23” 
mouth of Horse Creek, 
SCONE S EUG. cc .c.csc ccesccece soncvaase saneve 41° 48’ 25”.7 | 103° 45’ 02” 
fn a4. lett, bank of, Platte River..ic..asscncceee 41° 45’ 39” 103° 21’ 44” 
ao.) ett bank of Platte River........-s.00ces 41° 33/22” 102° 45’ 10” 
‘¢ 46 | Ash Hollow; one mile from river........ ALP 17S. yes, 4029 1021.28” 
REIMER HON IK CONTA 8 Soin neiteninsasephEhlvenecacaa 40° 38’ 45” 98°08) 4” 
‘© 48 | Oak Grove; Little Blue River............. 40° 13’ 41.2 97° 54’ 36” 
«¢ 49 | Road at leaving Little Blue River........ 40° 11’ 16” 97° 39’ 02” 
«« 50| Fort Leavenworth, according to the| 39° 21’ 14” 94° 44’ 00” 


determinations of Major Emory and 
M. Nicollet. 


REMARKS. 


The longitudes given in the table depend upon those assumed 
for Fort Leavenworth, and a point in Great Salt Lake Valley, 
taken from Nicollet and Frémont. Their means for fixing these 
points were so superior to ours that they are necessarily adopted ; 
our sextants being imperfect, and the mode of transporting chrono- 
meters, of the kind furnished, interfering with their regularity. 
On leaving Fort Leavenworth, the two chronometers differed but 
8.5 seconds; they were put into leather boxes, carefully adjusted 
upon two mounted men, and at the end of the first six miles differed 
11 minutes and 11 seconds. From subsequent observations the 
loss was attributed to No. 1961. They were then carried in a 
spring-wagon with the other instruments. The average daily differ- 
ence for the entire route was a little above one second. ‘The re- 
lative rate was, for sixteen days, a losing one of half a second; 
then, to Laramie, one and a half second’s gain. Thence they were 
carried in another light wagon, and the rate was about one second. 
Whatever important changes one received, was therefore partaken 
by the other. 


302 APPENDIX B.—GEOGRAPHICAL POSITION. 


Allowing the entire change of the first day to be due to No. 1961, 


and taking the longitudes of the termini as stated above, 


The average rate of gain for chronometer No. 1681, daily, is............ses0e 8.031 
Average rate at beginning and end of Jourmey..........sesssee csessecee socsee seeees 7.010 
At Fort Leavenworth... 2.0 caisceusunavewavectveseve'ceasaussescveunwaepieansecuneeenne 10.000 
At Great Salt Lake Citys... icc.c. sececaet tvacnctle duces ai castes sietuah edbuan ese eeeeeeeeeee 4.020 
hom. s@ 
Time by sextant observations at Great Salt Lake City.............cssceeeees 7 49 43 
ce -y q Fort Leavenworth... .-asscss.00s50-a-cerenee 6 28 31 
Difference Chronometer: HME, «2.2... 60/0eaveren voparnees/aposwsews guncenne nines seneeeet 1 21 12 
2 in mean time by assumed longitudes............. sesssesseceeeeeees 109 27,2 
CHTOROMELMIC Pale 222. 0s. ccosadacalncscsc cal edeeesans slsonsaswicesahusee Sibeesuseeaamne 11 54.9 
Time by sextant observations at Fort Laramie ... ...........sssscssceseseesens 7 17 29 


The number of days between the respective observations, was, 
from Leavenworth to Laramie, fifty days; thence to Great Salt 
Lake City, thirty-nine days. The proportion of gain is therefore 
6 min. 41.6 sec., and 5 min. 13.3 sec. Making these corrections, 
and converting into siderial time, the longitude of Laramie will be 
given at 105° 19’ 50”. 

But taking the rate at the termination of the journey, and ap- 
plying it to the time given at Laramie, we have, 


Longitude, Great Salt Lakes). csc. 1.j.cscsjesesesdeenovseescs ene 112° 06’ 08” 
as BGC Se oc oa ove saamncnnincannananeesomacee eee 104° 40’ 35” 


This would agree better with that of Fremont in 1842; but he 
observes in his book of 1843, that the longitudes of that year are 
thrown too far west collectively, and proposes to correct, at Fon- 
taine gut Bouit: taking the amount of correction given at that 
station in 1845, at 15’ 49’’, his observations place Laramie in 104° 
31’ 54’’, which is nearly that adopted in the accompanying map, 
resulting from measured distances and observations, on the home- 
ward journey. Relying upon the accuracy of the map of 1845, 
we have checked the work made up from course and distance, 
measured by an odometer, at the crossing of the North Fork of the 
Platte, and reduced the longitude of Camp 32 on Chugwater, at 
104° 56’ 45’... The chronometric difference of this camp and 
Laramie is taken, on account of the good apparent work of the 
time-keepers and the winding of the road, at 25’ 19/’, which gives 
the longitude of Fort Laramie at 104° 31’ 26’. Thence to Fort 
Leavenworth the intermediate points where latitudes were taken, 
we make up from course and distance as before. 

The lunars taken at Laramie and in the Salt Lake Valley, are 


ee a 


roe 


APPENDIX B.—GEOGRAPHICAL POSITION. 303 


disregarded, the instruments used being too imperfect for such 
work. On the return journey the chronometers were carried in 
panniers on the most gentle mules. Their comparison exhibits 
‘jumps and stops’’ so great as to forbid this mode of conveyance. 
Between Salt Lake City and Bridger’s Fort, the loss of 1631 upon 
its companion was 52, and 29.5 sec. on two respective days; on the 
last day’s travel 1961 stopped entirely, and at the arrival at Fort 
Laramie, 1631 was 23 minutes in advance of the time of starting. 

There can be no doubt that light balanced pocket chronometers 
are best suited to this method of determining difference of lon- 
gitude. They can be put on spiral springs with suitable packing, 
in boxes, and strapped so as to ride horizontally on the body of the 
mounted man. The chronometer should be put into its place after 
the carrier has mounted, (if on horseback,) and taken out by the 
astronomer at the halting, before alighting. But what is of great 
importance, the travelling rate of the instrument should be found 
previous to commencing the journey. This could be done by 
having it transported, in the manner intended on the route, the 
average daily number of hours, at the place of fitting out. With 
three well-tested chronometers, much confidence might be placed 
in their work, and if either one varied the others would detect it. 
Three persons should carry them. 

A remarkable fact is shown by those chronometers in our jour- 
ney. Their rates were given at Philadelphia. At Fort Leaven- 
worth they both had the same relative rate, but had increased 
from less than one to ten seconds gain, having been transported 
in stages over the mountains. After being some months stationary, 
in the winter at the Salt Lake, they together returned to nearly the 
rates at Philadelphia. It would appear, therefore, that such de- 
rangements do not obtain and affect these instruments merely while 
moving, but that it is gradually recovered from when returned to 
permanent rest. 


J. W. GUNNISON, Lieut. Top. Hng’s. 


In charge of Astronomical Department. 


Capt. H. Stanszury, Top. Hing’s. 
Commanding Expedition to Great Salt Lake. 


AV, eT Ph ey ee wh ed 7 yy SA we . path 


Pine) \ CORE DL” TM SERA ae ee ARTE 


Pe! ead) davhos cat ‘ pened Teac’ pbtralent tlt ide ue 
. + ie 
wi , ively dat Penge grey Sy ete ew yee eee Y Cea ‘Perr eee CAC ie 


; i ur ty) by 
{ f s ‘ : : 7 ; : VE 
levireet 4iff E fag 
i 
Vi fee TAM? BEE 9 Pettey * aro Be 


Powaiol . ax ee Te 
2) i ‘ ¥ eer f ’ a i! : ’ . i : ; , I a 
INGE... tho. noted beh soglee- adh te eiyot Ay wad 


: P : =% hou 
. ‘ ‘ hy gti ls oa e € i , ses ~ one ' r 
bee ii rot ab ORR Os oe DS Wa nm) t } ait eve Dod 


= ‘ fe. # ’ ° 


Jo geste Pisor eros Le} Ae vs 


hy bent DORR ee AMER EIS h.8 “el ary Pe a 
WTS s > : : ee 


° ; oe ; a a oe cone a pre. 
Bscindy teh ile f hee did Lo TE te aed: Ces a) fies 
tien TORT HAO AEA TeT eS PITA SR POT Oe: 


vib dA eA Oe de el ORTH in eCn ne a 
sik YPruion BE Lbthortey Fr otod ede AMT hee eee 
eral) Hi: _- PR weebdaarkt \aaoce? 7 tr DD (ia hiya 
ne bhar se ay rene BY HOT iNT ond) paren inbeter Powe a 
ey arrest Gone iy Fee ease om 6b peeerts © RP ae ate 


teal aN en a ai te 
“oeraagry gg! Samtpsed vee, Sipagtnts iis At 

i EAN PIR BS Nepean site — PSE a 

pethn:cte Sein te ee muRE we af yale aba Soe as 5 a ak xt 
ci Konto iat Oa 
a Pe Cy Oa PERT mae ta tq 


oe Re: spiaicihek gyn» a, 
Wi {ng Vana EE Ramm? ROM Datars 


4 + OR 1 j Way . se at i 
LG See. ae YR ae Lai Fo wae a ator bal ik 


APPENDIX C. 


ZOOLOGY. 


- QUADRUPEDS AND BIRDS, BY PROF. SPENCER F. BAIRD. 
| REPTILES, BY PROF. BAIRD AND CHAS. GIRARD. 
_ ss INSECTS, BY PROF. HALDEMAN. 


APPENDIX C. 


ZOOLOGY. 


It is much to be regretted that the circumstances of the Salt 
Lake Expedition were such as to prevent as much being done in 
the way of collections in Natural History as the accomplished 
head, Captain Stansbury, and his assistant, Lieutenant Gunnison, 
had intended. Called upon to start almost at a day’s notice, 
they found it utterly impracticable to obtain the proper preserva- 
tive materials, apparatus, and other necessary outfit for making 
collections, in the limited time allotted to them. Nevertheless, it 
will not be a matter of surprise to those who are acquainted 
with the gentlemen concerned, to learn how much was actually 
accomplished, as will be shown by the published results. In fact, 
no Government expedition, since the days of Major Long’s visit to 
the Missouri, has ever presented such important additions to Natu- 
ral History. Of the great advancement of geographical know- 
ledge, the reports of the officers will speak for themselves. 

The mammals observed, as might be supposed, belong mainly to 
the Rocky Mountain series. The most interesting fact in their 
history is the determination of the existence in Utah of the great- 
tailed fox, now for the first time described, although mentioned 
by various travellers. 

The birds brought in by the expedition, belong chiefly to the 
waders and swimmers. ‘The number is not sufficient to draw any 
general conclusion as to the ornithological fauna of Salt Lake 
Valley; although the indications are that this forms a meeting 
point for the species of the Saskatchewan, the Pacific, the Mis- 
souri, and of New Mexico. 

A South American duck was obtained for the second time in 
North America, (Pterocyanea rafflesii,) the single specimen previ- 
ously found in the country having been shot in Louisiana. A new 
bluebird, (Sialia macroptera,) appears to be abundant. An ex: 


ceedingly interesting fact is found in the determination of the 
307 


308 - APPENDIX C.—INTRODUCTION. 


winter quarters and range of Lewcosticte tephrocotis. The only 
specimen previously seen of this bird was obtained by Dr. Rich- 
ardson on the Saskatchewan. 

The reptiles procured are all new, excepting two: of these, one, 
Holbrookia maculata, was recently described by Mr. Girard, from 
the Platte, and the other, Phrynosoma douglasit, from Oregon, by 
Bell. None of the new species have ever been found elsewhere. 
Among these is the finest species of Cnemidophorus in North 
America. ‘To Mr. Girard, equally with ourselves, is due the cre- 
dit of the article on the reptiles, especially for the monograph of 
Phrynosoma, which is entirely from his pen. 

Many specimens of insects were procured, but unfortunately 
injured or lost on the return. The few preserved have been ably 
determined by Professor Haldeman, as will be seen by his report. 
The principal entomological result is the precise determination of 
the destructive grasshopper, which, but for the interposition of a 
species of tern, at one period was near turning the “Garden of 
the Mountains” into a desert. | 

Of shells and other invertebrate forms, no specimens were 
brought in, excepting in too fragmentary a state to admit of de- 
termination. 

By the kind permission of Colonel J. J. Abert, Chief of the 
Topographical Bureau, so well known for his liberality and love of 
science, we have been permitted to examine a small but exceed- 
ingly interesting collection of birds and mammals, procured by his 
son, Lieutenant J. W. Abert, in New Mexico. Among them we 
found a new species of bird and one new mammal, descriptions of 
which, with lists of the other species, we have subjoined. We have 
also ventured to include, in the article on reptiles, descriptions of 
some new species from Oregon, Texas, and New Mexico, collected 
by officers of the army. With the cheerful acquiescence of Cap- 
tain Stansbury, we have likewise appended to the article on birds 
a complete list of all the trans-Mississippi1 species not imcluded in 
Audubon’s American Ornithology, adding the few found since his 
time east of this great natural boundary. For assistance in cor- 
recting and extending this list, we are under great obligations to 
Mr. John Cassin, of Philadelphia, who is now engaged in publishing 
a continuation of Audubon’s Ornithology, to include all the species 
mentioned in it. 


S. F. Barrp. 


APPENDIX ©.—MAMMALS. 309 


MAMMALS. 


BY 8. F. BAIRD. 


1. Vunpres MAcRouRuUS, Baird.—Great-tailed Fox. 


WE greatly regret that the specimens of this, the most import- 
ant addition to our fauna made by Captain Stansbury, should be 
of such character as not to admit of a description based upon the 
skull. None were found by the party, all brought in being cased 
skins purchased of hunters in Salt Lake Valley, and, as usual, 
without the skull. 

In general appearance, this species resembles the red fox, 
Vulpes fulvus, in its different varieties. From these, however, 
it may be at once distinguished by the great length of the tail, 
which exceeds that of the latter species by six inches, and more. 
In the best specimen procured, the back is of mixed grizzled 
gray colour as in the gray fox or badger, the hairs being dark 
brown at the base, then yellowish white, and finally, tipped with 
black. These hairs are interspersed in a very abundant soft fur, 
of uniform colour, varying in tint with the region of body. A de- 
cided black colour prevails on the muzzle, sides of face, top of head, 
and upper part of neck, separated from that of the back by a rather 
uniform ferruginous in front of the shoulder. On the shoulders, 
the gray exhibits a slight tendency to a cross, and widens poste- 
riorly, including the outside of thighs, and extending a short dis- 
tance on the tail. There isa considerable admixture of black 
around the eyes. The sides of the neck and body, concavity of 
the ear, basal anterior portion of the convexity, and space across 
lower neck, are light ferruginous; the remaining portion of the 
convexity of the ear black; chin, throat, legs, belly, top, and sides 
of the tail, black. The tip of the tail is dirty white; beneath, 
with region about the arms and posterior edge of thighs, light fer- 
ruginous; whiskers, black. 

Length (approximate) from snout to base of tail 33 inches. 
Pail to end Of YerteDra.....ccoprocearser versonseoseens ee ares 


Ge Meee UO OU AUNT Se snrests' gis tiuiiev sale sonign atseamalvesnae aa. 
Breadth of tail, Hatten” fo.ccses ccvssveacvtaevesteccn o pee 


310 APPENDIX C.—MAMMALS. 


Hairs on sides-of batlic. 2, vsiciscusdaweanseastenee kansas 43 inches. 
C6 SERED) A sens cotWeetvin deans dine seals aeete eb au eee A yee 

WP OVEREDA). oixsnnensicanpancpainsesknivceapsapevne snip pasneeecen 10 48 

PIG Ol BANG cseac asks cisssasccineewasnce scenauss mepdawnss ose 


Another specimen, much like the last from Fort Laramie, has 
the top of the head and posterior half of the back grizzled yellow, 
gray, and black, anterior half of back and across shoulders nearly 
to elbow with a much greater predominance of black; basal half 
of the convexity of the ears and the entire concavity, with the 
edges, sides of neck and of body behind the forelegs, light-yel- 
lowish. Legs and beneath black, as also muzzle and ears, with 
the exception just stated. The fur is very full and soft, and the 
feet densely clothed with long, crimped, soft hair. 

Another specimen indicates quite a different variety, with a 
much closer resemblance to the red fox. The colour above is 
light ferruginous, deeper toward the dorsal line; beneath, white. 
The hairs at the base are, as usual, lead colour. The fur, how- 
ever, along and toward the dorsal line is terminated for the 
greater part of its length by a rich chestnut, rather darker behind. 
The long scattered hairs on the back are mostly black, tipped 
with light yellow; laterally, this fur is light ferruginous, fading off 
into white toward the belly. This ferruginous is more distinct 
immediately over the back. Inner sides of legs, sides of head, and 
concavity of ears, likewise yellowish white. Upper part of muz- 
zle, around the eye, and on top of head, grizzled chestnut, like 
the back. Convexity of posterior surface of ear, black. The 
sides of the soles also indicate black, although the legs are too 
much mutilated to show distinctly the colour. ‘The general colour 
of the tail is yellowish white, deeper above: the long hairs of top 
and sides tipped with black. ‘Tip of the tail, white. The feet in 
all the varieties are densely covered with hair on the under surface. 

In this species we find all the varieties of the common red fox, 
Vulpes fulvus, as the chestnut, the black, the silver gray, cross, 
&c. How far its range extends we are at present unable to state. 
It probably, however, reaches the Pacific coast, and far to the 
north in the Rocky Mountains. Indeed, we consider it very im- 
probable that the ordinary red fox extends west of the Missouri. 
As regards the eastern range, we have seen specimens from Fort 
Laramie, and Audubon and Bachman refer to a skin from Fort 
Union, at the mouth of the Yellowstone, which may possibly be- 
long to the same species. _ 


APPENDIX C.—MAMMALS. 311 


2. Putorivus vison, Lin.—Mink. 


Putorius vison, Dekay, N. Y. Zool. pl. 1, p. 837;—Aud. & Bach. Quadrupeds, I. 250, 
pl. 33. 


The well-known and destructive mink appears to be common in 
the valley of Salt Lake, several specimens having been procured. 
The colour is more uniform than common in Eastern specimens, 
there being no trace of the yellow spot on the chin. In one indi- 
vidual, however, the edge of the lower lips is white. 


3. PUTORIUS ERMINEA, Lin.—Ermine. 


Putorius noveboracensis, Dekay, N. Y. Zool. pl. 1, p. 36. 
Putorius erminea, Aud. & Bach. II. 36, pl. 59. 


This species occurs abundantly throughout the northern and 
temperate parts of the whole northern hemisphere. 


4, MELES LABRADORIA, Sabine.—Badger. 


Meles labradoria, Sabine. Captain Franklin’s Narrative, p. 649;—Richardson, F. 
B. A. L. p. 87 ;—Aud. & Bach. Quad. I. 360, pl. 47. 


The American badger is found in the interior of North Ame- 
Tica, especially in the regions bordering on the Rocky Mountain 
ranges. ! 


5. GuLo Luscus, L.—Wolverene. 


Orsus luscus, L. Syst. Nat. 
Gulo luscus, Rich. F. B. A. I., 41;—Aud. & Bach. Quad. I. 202, pl. 26. 


The wolverene, known also as carcajou and glutton, is an inha- 
bitant of the arctic regions of the northern hemisphere, extending 
as far north as lat. 75. In North America, it is an inhabitant of 
the whole of the British and Russian possessions. It is found 
sparingly in Maine, Massachusetts, and Northern New York, 
although exceedingly rare. Farther west it is more abundant, 
particularly along the upper Missouri and the Rocky Mountain 
ranges. The locality here assigned, of Salt Lake, is the most 
southern limit yet given. 

To the traveller and trapper on. the prairies or among the 
mountains, the wolverene is the greatest nuisance imaginable. 


312 APPENDIX C.—MAMMALS. 


It ferrets out the caches of provisions and skins, and devours 
their contents greedily, its enormous strength being such as to 
enable it to remove almost every weight which may be placed upon 
the articles concealed. It destroys the traps set for other ani- 
mals, and tears to pieces their contents. Indeed, in the Northern 
United States, this animal is dreaded more than the panther or 
bear, being invested with fabulous attributes of ferocity and dan- 
ger. This is, to a much less degree, the case in the Rocky Moun- 
tains ; but everywhere the wolverene is attacked with caution. 


6. FIBER zIBETHICcUS, L.—Muskrat. 
Fiber zibethicus, Aud. & Bach. Quad. I. 108, pl. 13. 


The muskrat abounds over the greater part of the American 
continent, extending from the Atlantic to the Pacific. Its north- 
ern and southern limits are not well ascertained: those assigned 
to it by Audubon and Bachman are lat. 69° to lat. 30°. 


7. SPERMOPHILUS 13-LINEATUS, Mitchill. 


Sciurus 18-lineatus, Mitchill’s Medical Repository for 1821. 
Spermophilus hoodi, Sabine. 
S. tridecem-lineatus, Aud. & Bach. I. 224, pl. 39. 


The little prairie squirrel, so common in Wisconsin, Michigan, | 
and Minnesota, would seem to have a very extensive range, in being 
found by the expedition on the Platte beyond Fort Laramie. 
The specimen is, however, immature, and it is quite possible that 
further investigation may show this to be a species distinct from 
the St. Peter’s specimen described by Dr. Mitchill. 


8. Ovis MonTANA, Desm.—Bighorn. 
- Ovis montana. Rich. F. B. A. I. p. 271;—Aud. & Bach. Quad. II. 164, pl. 73. 


The specimen of bighorn, or Rocky Mountain Sheep, brought 
home by Captain Stansbury, was shot on Chug-Water. It is the 
largest individual we have ever seen, although itself possibly not 
of maximum size. It differs somewhat from the description in 
Audubon and Bachman, in having the posterior line of all the 
legs yellowish white, this colour extending to the axillee in front, 
and confluent behind with the white of buttocks, scrotum, and 


APPENDIX C.—MAMMALS. ‘. pL 


thighs. The dorsal line is inconspicuous, except on the darker 
tips of the short mane. 
Circumference of horn at the base...... 17 inches. 


Length of horn along the convexity.....364 
Distance between the tips of horns.....18  ‘ 


The bighorn, at one time erroneously supposed to be the same 
as the old-world argali, is common in the ranges and hills be- 
longing to the Rocky Mountain system. 


COLLECTED BY LIEUTENANT ABERT. 


1. PSEUDOSTOMA CASTANOPS, Baird. 


This beautiful species was collected by Lieutenant Abert along 
the prairie road to Bent’s Fort. In general colour it is of a pale 
yellowish-brown, with an ample patch of light chestnut on the side 
of the head and face, deepest above. The dorsal line is not darker 
than the rest of the fur. In size it is intermediate between P. 
borealis and bursarius. 

The colour of the fur above is slightly grizzled, and much lighter 
than in P. bursarius ; beneath, paler. Throat, space between the 
forelegs and sides of arms, pale rusty. The chestnut marking, 
on the side of the head, is very strongly defined, occupying on 
each side a nearly circular space of about one and three-quarter 
inches in diameter, with the nearly obsolete ear as a centre. 
These chestnut spaces do not quite meet on the crown and occiput, 
but leave a rectilinear interval, coloured like the rest of the back, 
of about one-eighth of an inch in width. On the muzzle, however, 
from above the eyes, the colour of opposite sides is confluent. 

The hind feet and toes are thinly covered with whitish hairs, 
which on the fore feet appear more ferruginous. The claws are 
white, but sufficiently transparent to allow the coagulated blood in 
the phalanges to show through them. 


Length to base of tail (approximate).,.........00+ 8 inches. 
AM rec cbe acetates ajscnbaps weapasieay scsesascs sou ducerdecrnss oe) 66 
Hand (along the palm)..........csccocssecseesseveeceere 1% . 
Length of exposed part of middle anterior claw. 3 “ 
Hind feet (along sole) from heel...........s.0.essees 1” “ 


Middle: ClawWirnsarccgeesassonanavertorsauariercenensy such ooeee’ 42 «6 


314 APPENDIX C.—BIRDS. 


BIRDS. 


By SPENCER F. BAIRD. 


1. BuTEO BOREALIS, Bp.—Red-tailed Hawk. 
Faleo borealis, Wils. VI. 72, pl. 52.—Aud. Biog. I. 265, pl. 51. 


Salt Lake. Found by Gambel in California. 


2. ACCIPITER FUScUS, Bp.—Sharp-shinned Hawk. 
Falco velox, Wils. VII. 110, pl. 45, fig. 1 (young female). 


Faico pennsylvanicus, Wils. VI. 18, pl. 16, fig. 1 (adult male). 
Falco fuscus, Aud. Biog. IV. 522, pl. 374. 


Salt Lake. 


3. ATHENE HYPUGAEA, Cassin.—Burrowing Owl. 
Striz hypugea, Bp. ‘Am. Orn. I. 12;°p)., Fs 


Strix cunicularia, Aud. Biog. V. 264, pl. 482, f. 2. 
Athene socialis, Gambel, Pr. A. N. S. III. 47. 


Abundant in the valley of Salt Lake and on the plains east 
and west of the mountains. 


4, SIALIA MACROPTERA, Baird.—Long-winged Bluebird. 


‘Marz.—Salt Lake City, March 18, 1850. 


“A specimen of Sialéa was procured by Captain Stansbury, 
which, at first sight, was referred to S. arctica of Swainson. 
On comparing it with others from Fort Union, the differences 
were found to be sufficiently great to constitute a distinct species. 
The Fort Union specimen was clearly referable to S. arctica of | 
Swainson shot at Fort Franklin, Great Bear Lake, so that it be- 
comes necessary to impose a new name upon the one from Utah. 
Not having at hand specimens from the Pacific coast, it is im- 
possible to say to which species the birds described by Audubon 
as S. arctica belong, (probably arctica.) 

The principal difference between the two allied species is to be 
found in the longer wings, and much smaller and weaker claws, 
with rather longer toes, of the S. macroptera, as will be sufficiently 


APPENDIX (.—BIRDS. 315 


evident from the table of comparative measurements. ‘The bill, 
too, of the latter, although of much the same proportions, is de- 
cidedly smaller. 

As in S. arctica, the upper parts are a bright azure blue, more 
lustrous and deeper on the wings, rump, and tail-coverts. The 
lower parts are of a light greenish blue, excepting the lower belly 
and sides, abdomen, and tail-coverts, which are white. 

The white of the lower parts is clearer and extends higher up 
on the belly than in S. arctica, and the quills and tail-feathers 
are much bluer, this colour greatly predominating over the brown 
on the inner webs and inner faces of the feathers. The clove- 
brown shows somewhat conspicuously as a broad tip to the quills, 
which, besides, are very narrowly margined, terminally and inter- 
nally, with whitish. The outer margins of the primaries, towards 
the tips, shade into greenish blue. Owing, perhaps, to the plum- 
age not being quite full, many of the feathers of the back and 
breast have grayish tips. 


/ DIMENSIONS. 

Sialia macroptera, (Salt Lake, Male.) S. arctica, (Fort Union, Male.) ! 

ONE. 5500. fos ice csiewedcasoscsieves 4 inches. (Skin contracted) 6% inches. 
EN aera sg Sv nclasisha viaise siended ged ees 143, “ ‘ 
SOIC ct. cduaaesssnacosineesos Aig « ct 444 « 
Tail to insertion of middle feathers 3%, “ aN Pape 
FE og osacnvens vecenvcsncssese a (Tail worn) 4g 
Projection of longest primary be- 

yond longest secondary or ter- 

tiary, (wing not shut close)....... Lye “ 

Do. wing shut close..........css.c0000 3 acme sueenatadeesee ideseusece lair te 
Longest primary beyond shortest, 

oD =a ee soidvcnan ance ue nautls Tee at. sGeidacblasaweuccschesacaneis dh tame’ 
Second primary longer than first, 

MEUEIOUS oe acl cadc sree LS ngs tebe atte 3 2, canhenmUte EP are rman Mabe widaiaeses 28g « 
RRO A ad. 5 adecseyocecdven, | ae Sl. daiguadawka coduedenctt Betieene SA “ 
ice oho tsis wall diainn vdidense’s GO, BL BCRP CEPR ESELe Eee Mrs Ras Pe ae 
asin v saitavin Laswaden oiies signe SB Ul Fon ered ontuids Se oeuae yeaa 10K“ 
EMCO 5 nl cascesesevesccsscsssccs eda oe elaswaseas sasetcatidencenshgdi 14 

BN WAGI, CLAW, oc ascassececesens Pa as aciatas dassie cibtcianis vee Sel an 1,074 “6 
Lateral toes, (equal)..............0000 BRT aa cicdty oltideaidieseismatmnteateitene 44%“ 

Bemeee. Vwith Claw. ..cc.escccesse Big! a MM an dae « halen nade eaene rare tz “ 
BRUNE OG oie 50.'¢ sesso rccesdsacestanss coed API PL loeecasedldar eetmareaetat satis vz “ 

Rte oe WHEN CLARY fo ciesssasenisadissess Genes hier spies saulemnmameat stents’ tf “ 


(In S. macroptera.) 
1st quill spurious; 3d, longest; 2d, little shorter; 4th, rather less than 2d, 


816 APPENDIX C.—BIRDS. 


5. PARUS SEPTENTRIONALIS, Harris.-—Black-head Titmouse. 
Parus septentrionalis, Harris. Proceed. Acad. Nat. Sc. Phil. II. 300 (Dec. 1845). 


A single individual of this rare species was procured by Cap- 
tain Stansbury. This bird was first described by Edward Harris, 
Esq., from a specimen shot on the Yellow Stone River in July, 
1843, and is the largest of the American species of true Black-cap 
Titmice, three in number. It is not improbable that two species 
may be confounded under the name of septentrionalis, as this spe- 
cimen is quite different from one collected by R. H. Kern, in 
New Mexico. The latter is, however, too much mutilated and 
faded to serve as a proper standard of comparison, for whick we 
must wait to get better specimens, 


6. STURNELLA NEGLECTA, Audubon.—Western Lark. 
Sturnella neglecta, Aud. Biog. 2d ed. VII. 340 (1848). 


The distinctions between the old Sturnella ludoviciana and the 
present species are quite obscure. A specimen from Fort Union, 
presented to us by Mr. Audubon, agrees with the published cha- 
racters in nothing but the bands on the middle tail-feathers, which 
replace the scolloping seen in S. ludoviciana. ‘The tail is quite as 
much rounded, and the bill of the same size. The Salt Lake bird 
has the tail more square, and the bands on the middle tail-feathers 
still more distinct than in the one from Fort Union. ‘The size is 
fully as large as that of the common species. The specimen was 
shot March 18, 1850, in the cafions between Salt Lake City and 
the Hot Springs. This lark utters a single rough note like that 
of the European starling. 


LGA sserewes oss stsedetoanescas antoveeudevicaaaddapeece 104 inches. 
EEO, Sancicwan < ohdcdeedsrviveeconls asqucnswanvalasdadtadees 165 


7. NipHa@a orEGoNA, Audubon.—Oregon Snowbird. 
Fringilla oregona, Towns. Journ. Acad. Nat. Sc. Phil. VIT. 188 (1837).—Aud. 
Biog. V. 68, pl. 298, fig. 3, 4. 
Fringilla hudsonica, Var. Licht. Abh. Ac. Wiss. Berl. for 1838, 424. 
Fringilla nortonensis, Gm. I. 922, 87. 


This interesting species, so similar to WV. hyemalis, or common 
snowbird of the Atlantic’ region, replaces it in the Pacific. It 


iw: 


APPENDIX C.—BIRDS. 317 


occurs abundantly in Oregon and California, as well as in New 
Mexico and Utah. 


Length of specimen shot March 21st........ a dead 52 inches. 
FORECMIG Sanath inne a gtbee fea edad sect acdc sceis sib cosebwacd Get, 


8. PrucHA LINCOLNII, Audubon.—Lincoln’s Finch. 


Fringilla lincolnii, Aud. Biog. II. 539 pl. 198 (18384). 
Peucea lincolnti, Aud. Syn. 118 (1839). 


A specimen of this bird was shot at Salt Lake, March 21, 1850. 
The species was first described from individuals killed in Maine, 
since which it has been found in very small number—more abun- 
dantly about Carlisle, Pa., than anywhere else. The Salt Lake 
Specimen agrees with one brought from the upper Missouri by Mr. 
Audubon, in having a more grayish tinge than that usually seen 
in individuals from the Atlantic coast. The black marks on the 
dorsal feathers are also larger and more decided. The bill, too, 
appears a little smaller. These differences, however, are hardly 
specific. 

Length of Salt Lake specimen.............s000 esssseees 52 inches. 
PPMECMG Sei tecd pte de ctrveusaiel svacneddelecdssuwdeben st obwaced Bip FE 


9. LEUCOSTICTE TEPHROCOTIS, Swainson.—Gray-crowned Finch. 


Linaria (Leucosticte) tephrocotis, Sw. Fauna Bor. Amer. II. 265, pl. 50 (1831). 
Lrythrospiza tephrocotis, Aud. Synopsis, 125.—Nuttall’s Manual, 2d. ed. I. 632. 
Fringilla tephrocotis, Aud. Biog. V. 2382, pl. 424, fig. 3. 


Leucosticte tephrocotis, Bp. & Schl. Monog. des Loxiens, pl. 42.—Gray’s Genera 
Avium, 536. 


This exceedingly interesting bird was first described by Swain- 
son and Richardson, from a specimen procured by the latter, 
May, 1827, on the Saskatchewan River, in lat. 54°. But a single 
individual was obtained, which was subsequently presented to the 
Museum of the Zoological Society of London. From this, all the 
published descriptions have been made, even that by Mr. Audubon, 
who was unable himself to procure a specimen. For the sake, 
therefore, of multiplying comparisons, we shall present an original 
description taken from the bird brought home by Captain Stans- 
bury. This was procured on the 21st of March, 1850, in Salt 
Lake City. 

Matz. General colour of back, scapulars, hind neck, belly, 


318 APPENDIX (0.—BIRDS. 


breast, and indeed of the entire foreparts, (excepting the crown,) 
of a dull chestnut-brown, darkest on the chin, throat, and cheeks. 
Feathers of the back and breast with light margins. Upper part 
and sides of the head, including the lores, lower eyelid, and occi- — 
put, (excluding ear-coverts,) ash gray, lighter behind. A patch 
of dull black on the crown and forehead. Lesser wing-coverts, 
rump, under and upper tail-coverts, sides of body, abdomen, and 
thighs, having the feathers tipped with a beautiful purplish pink, 
giving to the rump the appearance of transverse bands. Tail- 
feathers dark-brown, narrowly margined externally with rose; 
inner secondaries and primary coverts more broadly with dull 
white, all the quills being faintly tipped with brown. Nostrils 
covered by a tuft of whitish recumbent feathers; a similar tuft at 
the side of the mouth. Bill, feet, and wings as described by 
Richardson. | 


DGGE vce sescey adouvesk Guest ates sa seueuce ines ener bals ieee 74 inches. 

ECECU i sone teow dels naw eareecmun ewes ae noe tee edeis's weusars 12 ie i 
Pleated. . tuscan cheetas is ins aaallsbebehy ann. penne 4h k: 
EUs i aE em octane eee oe eee elie ee eee ys a 
BD SE reeecie eau eaweisinw es ewe penile we ainteeme hey ate near ee 


This species comes nearest to Leucosticte griseinucha of Brandt, 
1842; L. griseigenys, Gould, an inhabitant of Russian America 
and the Aleutian Islands. This latter species may, however, be at 
once distinguished by the possession of gray cheeks and ear 
coverts. 


10. OrTocoRIS occIDENTALIS, McCall. 
Otocoris occidentalis, McCall, Pr. A. N. S. Phil. V. 118 (June, 1850). 


This species of sky lark is founded by Colonel McCall upon 
an immature bird, shot near Santa Féin July. Captain Stans- 
_bury’s specimen was killed near Salt Lake City, March 18, 1850, 
und is, consequently, an adult, in winter plumage. It differs from 
winter specimens of Otocoris alpestris, in having no yellow on the 
throat and superciliary stripe, more black on the cheeks, and 
Jess on the breast, and a very slight ferruginous tinge of the upper 
parts and sides of the body. The white across the forehead is 
more distinct. The bill is shorter, more slender, and more curved. 


ce 


pa erga lio Mel OB 2 Sn Og 


APPENDIX C.—BIRDS. 319 


From Alauda flava of Audubon it differs in the larger size, 
and in having the middle tail-feathers like the upper coverts, in- 
stead of being black. 


11. Prous Torquatus, Wils.—Lewis’s Woodpecker. 


Picus torquatus, Wils. III. 31, pl. 20, fig. 3 (1811).—Aud. Biog. V. 176, pl. 416, 
fig. 7,8. 


This beautiful woodpecker belongs to the Pacific fauna, not hav- 
ing been observed east of the mountains. It occurs abundantly 
along the western coast. 


12. TeTRAO UROPHASIANUS, Bp.—Cock of the Plains, or Prairie 
Cock; Sage Cock. 


Tetrao urophasianus, Bp. Zool. Journal, III. 214 (1827).—Aud. Biog. IV. 503, 
pl. 87; Syn. 205. 


A single specimen of this magnificent bird was shot near the 
mouth of Bear River, on the eastern side of the Lake, May 8th. 
It is found on the plains skirting the Rocky Mountains, seldom 
coming down to the Missouri, except far to the north. It is not 
yet recorded as being found on the coast of California, although 
abundant along the Columbia River. Its flesh is not usually con- 
sidered edible, from feeding so much upon the artemisia or sage. 


SELL Eke atnixarais ons stated aris cnciarisiistasiicte sins ei depsincaieaieteoerdemece 28 inches. 
IP LOM ras «welees anas'stee des sawalcecaecgeslsewatoins deea setecaces So ee 


13. CHARADRIUS vocIFERUS, L.—Killdeer. 
Charadius vociferus, L. 258, 3;—Aud. Biog. IV. 191, pl. 225. 


Common across the continent. 


14. GRUS CANADENSIS, Temm.—Brown Crane. 


Ardea canadensis, L. Syst. Nat. 234, 3. 
Grus canadensis, Aud. Biog. III. 441, pl. 61. 


The brown cranes were found during fall and winter in immense 
flocks in the marshes along Salt Lake. They presented their 
usual watchfulness and difficulty of approach. No white ones 


320 APPENDIX C.—BIRDS. 


were seen. Occurs in large flocks throughout the whole interior 
of North America. 
Length of, Pomigl gd ..cascseap racine icctanpen eecdeeaeeee eens 45 inches. 
EEX HO tt 55s vis nuan: «cdbesikn wane aan Mena teee aout Mabe eae teen Tho 
Legs, bill, and feet, black; eyes, orange. Male, 41 by 69. 


15. BoTauRuS LENTIGINOSUS, Montagu.—Bittern. 


Ardea lentiginosa, Mont. Orn. Dict. Suppl. 18138.—Rich. F. B. A. II. 874.—Nutt. 
Man. II. 60;—Aud. Syn. 263. 


Ardea minor, Wils. Am. Orn. VIII. 35, pl. 65, fig. 2 (1814).—Aud. Biog. IV. 296, 
pl. 337. 


This bird appears to be a great wanderer. Although an exceed- 
ingly rare visitant in Europe, the species was first described from 
a specimen shot in Ireland. It occurs throughout the United 
States, West Indies, California, and the fur countries up to lat. 58°. 


16. NUMENIUS LoNGIROSTRIS, Wils.—Long-billed Curlew. 


Numenius longirostris, Wils. Am. Orn. VIII. 28, pl. 64, fig. 4 (1824).—Aud. 
Biog. 240, pl. 231. 


A specimen was shot on Antelope Island. The species occurs 
abundantly throughout the interior of this country, along the Mis- 
souri, and on the prairies. Common also in New Mexico and Cali- 
fornia. 


17. SYMPHEMIA SEMIPALMATA, Hart.—Willet. 


' Scolopax semipalmata, Gm. 1. 659, 83.—Wils. Am. Orn. VIII. 27, pl. 56. 

Totanus semipalmatus, Rich. F. B. A. II. 388, pl. 67.—Aud. Biog. IV. 510, pl. 274. 
Catoptrophorus semipalmatus, Bp. Syn. 323 (1828). 

Symphemia atlantica, Raf. Jour. de Phys. vol. 88, p. 417 (1819). 


Shot on Salt Lake. Common both on the Atlantic and Pacific 
coasts. 
18. RECURVIROSTRA AMERICANA, Gm.—Avoset. 
Recurvirostra americana, Gm. 698, 2.—Aud. Biog. IV. 168, pl. 318. 


Salt Lake, March, 1850. Not noticed on the coast of the Pa- 
cific. 


ee 
———- ts 


APPENDIX (C.—BIRDS. ool 


19. Cyanus AMERICANUS, Sharpless.—Swan. 


Cygnus americanus, Sharpless, Silliman’s Journal XXII. 83 (1831).—Aud. 
Biog. V. 133, pl. 411; Synopsis, 274. 
Cygnus bewicki, Richardson, Fauna Bor. Am. II. No. 224. 


Of two specimens shot March 10, 1850, in Jordan River, one 
was in full plumage. The other, a male, not quite mature, had 
the space in front of the eyes and the flattened space at the base 
of the bill above, covered with scattered, minute feathers. The 
orange spot is indicated through the feathers. The bill is black 
at the tip, base, and along the commissure, and of a dull yellow- 
ish in the intermediate area. Feet mottled. The older one mea- 
sured 51 inches in length and 76 in extent. Weight, 15} pounds. 
The young one, 47 by 71; weight, 94 pounds. 

The range of this species is quite extensive. Inhabiting the 
Atlantic in winter, especially Chesapeake Bay, it was found on 
the Saskatchewan, in lat. 64°, by Sir John Richardson, and on 
the Columbia River by Dr. Townsend, thus extending across the 
continent. 


20. ANSER ERYTHROPUS, L.—White-fronted Goose. 


Anas erythropus, L. I. 197, 11. 
Anas albifrons, Gm. 509, 64. 
Anser albifrons, Nutt. Il. 346.—Aud. Biog. III. 568, pl. 280; Syn. 270. 


Jordan River, Salt Lake, in March. Found in California, Ore- 
gon, and east of the Rocky Mountains generally. 


21. ANSER CANADENSIS, Vieill.—Wild Goose. 


Anas canadensis, L. Syst. Nat. I. 198, 14.—Wils. Am. Orn. VIII. 52, pl. 67, fig. 4. 
Anser canadensis, Rich. F. B. A. II. 468.—Nutt. Man. II. 849.—Aud. Biog. III. 6, 
pl. 201; Syn. 270. 


This common bird occurs entirely across the continent, being 
found abundantly in California, Oregon, the Atlantic States, and 
the intermediate country. In summer it goes northward to breed, 
extending almost to the shores of the Arctic seas. This goose 
constitutes the principal summer food of the inhabitants of the fur 
countries, large numbers being salted down for winter use. 


The specimens procured were shot on Salt Lake. 
21 


$28 APPENDIX C.—BIRDS. 


22. ANAS BoscHas, L.—Mallard; Green-head. 


Anas boschas, L. I. 205, 40.—Wils. Am. Orn. VIII. 112.—Nutt. Man. II.—Aud. 
Biog. 164, pl. 221; Syn. p. 276. 


Abundant on Jordan River. Found throughout the United 
States, California, Oregon, and fur countries. 


23. MARECA AMERICANA, Steph.—Bald-pate. 


Anas americana, Gm. Syst. Nat. I. 526, 97.—Wils. Am. Orn. VIII. 86, pl. 69, fig. 4.— 
Aud. Biog. IV. 337, pl. 335. 


Mareca americana, Steph. Shaw. Zool. XII. 185.—Rich. F. B. A. II. 445. 


The bald-pate, so well known in the Chesapeake Bay region, 
for the impudence with which it robs the canvass-back of its 
favourite food, the celery grass, was found in considerable num- 
ber on the Jordan River. It is abundant in California and Ore- 
gon, as well as the country east of the mountains. 


24, QUERQUEDULA CAROLINENSIS, Bp.—Green-wing Teal. 


Anas carolinensis, Gm. I. 533, 103.—Aud. Syn. 28. 
Anas crecca, Wils. VIII. 101, pl. 70, fig. 4. Aud. Biog. III. 218, pl. 228. 


Jordan River, March 26, 1850. This species appears to be 
very abundant about Salt Lake. It is equally common in Califor- 
nia, as well as east of the mountains. | 


25. PTEROCYANEA RAFFLESII, King.—Red-breasted Teal. 


Anas rafilesii, King. Zool. Jour. IV. 87, Suppl. pl. 29 (1828).—Jard. and Selby’s 
Mlust. N. S. pl. 23.—Cassin, J. A. N. 8. IV. 195 (1841). 


“¢ Anas ceruleata, Licht.” 
<¢ Anas cyanopterus, Vieill.” 


This beautiful species is now for the second time presented as 
an inhabitant of North America. In 1849, Dr. Pilate, of Ope- 
lousas, Louisiana, sent a specimen, shot in his vicinity, to the 
Philadelphia Academy of Natural Sciences, which was announced 
by Mr. Cassin as new to our fauna. The species was first founded 
on specimens obtained by King, in the Straits of Magellan. It 
frequents the coast of Chili, whence specimens have been sent by 
Lieutenant J. M. Gilliss. 


APPENDIX C.—BIRDS. 323 


The Red-breasted Teal appears to be a common bird in Utah, 
three having been shot in Jordan River. As the species has never 
been described from a North American specimen, we subjoin the 
following, taken from those brought by the expedition. 

Ma.ze.—Head, neck, anterior part of body, sides, flanks, bar 
across the vent, rich purplish chestnut, deepest and most lustrous 
across the breast, faintest across the vent. Upper part of the head 
to occiput, the chin, and lower tail-coverts, dull-brownish black. 
Lower part of breast, belly, and abdomen, (encircled by the chest- 
nut just mentioned,) obscure brown, and faintly glossed with gray- 
ish and chestnut. Lower wing-coverts, subscapulars, and tips of 
primary coverts, white, showing on the latter as a conspicuous 
white patch. Shoulders, lesser coverts, and greater part of two 
of the longest scapulars on each side, bright blue, (darkest on the 
latter.) Speculum, grass green. Longest scapulars with a cen- 
tral streak of yellowish white, and variously margined and mot- 
tled with chestnut. Back, rump, and upper tail-coverts, margined 
with dull chestnut. Bull black, feet yellow. Bull long, narrow, 
somewhat wider towards the tip; nostrils oval, lamellar, rather 
long. ‘Tarsi short, and feet small. Outer toe, without its claw, 
about equal to the inner with claw. Tail rather long, wedge- 
shaped, of fourteen feathers. 


Total length (skin much stretched).........+sccsscsseessseesees 20 inches. 
Jon) SND D0) 8 cect ocase poe g ose Bocnesons coer nmere ere anonne endbas <eaneroiar Aaa 
set TEL Ig Sep ea aah nes does oaarannadnisowaceitan con ceason Su'cuigsiesmaict 275 “ 
CUDISS To -055 sein dnee Sa bre cag OSEP Es | Sos 0 gdee a cn cosddeoce rece er ede 1,4, “ 
ITO.CHS Gye CRG UTE AR Gannn sceene: © ochore Opes Seprenec ose sprr 6 lit“ 
aor ONAL se retcece aan nicnti's ec viaiiosisloeisaisiawas solos’ ageinaae sia Soy 
1 Ditka coesoits padseden oeaceancaecena So cdd Hes" secu saushoaos HerSach ace inec 3), aie 


FEMALE.—Similar in general pattern of colouration, the chestnut, 
however, replaced by the mottled yellowish and brown character- 
izing the female ducks. This pervades the whole inferior portions, 
not excepting the tail coverts. The black of the head, and the 
blue on the shoulders, (not on the scapulars,) are retained. 


26. DaFILA AcuTA, Bp.—Sprig-tail Duck. 


Anas acuia, L. I. 202, 28.—Aud. Biog. III. 214, pl. — 


Salt Lake. Found across the continent. 


324 APPENDIX C.—BIRDS. 


27. FULIGULA AFFINIS, Eyton.—Little Black-head; Shuffler. 


Fuligula afinis, Eyton, Mon. Anatide (1838). 
Fuligula mariloides, Vig. Zoology of Beechy’s Voyage (1889). 
Fuligula marila, Aud. Biog. IIT. 226, pl. 229. 
Fuligula minor, Giraud, Birds of Long Island, p. 323 (1844). 


Salt Lake, March 21, 1850. Found across the continent; very 
common throughout the interior. 


28. CLANGULA ALBEOLA, Bp.—Butter-ball. 


Anas albeola, L. Syst. Nat. I. 199, 18.—Wils. VIII. 51, pl. 67, f. 2, 3. 
Fuligula albeola, Aud. Biog. IV. 217, pl. 325. 


Provost Fork, February 22, 1850. Occurs from the Adloniie 
to the Pacific. 


29. PELECANUS TRACHYRRHYNCHUS, Lath.—White Pelican. 
Pelecanus Americanus, Aud. Biog. IV. 38, pl. 311; Syn. p. 309. 


The only specimen in the collection is in the form of a skeleton. 
This wants the peculiar vertical lamina of the bill, but in all pro- 
bability belongs to the above species, the female of which is usually 
without this appendage. 

It is mentioned by Gambel as common on the coast of California. 
In winter it is found in the Southern Atlantic and Gulf States, 
and to some distance up the Mississippi yee Exceedingly 
abundant about Salt Lake. 


30. PHALACROCORAX DILOPHUS, Sw.—Cormorant. 


Pelecanus (Carbo) dilophus, Sw. F. B. A. II. 473 (1831). 
Phalacrocoraxz dilophus et floriduus, Aud. 


Salt Lake. 


31. CoLYMBUS GLACIALIS, L.—Loon. 


Colymbus glacialis, L. Syst. Nat. I. 221, 5.—Sw. Faun. Bor. Amer. II. 474.— 
Nutt. Man. II. 578.—Aud. Biog. IV. 48, pl. 306; Syn. 353. 


This species of loon, shot on Salt Lake and brought in by Cap- 
tain Stansbury, enables us to give to it a locality more western 


APPENDIX (C.—BIRDS. 325 


than any yet recorded. It is abundant throughout the United 
States, where the difficulty of shooting it has passed into a proverb. 


BIRDS COLLECTED IN NEW MEXICO BY LIEUT. ABERT. 


1. Fanco spaRveRIvS, L.—Sparrow-hawk. 
2. PIPILO ABERTI, Baird. 


This species at first sight exhibits a strong resemblance to P2- 
pilo fusca from Monterey, from which, however, it differs in many 
characteristic features. ‘The colour above is of a nearly uniform 
rusty-brown, or olive, no material difference in tint being descerni- 
ble on the head and rump, as in P. fusca. Beneath, and on the 
sides of the neck, the colour is much like that of the back, with a 
stronger tinge of ferruginous, however, which becomes very decided 
about the lower tail-coverts. The markings around the bill are 
not very distinct, owing to the mutilated state of the specimen; 
there appears, however, to be a tendency to black on the loral 
feathers; the bristles also are black. The throat seems to be uni- 
form in colour with the neck and breast, and unspotted. The tail 
is uniformly coloured, and is destitute of the light tip of P. fusca. 

The bill is much stouter than in P. fusca, as well as more curved. 
The claws also are much stronger and larger, the tip of the outer 
reaching to the middle of the middle one; while in P. fusca it only 
extends to the base. 

The general tint of plumage in P. aberti has decidedly more of 
ferruginous than in P. fusca. The throat is uniform with the 
breast, and unspotted; the rump too is uniform with the back; in 
both these particulars differing from P. fusca. 

The following table exhibits the relative dimensions of the two 
species :— 

P. aberti. PP. fusca. (Male.) 


Menath: (approximate).........sceseneseneasrssnes 9 inches. 

Wing folded and slightly curved............ 37% “ 37° inches. 
MM TOTASCLON QUIS ys... 4.2.02 ssecescestnoee ens PT e: " 
feealome, the ridge. Jil io.vscvoesvcesidedessscees ae Golway 
Greatestidepth: of bill i.iiesssiesscieasecereeeenes Battle aq 
Middle tail-feather beyond outer,............ +3“ 7 ey 
Pe aN Dae See ges Usha o vile aatentiasenphls« dus snaiedioiatis'svele 43 OS ae 
BANE ane den anictevaie talasidls noni eal Ue d's +0 deen sies'es oi Rah ee een 
PROLIANT: TOD. vo he emesis edipisieaosdiees%solntnaieteras RS Fale a 
PRM CLAT. scacct Costot Souk ses Peacicaececaducaees RR 6 4a CO 


326 APPENDIX C.—BIRDS. 

’ We have dedicated this species to its accomplished discoverer, 
Lieutenant Jas. W. Abert. 

. AGELAIUS XANTHOCEPHALUS, oe ee Blackbird. . 
. Prous varius, L.—Yellow-bellied Woodpecker. 

. CoLUMBA LEUCOPTERA, L.—White-winged Hott 


. CALLIPEPLA SQUAMATA, Vig.—Blue Partridge. 


=f - woe’ Oy pet OD 


. CALLIPEPLA GAMBELI, Nutt.—Gambel’s Partridge. 


Callipepla gambeli, Nutt. Pr. A. N. S. Phila. 1848. 
Callipepla venusta, Gould, Pr. Zool. Soc. Lond. 1846. 


Upper parts and breast, lead colour; crown and occiput, chestnut. 
Base of bill, lore, chin, throat and sides of the neck, black, bounded 
by a white band, broader on the sides. The eyes are included in 
this black hood. The chestnut of the crown is margined with white, 
bordered on each side with black. The lead-coloured feathers of 
the sides and back of the neck are streaked with black along the 
shafts. Sides, deep chestnut, broadly lineated with yellowish 
white. Lower breast, whitish yellow. Indications of a black patch 
on the belly. . Under tail-coverts dull white and black. A long 
recurved crest of six (remaining?) keeled black feathers, the long- 
est measuring 1? inches in length. Inner webs of scapulars mar- 
gined with reddish. Tail much rounded, of twelve feathers; outer 
14 inches shorter than the central. Wing, 4% inches; tail, 4; tar- 
sus, 1. : 


8. ACTITURUS BARTRAMIUS, Wils.—Field Plover. 


9. RECURVIROSTRA OCCIDENTALIS, Via.—White-headed Avoset. 


This species resembles &. americana in the general pattern of 
its markings, but differs in having a pale-grayish white to replace 
the buff of the head and neck. It is also decidedly larger, the 
uncovered part of the tibia measuring 22 inches instead of 2 inches; 
and the tarsus 4 instead of 84 inches. The hind claw is nearly 
obselete. Bill 32 inches long. 

Found abundantly in New Mexico and California. 


a 


APPENDIX (.—BIRDS. oot 


LIST OF BIRDS 


INHABITING AMERICA, WEST OF THE MISSISSIPPI, NOT DESCRIBED IN 
AUDUBON’S ORNITHOLOGY.* 


ARCHIBUTEO FERRUGINEUS, Licht. Abh. Ac. Wiss. Berlin for 
1838. California. 


ROsTHRAMUS SOcIABILIS, Vieill. Nouv. Dict., vol. 18, p. 318. 
Miami River; Cape Florida. 


STRIX FRONTALIS, Licht. Abh. Ac. Wiss.. Berlin. 18388. Cali- 


fornia. 


ACANTHYLIS VAUXII, Towns. Jour. Acad. Nat. Sc., Phila., III. 1, 
p- 148. 18389. Columbia River. 


CHORDEILES BRASILIANUS, (Gm.)—Lawr. Ann. N. Y. Lye. oe 
1851. Rio Glande. Texas. 


ANTROSTOMUS NUTTALLI, Aud. Biog. Birds, 2d ed. VII. 361, 
pl. 495. 1847. Fort Union; California. 


CERYLE AMERICANA, Boie.—Lawr. Ann. N. Y. Lyc. April, 
1851. Texas, Rio Grande. 


ORNISMYA cOsT#, Bourcier, Rev. Zool., 1839, p. 294. Cali- 
fornia. 


CoNIROSTRUM ORNATUM, Lawr. Ann. N. Y. Lyc., April, 1851, 
pl. 4. Rio Grande, Texas. 


PICOLAPTES BRUNNEICAPILLUS, Laf.—Lawr. Ann. N. Y. Lye. 
April, 1851. Rio hind Texas. 


TROGLODYTES ALBIFRONS, Giraud. Texas Birds. 1841. 7. 
mexicanus, Sw. ‘Texas. 


* The list includes a few specimens recently described from the region east of 
the Mississippi. As already stated, the birds mentioned here will all be described 
and figured by Mr. Cassin of Philadelphia, in his forthcoming work on the Birds of 
North America, entitled, ‘Illustrations of the Birds of California, Texas,” &c., 
in continuation of Audubon. 


328 APPENDIX C.—BIRDS. 


VIREO HUTTONI,.Cassin, Pr. A. N. 8. Phila. V. 150. Feb. 1851. 
Monterey. 


VirEO BELLI, Aud. Orn. Biog. 2d ed., VII., 333, pl. 485. 
1843. Fort Union; California. 


VIREO ATRICAPILLA, Woodhouse, Pr. A. N. §S. Phila. VI. 
April, 1852. 


VIREOSYLVA PHILADELPHICA, Cassin, Pr. A. N. S., V. 153., 
Feb. 1851. Philadelphia. 


VIREOSYLVA ALTILOQUA, Vieill. Ois. Am. Sept. 1. Pl. 38. 
1807. Florida. 


SIALIA MACROPTERA, Baird, Stansbury’s Report. 1852. 
Lanius ELEGANS, Sw. F. B.A. 1831. Oregon; Fur countries. 


LANIUS EXCUBITOROIDES, Sw. F.B. A. 1831. Oregon; Fur 
countries. 


HyYPocoLIUs AMPELINUS, Bp. Con. Gen. Ay. I. 336. 1850. 


California. 


IcTERIA VALASQUEZII, Bp. Proc. Zool. Soc. London. 1887. 
California. 


CULICIVORA ATRICAPILLA, Sw. Zool. [].—Lawr. Ann. N. Y. Lye. 
Sept. 1851. Texas. 


SYLVICOLA OLIVACEA, Giraud, Texas Birds, pl. 7. 1841. Texas. 
VERMIVORA BREVIPENNIS, Giraud, Ann. N. Y. Lyc. Texas. 


TURDUS RUFOPALLIATUS, Lafresn., Rev. Zool. 1840, p. 259. 


Monterey. 

MERULA OLIVACEA, Brewer, Proc. Boston Soc. Nat. Hist. I. 
p. Fou. 

Mimus LEUCOPTERUS, Vig. Zool. of Blossom, 18. 1839. Western 
N. America. 


Mimus Lonerrostris, Lafresn., Rev. Zool. 105, 1838. California 
and Mexico. 


ToOXOSTOMA REDIVIVA, Gambel, Pr. <A. N. S. Phila., II. 264, 
Aug. 1845. Monterey. 


APPENDIX C.—BIRDS. 329 


TOXOSTOMA CURVIROSTRIS, Swainson, Matamoras. 


ToxosToMA LECONTEI, Lawr. Ann. N. Y. Lyc., Sept. 1851. 
Gila River. 


Moracinua LEUCOPTERA, Vig. Zool. of Blossom, 1839. Western 
N. America. 


AGRODOMA SPRAGUEI, Aud. Orn. Biog. 2d ed., VII. 335, pl. 
486, 1843. Fort Union. 


SAXICOLA G@NANTHOIDES, Vig. Zool. Beechey’s Voyage, 1839. 
N. W. Coast of America; Labrador? 


SAUROPHAGUS SULPHURATUS, Swainson.—Gambel, in J. A. N. 
S. Phila., I. 39. Gulf of California. 


SAUROPHAGUS BAIRDII, Gambel, J. A. N. S. Phila. I. 40, 1847. 


California. 


Tyrannus cassintt, Lawrence, Ann. N. Y. Lyc. June, 1850. 
Texas. ) 


TYRANNULA CAYANENSIS, Gm. ‘Texas. 


TYRANNULA LAWRENCEII, Giraud, Texas Birds, pl. 2, 1841. 
Texas. 


TYRANNULA CINERASCENS, Lawrence, Ann. N. Y. Lyc. Sept. 
1851. Texas and California. — 


TYRANNULA FLAVIVENTRIS, Baird, Pr. A. N. 8. Phila., I. 283, 
July, 1843. Carlisle, Pa. 


TYRANNULA MINIMA, Baird, Pr. A. N. 8. Phila., I. 284, July, 
1843. Carlisle, Pa. 


PYROCEPHALUS RUBINEUS, Bodd, (P. coronata, Gould)—Law- 
rence, Ann. N. Y. Lyc. April, 1851. Rio Grande, Texas. 


SETOPHAGA VULNERATA, Wagler. ‘Texas. 


SETOPHAGA BELLI, Giraud, Texas Birds, Pl. 4, fig. 2, 1841. 
Texas. 


SETOPHAGA RUBRA, Swainson, Phil. Mag. 1830. Texas. 
SETOPHAGA PICTA, Swainson, Phil. Mag. 1830. Texas. 


SETAPHAGA RUBRIFRONS, Giraud, Texas Birds. pl, Ty. fige alg 
1841. Texas. 


330 APPENDIX C.—BIRDS. 
EMBERNAGRA RUFIVIRGATA, Lawrence, Ann. N. Y. Lye., pl. V. 
fig. 2, April, 1851. Rio Grande, Texas. 


EMBERNAGRA BLANDINGJANA, Gambel, Pr. A. N. 8. Phila., I. 
p- 260. Rocky Mountains. 


SALTATOR RUFIVENTRIS, Vig. Zool. Blossom, 19, 1839. Western 
N. America. 


EUPHONIA ELEGANTISSIMA, Bp. Proc. Zool. Soc. Lond., 1837. 
Texas. 


SPERMOPHILA ALBOGULARIS, Swainson.—Lawrence, Ann. N. Y. 
Lyceum, Sept. 1851. Texas. 

RHAMPHOPIS FLAMMIGERUS, Jard. Ill. III. pl. 131. Columbia 
River, Oregon ? 

CurysopoGa TYPIcA, Bp. Con. Gen. Av. I. 480, 1850. Calli- 
fornia. 


FRINGILLA MERULOIDES, Vig. Zool. Blossom. Monterey. 


ZONOTRICHIA QUERULA, Nutt. Man. I. 555, 2d ed. 1840. 
Z. comata, De Wied. Missouri River. 


ZONOTRICHIA GAMBELI, Nutt. Orn. I. 557, 2d ed., 1840. Co- 
lumbia River. 


ZONOTRICHIA CASSINII, Woodhouse, Proc. ae N. 5S: "Pitas 
April, 1852. Texas. 


CHRYSOMITRIS LAWRENCEII, Cassin, Pr. A. N. S. Phila., V. 105, 
pl. 5, Oct. 1850. San Diego, California. 


PrprLo FuscA, Sw. Phil. Mag. 1827. California. 
PIPILO OREGONA, Bell, Ann. N. Y. Lye. 1848. Oregon. 


PIPILO ABERTI, Baird, Stansbury’s Report. 1852. New 
Mexico. 


EMBERIZA LECONTEI, Aud. Biog. Birds, 2d ed., VII. 338, pl. 
488. 18438. Fort Union. 


EMBERIZA BAIRDII, Aud. Biog. 2d ed., VII. 359, pl. 500. 
1843. Fort Union. 


EMBERIZA BILINEATA, Cassin, Pr. A. N. 8., Phila. V. 104, pices 
Oct. 1850. Rio Grande, Texas. 


APPENDIX (.—BIRDS. 8el 


Emprriza BELLI, Cassin, Pr. A. N. 8. Phila., V. 104, pl. 4. 
Oct. 1850. San Diego, California. 


Carpopacus ozBscurnus, McCall, Pr. A. N. S. Phila., V. 220. 
June, 1850. Santa Fe. 


CARPODACUS FAMILIARIS, McCall, Pr. A. N. S. Phila., VI., 
April, 1852. New Mexico. 


CoccoOTHRAUSTES FERREO-ROSTRIS, Vig. Zool. Jour. IV. p. 352. 
1828-9. N. W. coast America; California. 


CARDINALIS sINUATUS, Bp.—Lawrence, Ann. N. Y. Lye. 
April, 1851. Rio Grande, Texas. 


PYRRHULA INORNATA, Vig. Zool. of Blossom, 20. 1887. 
Western N. America. 


LEUCOSTICTE GRISEINUCHA, Brandt, Orn. Ross. 1842. Aleutian 
Islands. (L. griseogenys, Gould.) 


PLECTROPHANES MACCOWNII, Lawrence. Ann. N. Y. Lye. Sept. 
1851. Western Texas. 


PASSERELLA UNALASCHENSIS, Bp. Con. Gen. Av. 477. 1850. 
Unalascha. 


PASSERELLA RUFINA, Brandt, Orn. Ros. Sitka. 


HuspizaA arctica, Bp. Con. Gen. Av. 469. 1850. (Hmberiza 
chrysops, Pall.) N. W. Coast. 


ALAUDA RUFA, Lath.—Aud. Orn. Biog. 2d ed., VII. 353, pl. 497. 
1843. Texas. 


OTOCORIS OCCIDENTALIS, McCall, Pr. A., N. 8. Phila., V. 218. 
June, 1851. Santa Fé. Salt Lake City. 


STURNELLA NEGLECTA, Aud. Biog. 2d ed. VII. 389, pl. 489. 
1843. Upper Missouri; Utah; New Mexico; California. 


QUISCALUS MAcRoURUS, Sw.—Lawrence, Ann. N. Y. Lye. April, 
1851. Rio Grande, Texas. 


SCOLECOPHAGUS MEXICANUS, Sw. 24 Cent. Birds, No. 66. 1838. 
(Quiscalus brewertt). Fort Union, Missouri; California. 


PENDULINUS CALIFORNIANUS, Less. Rev. Zool., 1844, p. 486. 
California. 


832 APPENDIX (C.—BIRDS. 
PSAROCOLIUS AURICOLLIS, De Wied., Reise, 367. Missouri 
River. 


XANTHORNUS MEXICANUS, Briss.—Vigors, Zool. Blossom. Pacific 
Coast N. America. 


XANTHORNUS AFFINIS, Lawrence, Ann. N. Y. Lye. April, 1851. 
Rio Grande, Texas. 


IcTERUS cUCULLATUS, Sw.—Lawrence, Ann. N. Y. Lye. April, 
1851. Rio Grande, Texas. 


IcTERUS MELANOCEPHALUS, Wagler, Isis, 1829, p. 756. Texas. 


IcTERUS VULGARIS, Daud.—Aud. Orn. Biog., 2d ed., VII. pl. 499. 
1843. South Carolina. 


IcTERUS FRENATUS, Licht. Isis, 1843, p. 59. Greenland? 
Mexico. 


CHAMEA FASCIATA, Gambel, Pr. A. N.S., Phila., II. 265. Aug. 
1845. California. 


LOPHOPHANES SEPTENRIONALIS, Harris, Pr. A. N. §., II. 300. 
Dec., 1845. Upper Missouri; Rocky Mountains; Salt 
Lake. 


LOPHOPHANES INORNATUS, Gambel, Pr. A. N. 8. Phila., II. 265. 


California. 


LOPHOPHANES WOLLWEBERI, Bp. Comptes Rendus. Sept. 
1850. (P. annexus.) Cassin, Oct. 1850, Rio Grande, 
Texas. 


LOPHOPHANES ATRICRISTATUS, Cassin, Pr. A. N.S. Phila., V. 103, 
pl. 2. Oct. 1850. Rio Grande, Texas. 


PARUS MONTANUS, Gambel, Pr. A. N.S. Phila., I. p. 259. New 
Mexico. 


GYMNOKITTA CYANOCEPHALA, De Wied., Reise. Upper Missouri; 
Rocky Mountains. 


OYANOCORAX cOoRONATUS, Sw. Phil. Mag. 1827. Texas. 


CYANOCORAX LUxUosUS, Lesson.—Lawrence, Ann. N. Y. Lye. 
April, 1851. Rio Grande, Texas. 


CYANOCORAX CASSINII, McCall, Pr. A. N. S. Phila, V. 216. . Fane, | 
1851. Santa Fé. 


APPENDIX C.—BIRDS. 333 


GARRULUS CALIFORNICUS, Vig. Zool. Blossom, pl. 5. 1839. 
Monterey, California. 


Pica BEECHEYII, Vig. Zool. Jour. IV. 358. 1828-9. Mon- 


terey, 

CKOTOPHAGA ? New Orleans. Mr. Audubon’s collection. 

PIAYA CAYANENSIS, Gambel, J. A. N.S. Phila., I. 25. Gulf of 
California. 

GxOococoyx AFFINIS, Hartlaub, Rev. Zool., 1844, p. 215. Cali- 
fornia. 


GeEococcyx viaTicus, Wagler—McCall, Pr. A. N. 8. Phila., V. 
220. June, 1851. Texas and New Mexico. 


MELANERPES ALBOLARVATUS, Cassin, Pr. A. N. S. Phila., V. 
106. October, 1850. Sutter’s Mill, California. 


_ MELANERPES FORMICIVORUS, Swainson.—Nuttall’s Man., I. 166. 
, Santa Barbara, California. 


CENTURUS SANTACRUZII, Bp. Pr. Zool. Soc. Lond. 1887, p. 16. 
Western Texas. 7 


CENTURUS FLAVIVENTRIS, Swainson, 23 cent., Lardner, Cab. 
Cyclopedia. Texas. 


CENTURUS ELEGANS, Sw.—Lawrence, Ann. N. Y. Lyc. April, 
1851." 24 cent. Rio Grande, Texas. 


COLAPTES MEXICANOIDES, Lafres. Rev. Zool., 1844. California. 


CoLapres avREsi1, Aud. Biog. Birds, 2d ed., VII. 348, pl. 494. 
18438. Fort Union. } 


CoLAPTES COLLARIS, Vig. Zool. Blossom, p- 24, pl. 9.; Zool. 
Jour. IV. 854, 1828-9. Monterey. 


Picus scaPuLaRis, Vig. Zool. Jour. IV. 353, 1828-9. San 
Blas, California. 


Picus nurrauui, Gambel, Proc. A. N. S., Phila., I. 259. (P. wil- 
sonit, Malherbe.) California. 


Picus scaLaris, Wagler, Isis, 1829. California and New Mexico. 
Picus LecontvEI, Jones, Ann. N. Y. Lyceum, IV. 489. Georgia. 


CoLUMBA SOLITARIA, McCall, Pr. A. N.S. Phila., IT. 233, July, 
1847. Matamoras. 


334 APPENDIX C.—BIRDS. 


CoLUMBA FLAVIROSTRIS, Wagler—Lawr. Ann. N. Y. Lyceum. 
April, 1851. Texas, Rio Grande. , 


PENELOPE POLIOCEPHALA, Wagler.—McCall in Pr. A. N.S. Phila., 
V. 222. Matamoras and Rio Grande. 


ORTALIDA VETULA, Wagler.—Lawr. Ann. N.Y. Lyceum. April, 
1851. Texas, Rio Grande. 


CYRTONYX MASSENA, Gould.—McCall, Pr. A. N. 8. Phila., V. 
221. San Pedro and Rio Pecos, New Mexico. 


CALLIPEPLA GAMBELI, Nutt. Pr. A. N. 8. Phila, I. 260. April, 
1843.—Callipepla venusta, Gould, Pr. Zool. Soc. Lond., 
1846, p. 71. New Mexico and California. 


‘CALLIPEPLA prota, Dougl. Linn. Trans. Lond. California. 
CALLIPEPLA ELEGANS, Less. Cent. Zool., pl. 61. California. 


CALLIPEPLA DOUGLASSII, Vig. Zool. Jour., IV. 3538. 1828-9. 
Monterey, California. 


CALLIPEPLA SQUAMATA, Vig. Zool. Jour., V. 275. 1830. New 
Mexico. 


STREPSILAS MELANOCEPHALUS, Vig. Zool. Jour., IV. 358. 
1828-9. Monterey. 


NUMENIUS RUFIVENTRIS, Vig. Zool. Jour., IV. 856. 1828-9. 
Pacific coast of N. America. 


MACRORHAMPHUS SCOLOPACEUS, Lawrence, Ann. N. Y. me 
Mississippi Valley. (Limoea sbolopacea, Say). 


RECURVIROSTRA OCCIDENTALIS, Vig. Zool. Jour., IV. 356. San 
Francisco. New Texte, 


ANSER NicRIcANS, Lawr. Ann. N. Y. Lyc., 1846. Atlantic 
coast. 


ANAS UROPHASIANUS, Vig. Zool. Jour., IV. 353. 1828-9, 
_ N.W. coast N. America. 


DENDROCYGNA ARBOREA? Penn. Mexico. South Carolina? 


DENDROCYGNA AUTUMNALIS, Hyton.—Lawr., Ann. N. Y. Lye. 
April, 1851. Rio Grande, Texas. 


CYANOPTERUS RAFFLESII, King. Salt Lake, Utah. Louisiana. 


APPENDIX C,—BIRDS. 835 


OIDEMIA VELVETINA, Cassin, Pr. A. N.S. Phila, V. 126. Decem- 
ber, 1850. Atlantic coast. (O. fusca of former authors). 


LARUS BRACHYRHYNCHUS, Gould, Pr. Zool. Soc. Lond., 1841, pl. 
106. July, 1848. Zoology of the Sulphur, pl. 834. Wes- 
tern N. America. 


LARUS BELCHERI, Vig. Zool. Jour., IV. 858. 1828-9. Pacific 
coast of North America. 


STERNA ELEGANS, Gambel, Pr. A. N.S. Phila., IV. 129. Decem- 
ber, 1848. Mazatlan. 


Sterna oasprA, L.—Lawrence, Ann. N. Y. Lyc. May, 1850. 
Atlantic coast, U. 8. 


PROCELLARIA MERIDIONALIS, Lawrence, Ann. N. Y. Lyc. Fe- 
bruary, 1847. Indian River, Florida. 


THALASSIDROMA FuURCATA, Lath.—Gould, Zool. Sulphur. Rus- 


sian America. 


THALASSIDROMA FREGETTA, Kuhl.—Lawrence, Ann. N. Y. Lye. 
April, 1851. Florida. 


PHALACROCORAX PERSPICILLATUS, Pall. Zool. Ros. As., IL. 308. 
Gould, Zool. Sulphur, pl. 32. Russian America. 


PHALACROCORAX PENICILLATUS, Brandt. Monterey. 


Urnia BREVIROSTRIS, Vig. Zool. Jour., IV. 857. 1828-9. Pa- 
cific coast of N. America. 


MERGULUS CIRROCEPHALUS, Vig. Zool. Blossom. 1839. Pacific 
coast of N. America. 


- MERGULUS CASSINII, Gambel, Pr. A. N.S. Phila., II. 266. Au- 
gust, 1845. Coast of California. 


PTYCHORHAMPHUS ALEUTICUS, Brandt, Bull. Sc. St. Peters- 
burg, II. 1887. Aleutian Islands. 


BRACHYRHAMPHUS WRANGELI, Brandt, Bull. Sc. St. Peters- 
burg, IT. No. 20. Aleutian Islands. 


BRACHYRHAMPHUS BRACHYPTERUS, Brandt, Bull. St. Peters- 
burg. Unalaschka. 


336 APPENDIX (C.—REPTILES. 


REPTILES. 


———— 


By SPENCER F. BAIRD anp CHARLES GIRARD. 


me 


SIREDON LICHENOIDES, Baird. 
Ps... 


Spec. cuar.—Body uniform blackish brown, covered all over with licheniform 
patches of grayish yellow; snout rounded; tail compressed and lanceolated ; 
toes broad and short. 

The addition of an authentic new species to the genus Stredon 
will justly be considered as of great interest to herpetologists. 
Two species are now clearly ascertained to exist; perhaps a third, 
if the one mentioned by Prof. Owen* be really such. It is not impro- 
bable that many more exist, as we have accounts of many localities 
of «fish with legs,” in various parts of Mexico, New Mexico, and 
Texas, although, as yet, we have been able to procure only these 
two species. ‘The possession of these, however, allows a comparison 
of characters by which the absolute generic features of the group 
can be better ascertained. 

The figures hitherto published of S. mexicanus, and the imper- 
fect sketch of S. maculatus, are far from being satisfactory, and 
do not allow any accurate comparison to be made of their specific 
features. As these will have to be critically redrawn in order to 
meet the wants of science, we have endeavoured to obtain, 
and we hope with success, figures of S. léchenozdes that will enable 
future comparisons to be satisfactorily made. Our specimen is a 
little more than six inches and a-half in length, the figures being 
all of natural size. Whether this. be the absolute size of the 
species which it represents, we are unprepared to state. If such 
was the case, it is considerably smaller than S. mexicanus. The 
tail forms nearly the half of the entire length, and the head a little 
less than the fifth of the same. 

The head is ovoidal, much broader than deep, and the snout 
rounded, a character which at once will disttmguish our species 


* Annals and Magazine of Natural History, xiv: 1844, 23. 


piled SHClONSHIIT NOGaals , 


ane rm 


. ak 
24 oe 
' * “ ips ? 1 
ae vat.) > 
e 
° 
i 
. 
. 
- 
* 
*. 
~ 
. 
* 
al 
. ™ 
; 
= 
; 
‘- 
fs Pa ~ 
%, 4 
tind . 
re at 
« r i eC hat 
, 
adi. 
, 
. *’ 
7 “ . 
~* 
r ° 
° . 
ail * 
7 i 
‘ e * 
A 
. ¥, Ss 
ye 
ae 
* 
. HY * ral 
" - 
/ ' mn 
. 
- 
rd 
; 4 ‘ 
Pe 
i . 
z 
7 - - 
in 


- APPENDIX C.—REPTILES. 837 


from S. maculatus, the chief character of which consists in the 
truncated snout. The eyes are of medium size, situated behind 
the angle of the mouth. Their position and size in S. maculatus 
we do not know accurately, but in S. mexicanus these organs 
would be considerably smaller and more approximated to the end 
of the snout, judging of these peculiarities from the various 
sketches given of that species, and proportionally much smaller 
than in a specimen of Stredon in our possession from the city of 
Mexico, and of about the same size as our S. lichenoides. If these 
proportions are correct in the drawings of S. mexicanus, we would 
undoubtedly have a species which could not be accurately charac- 
terized until further information should be obtained. The nostrils 
are very small and near the end of the snout. The body is sub- 
cylindrical, subfusiform, broader and deeper at its origin than on 
any point backward. ‘The tail is very much compressed, elongated, 
and tapering intoa point. The dorsal membrane commences at the 
occiput, rising gradually until the middle of the tail, whence it di- 
minishes again toward its pointed tip. The membrane under the tail 
is lower than that above, extending from behind the vent to the tip 
of the tail, and reaching its greatest height on its anterior third, but 
diminishing more rapidly forward than backward. The anal open- 
ing is very large, elongated, and rendered very conspicuous by the 
great development of the fleshy masses which constitute its margin. 

The presence of four external flaps, provided with respiratory 
fringes, is a generic character, belonging to all the species hitherto 
known. Their real appearance has been misrepresented in many 
sketches, as we could satisfy ourselves by the examination of two 
Species preserved in alcohol. The branchial fringes do not extend 
all along the upper edge of the branchial flap. They occupy 
densely the lower edge of that cutaneous appendage from its 
origin to its tip, and thence for a short space above, but much 

less developed here than below, as we have endeavoured to show 

in the profile of fig. 2. The fringes themselves are very much 
flattened, tapering, and disposed upon a double row, so that each 
of them appears as if double; but it is easy to ascertain that the 
row on either side does not combine with the other. 

The fore and hind legs have nearly the same length when mea- 
sured from their bases to their extremities; the hind ones, however, 
are much thicker, and the toes of both pairs are neither so slender 
nor so elongated as in S. mexicanus and S. maculatus. 


The ground colour is blackish brown; there are irregular patches 
22 


338 APPENDIX C.—REPTILES. 


of grayish yellow spread all over the body, head, and tail, remind- 
ing us of surfaces over which lichens grow, whence the specific 
name by which we designate this species. 

It was caught by R. H. Kern, Esq., in Spring Lake, at the head 
of Santa Fé Creek, in New Mexico, accordingly a member of the 
fauna of the basin of the Rio Grande del Norte. 

In a revision of the North American Tailed Batrachia, published 
in the Journal of the Academy of Natural Sciences, 2d series, vol. 
i, p- 281, (1849,) we intimated a doubt as to whether Siredon or 
the axolotls were adult animals. Their wonderful resemblance to 
the larve of Ambystoma punctata was our chief ground for this 
belief. Since then, however, we have seen the description and 
figures by Sir Everard Home of 8. mexicanus, in which he clearly 
indicates, in one individual at least, ovaries distended with eggs. 
We have also seen specimens with the genital apparatus present- 
ing the tumid and highly developed appearance of salamanders in 
general, when in the breeding season. 


CNEMIDOPHORUS TIGRIS, Baird and Girard. 
Pl. II. 


Spec. cHaR.—Scales on the subguttural fold small in size; four yellowish, indis- 
tinct stripes along the dorsal region. 


This species, one of the most elegant of its genus, is the third 
hitherto described as found in the United States, for we have no 
doubt that the lizard referred to in Long’s expedition, under the 
name of Amezva tesselata, will come under this genus, and be closely 
allied to our species. No specimen of this being extant at the 
present time in any known collection, a direct comparison with the 
other species of Cremidophorus was not possible. That A. tesselata, 
however, although closely allied to, is not identical with our species, 
we think that any one will be convinced on comparing Say’s de- 
scription with the figures in pl. IJ. These are of natural size, and 
exhibit most admirably the structure of the regions in their most 
minute details. The plates of the head above (fig. 3), below (fig. 2), 
and on the sides (fig. 1), need no further description. The minute 
scales of the back and upper portion of the legs contrast strikingly 
with the eight rows of large scales of the belly and those of the 
lower part of the hind legs, as well as with those of the inferior 
surface of the head and throat. On the tail again, the scales 
assume another character, well opposed too; they are longer than 


oes, nen es ; 


Avker usr bar's aT RYoadway wy : = 


a eee 


O° dba lh ok ae 


APPENDIX C.—REPTILES. 339 


broad, arranged in annular rows, or else verticillated and slightly 
carinated. The tail is cylindrical, and two and a-half times the 
length of the body and head together: it tapers gradually and 
terminates in a point. 

The ground colour appears to have been bluish yellow, marked 
with irregular patches of black. In some specimens, four longitu- 
dinal yellow stripes may be seen extending from the occiput to the 
base of the tail, and occasionally to a little distance on the latter. 
In the young state, the black patches predominate, unite, and form, 
as it were, the ground colour, and the yellow constitutes irregular 
small spots. 

A series of individuals of different sizes were collected by Capt. 
Stansbury in the valley of the Great Salt Lake. 


Genus CrotapHytus, Holbrook. 


Gen. cHar.—Head covered with small and polygonal plates. The occipital pro- 
per, minute. Teeth on the jaws, pterygoids and palatines, rudimentary on 
the latter. Broad auditory aperture. Femoral pores present; no anal ones. 
Tail very long. 

Of the genus Crotaphytus, a typical form peculiar to North 
America, but one single species has hitherto been discovered, the C. 
collaris, observed for the first time by Major Long’s party, on their 
expedition to the Rocky Mountains. The discovery of a second 
species of this genus within the limits of the United States will be 
received with interest: both of these have a very close generic re- 
lationship, the differences being found in some minor details of their 
structure. 

The generic characters are the following :—The occipital region, 
vertex, and front are covered with small and polygonal plates. On 
the superciliary region and around the nostrils the plates are 
scarcely larger than the scales of the back. The odd occipital 
plates are inconspicuous, being but very little-larger than the sur- 
rounding ones. The auditory apertures are broadly open. Teeth 
exist on the maxillaries, palatines, and pterygoids; conical, acute, 
and slightly curved on the anterior part of the jaws, they are com- 
pressed and tricuspid on the posterior. The palatine teeth are 
rudimentary. The skin is folded under the throat. The scales of 
the upper part of the body are polygonal, and smaller than those 
of the lower part, or belly, and tail; those under the head have 
nearly the size of those of the back. The femoral pores are very 


340 APPENDIX C.—REPTILES. 


distinct ; the anal ones are wanting. The tail is cylindrical, longer 
than the body and head together. 

The genus Crotaphytus differs from Holbrookia in having exter- 
nal auditory apertures, teeth on the pterygoids, and but a minute 
occipital plate. The shape of the head is likewise more elongated 
and pointed in front. 


CROTAPHYTUS WISLIZENII, Baird and Girard. 


PurateE III. 

Spec. coaR.—Head proportionally narrow and elongated. Cephalic plates and 
scales on the back very small. Yellowish brown, spotted all over with small 
patches of deeper brown or black. 

C. wislizenii has the same general form and appearance as C. 
collaris, exhibiting the same contracted neck and fold under the 
throat, the same compact body, the same cylindrical and elongated 
tail, and the same shape and proportions of locomotive members 
and terminating toes. The differences by which the two are dis- 
tinguished, although of a comparatively minor character, are readi- 
ly appreciable when both are directly compared. Thus the head 
of C. wislizenii is proportionally more elongated and narrower than 
that of C. collaris. The small and polygonal plates which cover its 
upper surface and sides are smaller, as well as those of the lip of 
the lower jaw. ‘The scales of the back are likewise smaller, and 
those of the belly larger. The tail 1s somewhat longer, and its 
scales larger in C. collaris; these are subverticillated in both 
species, and subcarinated from the middle of the back toward its 
extremity. The pores of the lower surface of the thighs are more 
conspicuous in C. collaris, independently of the fact that they are 
generally less so in the female than in the male of the same species. 
Immediately behind the vent, at the origin of the tail, there exists, 
in the male, a row of large scales more uniform in C. collaris than 
in C. wislizenit. The specimen figured on our plate III. being a 
female, these anal plates are not to be seen in fig. 4. 

In the colours of the body distinctive marks will at once be 
found. (. collaris possesses on the sides of the neck a double 
band of black bordered with white, which does not exist in O. wis- 
lizenti. The upper surface of the body of the former is scattered 
all over with small yellow dots, which indeed are found in the lat- 
ter, but are much smaller and more numerous, having in addition, 
intermixed with them, irregular roundish brown spots, extending 


Pv Pa 


a 


APPENDIX (C.—REPTILES. 341 


even to the head. The tail is irregularly annulated with brown 
and yellow. 

Nothing is known of the habits of the crotaphyti. We found 
in the stomach of one specimen of the C. wislizenit the remains of 
a species of Cnemidophorus allied to C. sexlineatus. 

The specimen which we have had figured was caught near Santa 
Fé, by Dr. Wislizenus, during the Mexican war. ‘To him we take 
great pleasure in dedicating it, in testimonial of his zeal for science, 
during the arduous duties of an army surgeon, while attached to 
Colonel Doniphan’s command. On his return, the specimen was 
sent to Dr. Le Conte, and by him kindly transmitted to the Smith- 
sonian Institution. The same species was obtained by Colonel J. 
D. Graham, between San Antonio and El Paso del Norte, while 
on his journey to the boundary-line, in May last. 


Genus HoLtBrooxta, Girard. 
Syn. Cophosaurus, Troscu. Arch. f. Naturg. (1850, I.) 1852. 


GEN. cHAR.—Head covered with small and polygonal plates. No auditory 
aperture. No teeth on the palatine bones. A fold of the skin on the breast. 
Femoral pores present, but no anal ones. 


This genus was established in 1850, upon a small lizard, much 
less remarkable in its general aspect than in its structure. In its 
appearance it is so similar to certain species of the Chilian procto- 
treti, that at first glance no one could suppose it to be different 
from the latter genus. But in examining it more attentively, we 
find no external opening to the ears, a fact that cannot but strike 
the observer. There are no teeth on the palatines—another 
character which distinguishes Holbrookia from Proctotretus. Both 
of these genera have a fold under the throat, but the former has 
femoral pores, which are wanting in the latter. The anal pores are 
absent in Holbrookia, while they exist in Proctrotretus. 

The genus Holbrookia will, no doubt, prove somewhat related 
to Crotaphytus, having, like the latter, the upper surface of the 
head covered with small and polygonal plates, and well-developed 
femoral pores. The elongated tail of the crotaphyti, although so 
disproportionate when compared to Holbrookia maculata, will no 
longer appear as a feature peculiar to the genus, so soon as we 
shall have an opportunity to give a description and a figure of 
another species of the same genus, and which was lately collected 
by Mr. John H. Clark, zoologist to Colonel J. D. Graham, while in 
charge of the survey of the United States and Mexican boundary. 


342 APPENDIX (C.—REPTILES. 


As these pages were passing through the press, we received the 
numbers iv. and y. for 1850, and ii. and iii. for 1851, of the «« Archiv 
fiir Naturgeschichte.” The present editor of that periodical, 
Dr. Troschel, describes, in the fourth number for 1850, a new 
saurian genus, under the name of Cophosaurus, a species of which 
was brought to Germany by the geologist, Ferd. Roemer, who visited 
Texas some years ago. Having the same saurian in our possession, 
collected in the same locality, as a second but undescribed species 
of our genus Holbrookia, we could readily identify the genus 
Cophosaurus. It is to be regretted that the description of our 
genus Holbrookia which reached Germany in the summer of 1851, 
that is before the publication of the number iv. of the Archiv. fiir 
Naturgeschichte for 1850, has been overlooked by this able 
German zoologist. The absence of auditory apertures, <aures 
externe null,’ which is the most striking character of our genus, 
would have struck Dr. Troschel, had he been aware of our descrip- 
tion in the Proceedings of the American Association for the Ad- 
vance of Science, fourth meeting, held at New Haven, in August, 
1850. 

As it may hereafter be questioned which of these generic names 
has the priority, inasmuch as the volume in which Cophosaurus is 
published bears the date of 1850, we deem it proper to call atten- 
tion to the fact that its publication took place in 1852. LHven at 
the date at which we write these lines, (April, 1852), the year 
1850 of the Archiv. fiir Naturgeschichte has not been completed. 
This fact shows how an author may be deprived - the fruit of his 
labours by an anachronism of this kind. 


HoLBROOKIA MACULATA, Girard. 
Pu. VI. Fic. 1-3. 
Syn. Holbrookia maculata, Girarp, Proc. Amer. Assoc. Ady. Sc., IV. (1850), 
1651, 201. 


Sprc. cHAR.—Tail about the length of the trunk. Head subcircular, slightly 
conical in front. Pectoral fold bordered with large scales. 


The general form of this species is rather thick and short than 
elongated, especially in the female: the young and the males are 
more slender. The body is subcylindrical, the tail conical and 
very stout at its origin, tapering however suddenly away. The 
entire length is between three and four inches, as shown by the 


pareoi (el 


mney 


379 Broadway NY 


Ackerman Lith 


4-8. PHRYNOSOMA MODESTA, Girard 


ia 


Fig 1-3. HOLBROOKIA MACULATA Girard. 


¥ 


APPENDIX C.—REPTILES. 343 


figures, all of which are drawn of natural size. The tail is of the 
length of the body, the head excluded. The latter is oval, broader 
than high; its summit being convex, and its snout truncated. It is 
‘covered with small, irregular, and polygonal plates, larger on the 
middle line of the skull than above the eye, the nose, and the nape. 
There is a supraorbital carina, with small elongated plates, 
scarcely to be seen with the naked eye. The infraorbital plates 
are less numerous, but longer. ‘The eyes occupy the middle of the 
length of the head. The eyelids are bordered by a row of minute 
and pointed plates, forming a serrated edge. The nostrils are 
nearer the end of the snout than the eye. ‘The angle of the mouth 
extends to the posterior rim of the orbit. ‘The upper jaw is bor- 
dered with a row of small, very elongated plates, obliquely imbri- 
cated. Margining the lower jaw there are two rows of small an- 
gular plates, the larger ones being at the angle of the mouth. 
There is a single row of small conical teeth on both jaws; those 
in front are acute and slightly recurved; those behind stouter and 
erect, with a carina separating the rounded crown from the body 
of the tooth. The posterior extremity of the tongue has a semi- 
lunar notch. As observed in the generic paragraph, there are no 
external auditory apertures; the tympanum is covered by scales 
altogether similar to those of the neck. On both sides of the 
neck and immediately behind the angle of the mouth, is a fold of 
the skin, which vanishes in a depression under the head. Farther 
backward, and on the breast, is situated another fold, constituting 
an elegant neck ring, which, however, does not extend higher than 
the shoulders. 

The anterior legs are shorter and more slender than the poste- 
rior. There are five toes, similar in each pair of limbs, elongated, 
slender, terminated by a compressed and recurved nail. The toes 
and nails of the posterior limb, however, are a little longer in pro- 
portion to the limbs themselves. The fourth toe is the longest. 
the two external ones the shortest, the second and third nearly 
equal. There are eleven femoral pores on each thigh. 

The scales are slighty imbricated, subcarinated on the back and 
sides, smooth underneath. They are smaller on the neck and at 
the base of the limbs than on the sides and back. ‘Those on the 
tail are indistinctly verticillated. The smallest ones are found 
under the head, in the region of the gros and behind the vent; 
they are larger on the abdomen than under the tail. The toes 
are entirely covered with scales. 


344 APPENDIX C.—REPTILES. 


The colour, according to a drawing made from life by an artist 
of great merit, Mr. William H. Tappan, while on the River Platte 
in 1848, is olivaceous brown, slightly violaceous on the sides of 
the head. On each side of the body there are two, sometimes 
three oblong spots, of a deep black ; and on the body two rows of 
quite large, irregular blackish-brown patches, with a band of a 
lighter colour between eachrow. ‘There is an indication of a third 
row of these patches, less apparent, however, in the male, in which 
again we find two orange-red lines, the uppermost extending to 
the end of the snout in passing over the eye; the other fol- 
lows the lower jaw. In advance of the eyes the orange hue of 
these lines passes into brighter yellow. 

This species inhabits the valley of the Platte River, as collected 
there by W. H. Tappan. We have received several specimens 
from Texas, collected by General Churchill, one of which is the 
original from which the accompanying drawings have been made. 
The individual represented is a female. 


Genus Uta, Baird and Girard. 


GEN. CHAR.—Upper part of body covered with minute scales; a pectoral fold; 
auditory apertures ; femoral pores, but no anal ones. 


The genus which we now establish will not fail to attract the 
attention of herpetologists, having a relation to both Sceloporus 
and Holbrookia. The former genus it resembles in having the 
upper surface of the head covered with similar scales, and in being 
provided with auditory apertures. On the other hand, the body is 


covered with scales, like those in Holbrookia, while on the tail they 


are much large than in any of the above genera, thus contrasting 
greatly with those of the back. The genus Uta, moreover, has a 
subgular fold of skin, constituting a neck-ring similar to that in 
Holbrookia. Its elongated tail would recall to mind the genus Cro- 
taphytus, were not the palatine teeth absent. Femoral pores 
exist, while anal ones are wanting. 

Besides the species described below, this genus embraces two 
others, one of which is entirely new to science, and was sent in by 
Colonel J. D. Graham from the boundary-line, and which we eall 
Uta ornata; while the other was described as a Sceloporus, and first 
as a mere variety of S. grammicus, afterward, however, separated 


under the specific name of S. microlepidotus. The minutenessof - 


APPENDIX C.—REPTILES. 345 


the scales to which it makes allusion becomes now a generic charac- 
ter. Thus the species included in the genus Uta are, generally 
speaking, of a smaller size than the scelopori proper. 


UTA STANSBURIANA, Baird and Girard. 
Pu. V. Fia. 4-6. 


Spec. cHAR.—Tail slender, elongated, and conical, provided with large scales 
arranged in verticils; a subgular fold in addition to the pectoral one. 


This species is one of the handsomest and most graceful among 
the small saurian forms peculiar to North America. In its general 
aspect it recalls to mind the cnemidophori, having, like the latter, 
the body covered with small polygonal scales, while on the tail 
the scales are large, elongated, and verticillated. The proportions 
in length between the body and tail, and the fold under the throat, 
are likewise similar in both Cnemidophorus and Uta. But when we 
compare the shape and structure of the head in the two, we detect 
differences which are not merely generic, but cause them to belong 
to different families. The upper surface of the head is covered 
with small and polygonal scales, as in Sceloporus, Crotaphytus, and 
Holbrookia; its shape is rather rounded than conical or elongated, 
and is broad and flattened, while it is narrower and higher than 
broad in enemidophori and allied genera. The body is subfusi- 
form, the tail slender, conical, and nearly one and a-half times the 
length of the body and head together. 

The occipital plate is polygonal and comparatively large; three 
or four superciliaries, the vertical and frontal plates, are a little 
larger than the many others by which they are surrounded. — 
The nostrils are very conspicuous, and approximated to the end 
of the snout; they are situated interior to the prolongation of the 
superciliary ridge, and composed of elongated and narrow plates. 
The rostral plate is narrow, and is scarcely to be seen when viewed 
from above, but much more conspicuous than in Holbrookia when 
looked at in front. The plates lining the margin of the Jaws are 
quadrangular, elongated, and much narrower on the upper than 
on the lower jaw. Under the chin, three or four pairs of polygonal 
scales are considerably the largest. Between the latter and those 
on the margin of the lower jaw two other rows may be seen, com- 
posed of scales a little larger than the subgular ones. ‘The audi- 
tory aperture is moderate, surrounded by irregular folds of the 
skin which extend under the throat; from its anterior edge pro- 


346 APPENDIX C.—REPTILES. 


ceed three or four conical and acute scales; the posterior pectoral 
fold is provided along its margin with a row of large, smooth, and 
subquadrangular scales. The anterior and upper surface of the 
locomotive members are covered with scales nearly of the same 
size of those of the belly, but carinated and a little narrower pos- 
teriorly. Scales, similar to those of the belly, extend on the lower 
part of the hind legs, are somewhat larger, especially on the an- 
terior margin. The toes are slender and terminated by short, 
pointed, and slightly recurved nails. ‘The small scales which cover 
the middle region of the back, from the anterior limbs to the 
origin of the tail, are somewhat longer than those of the occiput 
and sides from the ear to the hind limbs. ‘The posterior part of the 
thigh and groins are covered with minute scales similar to those 
of the sides of the body. The largest scales are seen on the tail, 
as already observed; they are verticillated and strongly carinated. 

The colour below is uniform greenish-yellow, except under the 
head, where the green predominates, intermingled with brownish, 
narrow bands. Above it is blackish-brown, marbled with greenish- 
yellow, or whitish-yellow irregular dots. Sometimes a double row 
of dorsal patches of a deeper black may be seen along the back in 
some individuals, recalling to mind a similar distribution of colour 
in Sceloporus scalaris, S. graciosus, and Holbrookia maculata. 

This species is from the valley of the Great Salt Lake, where it 
was collected by Captain Howard Stansbury ; in remembrance of 
whose services to the country and to science we have designated 
it under the name which it now bears. 


ScELOPORUS GRACIOSUS, Baird and Girard. 


Pu. V. Fic. 1-38. 


Spec. cHaR.—Head subconical; scales of the back proportionally large; tail of 
medium size, slender and conical. 


This small and graceful species has a much greater affinity with 
the Mexican Sceloporus scalaris than with S. undulatus of the 
United States. The most striking character, as compared with S. 
scalaris, consists in the marked difference in size between the 
scales of the back and those of the base of the tail. The latter is 
proportionally longer than in S. scalaris, and shorter than in S&. 
undulatus. The body of our species is subcylindrical, and rather 
short; the specimen figured being a female, the abdomen is repre- 


APPENDIX C.—REPTILES. 347 


sented in a state of too great expansion. The head is ovoidal, de- 
pressed, subconcave above. The occipital plate is heptagonal, 
broad, and linear posteriorly, pointed and very acute anteriorly. 
It is surrounded by twelve smaller plates, six of which, very small, 
line the posterior edge, while the dther six are distributed on the 
remaining circumference, three to the right and three to the left. 
There are two vertical plates, the anterior one the largest and pen- 
tagonal. In advance of the latter, seven or eight polygonal frontal 
plates form a conspicuous group, while near the extremity of the 
snout the plates are as minute as the scales on the neck. The 
plates which line the margin of the jaws are small, narrow, elon- 
gated, and inconspicuous. On the chin there are three pairs compa- 
ratively larger and conspicuous. ‘The scales on the back are spade- 
shaped and strongly carinated from the neck to the tip of the tail. 
On the latter region they are a little more acute posteriorly and 
verticillated. On the sides of the abdomen they are smaller, and 
their outline less regular. The abdominal scales are smooth, . 
irregularly lozenge-shaped, bidentated posteriorly as in S. scalaris. 
The opening of the ear is subtriangular, protected by several pro- 
jecting scales arising from its anterior edge. The toes and nails 
are very slender; the latter are proportionally long and slightly 
eurved. When the fore legs are stretched backward, the tip of 
their toes will reach to the knees of the hind ones when the latter 
are brought forwards. 

The brillianey of the general hue having disappeared by immer- 
sion in alcohol of the specimens collected, we are not prepared to 
describe this accurately. As to the markings, they differ somewhat 
from those of S. scalaris. The row of the large crescent spots 
along the back is more compact, and not so distinctly bordered 
with white. The yellowish band that runs from behind the eyes 
backward to the middle of the tail is much broader; and besides, 
there is a second similar band extending from below the snout, and 
passing under the eye and above the auditory aperture, to the in- 
sertion of the hind locomotive limbs; the sides, therefore, are not 
ornamented with vertical, slightly undulating dark stripes; the 
irregular patches that are seen on that region are entirely deprived 
of any white margin. The abdomen in the male is blue indigo, as 
in most species of the same genus. The neck and throat are uni- 
color in both sexes. 

This species inhabits the valley of the Great Salt Lake, where 
it was collected by Captain Stansbury and Lieutenant Gunnison. | 


348 APPENDIX (C.—REPTILES. 


ELGARIA SCINCICAUDA, Baird and Girard. 


Pu. IV. Fie. 1-8. 
Syn. Tropidolepis scincicaudus.—Skilt. Amer. Jour. of Sc. VII. 1849, 202, fig. 1-8. 
Spec. cuarn.—Dusky green above; light ash colour below. Eleven transverse 
black bands on the back, interrupted on the dorsal line; white dotted pos- 
teriorly, six or more on the tail. Thirteen to fourteen rows of scales, well 
carinated. 

The individual of this species which we have had figured, although 
not quite full-grown, exhibits, nevertheless, all the essential charac- 
ters for its identification and specific distinction. A much younger 
specimen was figured by Mr. Skilton, in the seventh volume of the - 
second series of the American Journal of Sciences and Arts, where 
it is given as a species of Zropidolepis. ‘The scales, indeed, are 
carinated, but this is not a character peculiar to the genus Tropi- 
dolepis. A closer examination soon reveals the characteristic 
features of the gerrhonoti, of which the genus Higaria is a mere 
subdivision. ‘The conical and tapering tail, which is longer than 
the body and head together, forms the prominent distinctive mark 
of the genus in which we place the present species. The longitudi- 
nal area of minute scales, which extends from behind the ears to 
the insertion of the hind legs, belong to gerrhonoti generally, and 
distinguishes them from Tropidolepis or Sceloporus. In elgaria 
and gerrhonoti, in general, there are neither femoral nor anal 
pores, while their existence in Sceleporus will enable any one to 
distinguish between them. 

The hind legs of H. scincicauda are slightly longer and stouter 
than the anterior, not taking the toes into consideration, as those 
of the hind feet are generally much the longest. The head is flat- 
tened, and has the shape of an acute triangle, the summit of which 
would be rounded. The snout therefore is rather elongated, with 
the end conical. The plates which cover its upper surface are 
smooth, and faithfully represented m fig. 3. The scales of the 
back constitute twelve or thirteen rows, each of which is provided 
with a medial carina, or ridge, extending over the tail. The scales 
of the belly and lower surface of the tail are smooth; on the ab- 
domen they form twelve rows, the middle one being composed of 
larger scales. Fig. 2 exhibits the differences of the scales under 
the head, throat, belly, and tail. The scales form transverse as 
well as longitudinal rows, not only on the tail, as is often the 
ease, but likewise on the body itself. Those on the throat and- 


. 


pues) y PLO ‘WANVINOLUAS NOGOLLSHId 9-9  — Prratg Be pareg' VONWOIONIOS WIAWO vo) tu 


Jue dt As tape Yu] Utada doy 


APPENDIX (0.—REPTILES. | 849 


lower surface of the head alone seem not subjected to any serial 
order. | } 

Possessing only specimens that have been immersed in alcohol, 
and therefore have lost their general hue, we can only, in allusion 
to the colour, remark that the upper part of body and tail have 
transverse and irregular, sometimes undulating, and sometimes 
angularly broken bands of deep brown or black. ‘Ten of these 
bands belong to the body from the occiput to the origin of the tail. 
On the tail itself they extend more or less toward the tip, accord- 
ing to the size of the specimen. 

This species inhabits Oregon, about the Dalles of the Colum- 
bia River, where it has been collected by Rev. George Geary. A 
specimen in our possession, from California, was collected by 
Colonel Frémont. That which is represented on our plate we owe 
to the kindness of Dr. Avery J. Skilton. 


PLESTIODON SKILTONIANUM, Baird and Girard. 


Pu. IV. Fic. 4-6. 


Spec. cHaR.—Head small, continuous with the body; tail stout, very long, and 
subquadrangular; olivaceous brown, with four broad bands of black. 


This is a species of skink which must strike any one familiar 


with the general appearance of the other species of the same 


genus inhabiting North America. The tail, although considerably 
developed in all the skinks of the genus Plestcodon, acquires in 
this species much greater proportions. In the specimen figured, 
the tip of that organ is wanting, but when restored, the whole 
organ would be nearly twice the length of the rest of the body and 
head. Its form is rather subquadrangular than conical, and pre- 
serves a general stoutness which is not seen in the other species, 
in which it tapers more suddenly from its origin to its tip. The 


body is subcylindrical, and nearly of the same thickness from the 


occiput to the tail, into which it passes almost imperceptibly. The 


_ head itself is rather small, subconical, rounded on the snout; it is 


continuous with the body, the neck being but slightly contracted. 
The plates of its upper surface are represented in figure 6, which 
will serve as a good term for comparison with the other species. 
The locomotive members are very short; the fore ones rather 
slender, the hind ones stouter. ‘The toes are terminated by deli- 
cate and slender nails, curved at their tip. The scales have a 


very smooth appearance; when examined attentively, those of the 


850 APPENDIX C.—REPTILES. 


back and tail, however, exhibit four or more longitudinal furrows 
or stripes on each of them. | 

The lower part of the head and throat is whitish, the belly 
bluish. Above, olivaceous brown, with two broad bands of black 
on each side, and perhaps two narrow ones on the back extending 
from the head to the anterior region of the tail. A whitish or 
perhaps yellowish stripe extends from the opening of the ear to 
behind the hind legs, and a similar one from the occiput to the 
origin of the tail. 

This species inhabits the same location in Oregon as Higaria 
scincicauda. The specimen figured, together with several others, 
were collected by Rev. George Geary, and sent by him to Dr. 
Avery J. Skilton, to whom we are indebted for their examination, 
and to whom we have dedicated it, as a slight acknowledgment of 
gratitude. 


Genus CHURCHILLIA, Baird and Girard. 


GEN. CHAR.—Three pairs of frontal plates; a very small loral, and several 
postorbitals. Scales carinated. 


In spite of the great uniformity in the plan of structure of 
ophidian reptiles, and especially of the genera Coluber and Tropi- 
donotus, we do not hesitate in separating generically from both of 
these, the species hereon described. 

The genus Churchillia has three pairs of frontal plates, one 
more than in both Coluber and Tropidonotus ; the middle pair, 
however, is subjected to some irregularities, by which, instead of 
one pair, there is only an odd plate. In either case it 1s a con- 
stant character proper to our genus. There is a small loral plate 
and several postorbitals. 

The scales of the upper part of the body are carinated, as in 
tropidonotus, but the lateral row, which is contiguous to the 
ventral shields, is scarcely larger than the other—while in Tropi- 
donotus, the scales composing that row are much the largest. 


CHURCHILLIA BELLONA, Baird and Girard. 


Spec. cHar.—Body yellowish, with a series of large subhexagonal patches of 
brown, bordered with black, and two or three rows of smaller patches on the 
sides. A brownish black band across the eyes, from top of head to the angle of 
the mouth. : ; 


This snake attains a considerable size. The tail, properly so 
called, is comparatively very short, forming only the ninth part 


en Lee eS 8 we 


APPENDIX C.—REPTILES. 351 


of the length. The head is moderate, conical, detached from the 
body by a somewhat contracted neck. The eyes are comparatively 
large, occupying the middle of the length of the head. There are 
two anterior orbital plates, the uppermost very large, vertically 
elongated, reaching the upper surface of the head; the lower one, 
on the other hand, is very minute, situated between the third and 
fourth labial shields. The postorbitals are four in number, of 
medium, but nearly of equal size, the upper ones, however, being 
slightly the largest. The lower one is separated from the small 
anterior orbital by the fourth labial shield, which reaches the orbit. 

The scales of the back and sides are keeled, elliptical in form, 
and a little more pointed posteriorly than anteriorly. On the 
occiput they are smaller, polygonal, and smooth. 

The coloration of this snake resembles at first glance that of 
Coluber eximius; the ground colour is a light yellowish-brown, 


-maculated with large patches of a deeper brown, margined with 


black, and much smaller patches of pure black. The dorsal row 
of brown patches is considerably the largest, as in Coluber eximius. 
On the abdomen there are two rows of small and semilunar black 
dots, the convexity of which is turned forward. A narrow band 
of black is seen on the upper surface of the head, in advance of 
the eyes, extending obliquely to the angle of the mouth, being 
only interrupted by the eye itself. 

‘This species was collected by General Churchill, on his march 
to Mexico, on the left bank of the Rio Grande, at the crossing 


_ near Presidio del Norte, in 1846; and it is with much satisfac- ‘ 


tion that we embrace the occasion to pay a tribute of respect to 
one, who, during the exercise of his arduous official labours, has 
always found time for the advancement of natural history, by 
securing specimens of whatever new or interesting species of 
animals might fall in his way. 


COLUBER MORMON, Baird and Girard. 


Spec. cHar.—Posterior frontal plates very large; vertical plate long and very 
narrow on its middle; eyes very large. 


The only specimen which we have seen of this snake may prove 
hereafter to be a young individual, as it is only one foot and a half 
long; but we are satisfied it will also prove to be a very distinct spe- 
cies. The slender and conical tail forms between a third and a fourth 
of the total length. The head is elongated and ovoidal, separated 


352 APPENDIX C.—~REPTILES. 


from the body by a contracted neck. The vertical plate is elon- 
gated and contracted on its middle. The posterior pair of frontal 
plates has almost twice the size of the anterior pair. The eyes are 
proportionally very large, protected below by the fourth and fifth 
labial shields, anteriorly by a large and a small orbital plate, and 
posteriorly by two of nearly the same size, if not of the same shape. 
The loral is proportionally very much developed. The nostrils are 
prominent. There are eight labial plates to the upper jaw and 
seven to the lower. The scales are smooth, elongated on the back, 
and posteriorly pointed, while on the sides they are broader, and 
the row contiguous to the ventral shield is composed of scales 
broader than long. On the posterior part of the head, as usual, 
the scales are subcircular, or polygonal, and much the smallest. 
On the tail they assume rather a lozenge shape. There are one 
hundred and seventy-five ventral shields from the throat to the 
vent. 

The colour is brown above and yellow underneath, with a bluish 
tint along the sides of the abdominal region. On the back there 
is a row of transversely elongated patches of deep brown, bordered 
with black. ‘The sides are spotted with three indistinct rows of 
small spots of brown and black intermixed. 

This species was found by Captain Howard Stansbury’s party, 
in the valley of the Great Salt Lake. 


e HETORODON NAsicus, Baird and Girard. 


Sprc. cuar.—Minute and numerous frontal plates instead of two large pairs; 
two brown stripes over the head; temporal patch very broad. 


The essential character which distinguishes this species from 
the H. platyrhinos and H. simus, which it most resembles, con- 
sists in the presence of numerous small scales between the vertical 
and rostral plates; that is, on the space which is occupied by the 
two pairs of frontal plates. In H. simus, it is true, very minute 
scales may be occasionally observed on the middle line between 
the vertical, rostral, and frontal plates; but when this is the case, 
the latter are not feaneed to the small s size which they have in the 
species which we here describe. 

The vertical plate in H. nasicus is circularly sibhesagamel 
The superciliaries are longer than the vertical, and thus longer 
than broad, while the occipital are broader than long. The nasal 


APPENDIX C.—REPTILES. Bs 


is proportionally large, and so is the lower loral plate, as there 
exists a second loral above the first, much smaller than any of 
those of the orbital circle. A specimen of H. simus examined by 
us has shown a second similar small loral on one side of the head, 
while on the other it evidently did not exist, so that it may prove 
either not to be constant in H. nasicus, or to be likewise found 
in other species of the same genus. ‘The rostral plate is directed 
upward, as in H. simus, having the same general shape as in the 
latter. ) | 

The tail has the same proportions with regard to the body as in 
Hi. platyrhinos, that is about the sixth of the entire length. The 
eyes are considerably smaller than in H. platyrhinos, as we have 
compared specimens of both species, having the same length and 
size of body. The only specimen of H. nasicus which we have 
seen is not a foot long, so that we have had to compare with it 
the young of H. platyrhinos. Now, if our H. nasicus be im- 
mature, it is the young of a species which differs from those already 
known of the genus, as shown by the very great difference in size of 
the eyes, and those other characters which we have just described. 

In coloration, our species resembles more that of H. platyrhinos 
than that of H. semus. The ground colour appears to be the same. 
There are in both species three rows of blackish-brown patches, 
bordered with a line of light yellow or white; but while these 
patches are elongated in H. platyrhinos, they assume a more circular 
shape in H. nasicus, although there is no regularity or sym- 
metry about them. ‘Two indistinct and much smaller rows exist 
on the sides. The lower part of abdomen and tail is almost en- 
tirely black. The throat and lower surface of head are unicolor, 
of the same hue as the ground colour of the body above. On the 
occiput, immediately behind the occipital plate, there is a lan- 
ceolated patch, and on both sides of this a much greater irregu- 
larly oval one. ‘Two brown stripes are seen on the top of the 
head, one over the anterior half of the eyes, extending vertically 
to the mouth, and another over the posterior half of the eye, hence 
obliquely backward, forming over the temples an elongated but 
much broader patch than in H. platyrhinos and H. simus. 

This species was collected in Texas by General Churchill: a 


Specimen is preserved at the Smithsonian Institution. 
23 


354 APPENDIX (.—REPTILES. 


A MONOGRAPHIC ESSAY ON THE GENUS PHRYNOSOMA. 


By CHARLES GIRARD. 


THE numerous specimens of nearly all the known species of 
this genus which are now preserved in the Smithsonian Institution, 
together with those at the Academy of Natural Sciences of Phila- 
delphia, have enabled us carefully to study and compare the 
different members of that most remarkable group, the result of 
which we propose here to present. 

Indeed, there are no genera in the saurian order that can so 
readily be distinguished as that of Phrynosoma. The body more 
or less circular in shape, always depressed, sometimes flattened, 
scattered all over with irregular and spine-like scales; the solid 
and subtriangular head provided with acute spines or tuberculous 
knobs, the short and conical tail covered with scales similar to 
those of the body, sometimes even more prominent, are as many con- 
spicuous features, which must strike any one at the very first glance. 
Their general aspect, perhaps their sluggishness, may recall to 
mind a frog or a toad: hence the vulgar name of horned toads or 
frogs. But the naturalist, with no hesitation, recognises in them 
true saurians, inasmuch as the body, instead of being smooth, like 
that of either toads and frogs, is covered, as just stated, with 
scales of a peculiar character. Besides the spines of the head, the ~ 
tail, although short, is another feature by which they disagree 
from both toads and frogs. So much when these animals are at 
rest: as soon as they move, the observer cannot fail to be struck 
with the fact that phrynosomas never jump or leap, as is the 
- case with the batrachians, to which they have been compared. 

If we look now more closely at the zoological peculiarities proper 
to the genus Phrynosoma we will see that the vertex is a promi- 
nent feature of the head, subtriangular or cordiform, with a sharp 
and projecting margin, forming a carina which overlaps the orbits; 
sometimes it is terminated posteriorly by two spines, one on each 
side. The occipital region generally presents the largest spines 
in those species in which these exist as a prominent feature. The 


APPENDIX (.—REPTILES. B50 


temporal region is very much developed and projects over the au- 
ditory aperture, and is provided along its projecting margin with 
spines or conical plates, the largest of which is most approximated 
to the occiput. The eyes seem as if situated in the middle of a 
groove extending from the snout to the occiput, on account of the 
projection of the superciliary ridge and temporal region. The 
lower jaw is generally bordered with a row or two of large plates, 
which vary in structure and shape according to the species, and 
furnish good discriminating characters between them. The snout 
is either truncated or acute. The nostrils are conspicuous, and 
situated near the extremity of the snout, either within the inner 
margin of the superciliary ridge, or on its direct prolongation. 
The upper surface of the head and sides not occupied by the 
spines or tuberculous knobs, are covered with small polygonal 
plates, varying in size according to the area over which they extend. 
The surface of these plates is rugose, wrinkled or keeled, as is also 
the surface of the spines themselves. ‘They are exceedingly sniall 
in advance and behind the orbits. The whole surface of the eye- 
lids is covered with minute scales of a granular appearance; the 
margin of the eyelid itself is ornamented with a double row of sub- 
quadrangular plates, a little larger than the granules of its surface. 
The lower part of the head from the chin to the breast is covered 
with small scales, characteristic in each species. The neck is 
generally very short, appearing as if contracted, the result of 
which contraction would be the presence of several folds of the 
skin, concealing the auditory apertures in conjunction with the 
temporal projection of the head. ) 
The scales on the upper surface of the body are very irregular 
in size and shape; on the neck, above and below, on the pits, 
along the sides of the back, and on the groins, they assume a gra- 
nular appearance, while along the middle of the back and on the 
tail they appear like thin lamellz, very irregular still, and cari- 
nated or subcarinated. All over the back, sides, tail, and hind 
legs, there are large, irregularly pyramidal scales, with an acute 
point and a wrinkled or carinated surface. The margins of the 
abdomen exhibit one or two horizontal rows of these pyramidal 
but soft scales, bent backward, extending from the fore legs to 
the hind ones. The species in which the scales of the back are the 
largest, is Phr. coronatum, which strikes every one by its rough ap- 
pearance; while those in which the scales are the least developed 
are Phr. platyrhinosand Phr. modestum, whose external appearance 


356 APPENDIX (C.—REPTILES. 


is in great measure destitute of that roughness which is generally 
associated with the idea of these reptiles. 

The abdominal scales are subquadrangular or lozenge-shaped, 
either smooth or slightly keeled, according to the species. On 
the breast and anterior portion of the shoulders several rows of 
the largest scales are seen, very prominent, very acute posteriorly, 
and strongly carinated or keeled. The anterior and upper portion 
of the thigh is likewise provided with large scales, but much less 
conspicuously keeled than on the shoulders. 

The tail is stout, always depressed at its base: it diminishes 
very rapidly posterior to the vent, and becomes cylindrical toward 
its tip. The pyramidal and raised scales are sometimes more con- 
spicuous on its sides and upper surface than on the body itself. 
The scales below the tail, in the vicinity of the vent, have the 
general appearance of those of the belly; in the post-anal groove 
some larger scales may occasionally be seen. Here the scales 
assume a subverticillated arrangement. On the conical portion of 
the tail they are carinated, while they are generally smooth about 
the vent. 

The anterior and posterior legs are nearly of equal size; the 
latter, however, are slightly stouter. The toes, five in number, 
are neither short nor very long; the first and fifth are the shortest 
and either of equal length, or the fifth may be a little longer; the 
second and fourth somewhat longer than the first and fifth, and 
likewise either of equal length, or the fourth somewhat longer 
than the second; the third is always the longest. The scales ex- 
tend all over the toes, overlapping even the base of the nails: they 
assume a subtriangular shape, with a very acute posterior summit, 
and a very distinct carina on their middle. The nails themselves 
are curved, compressed at the base, and very acute at the tip. 
On the inferior surface of the hind legs, along the thigh, a series 
of pores is observed, the femoral pores varying in number and con- 
spicuousness according to the species. ‘The anal pores are ‘ata 
absent in phrynosomas. 

We are thoroughly acquainted with six species of the genus 
Phrynosoma, viz. Phr. coronatum, Phr. cornutum, Phr. douglassit, 
Phr. modestum, Phr. orbiculare, and Phr. platyrhinos. 

Phr. harlant is identical with Phr. cornutum; Phr. wieg- 
mannt, with Phr. orbiculare; and Phr. blainvillet, with Phr. 
coronatum. | 

After a. mature examination of Prof. Holbrook’s description of. 


APPENDIX (C.—REPTILES. BOT 


Phr. orbiculare, we have been led to the belief that the species de- 
scribed under that name is neither the Phr. orbiculare of Wiegmann, 
nor any of the other just mentioned, and indicates a species the 
characters of which cannot be properly defined, since the original 
specimens were not preserved. The species is said to occur in 
Arkansas, Louisiana, and Texas, that is, within the geographical 
range of Phr.cornutum. Nowitis remarkable that among the nume- 
rous phrynosomas which have been received from these regions, all _ 
of them were found to belong to Phr. cornutum ; unless the species 
be very rare, this circumstance cannot well be accounted for. Phr. 
orbiculare is exclusively Mexican, and Phr. cornutum North Ameri- 
ean. If any other species be found with Phr. cornutum within 
the limits ascribed in the United States to Phr. orbiculare by the 
author of the North American Herpetology, we do not hesitate in 
pronounciny it distinct from Phr. orbiculare of Wiegmann. 

Dr. Wiegmann has indicated another species of the genus Phry- 
nosoma, under the name of Phr. bufonium, and Surinam was first 
given as its home. But in his «‘ Herpetologia Mexicana,”’ he is 
in doubt as to the locality whence that species comes. -Phr. 
bufoniwm is identified with Phr. cornutum by John Edward Gray in 
the catalogue of the British Museum. If Pahr. bufoniwm be an 
inhabitant of South America, we doubt the correctness of this 
identification. 

In the absence of authentic data in reference to Phr. bufonium, 
we would lay that species aside, and come back to the six ones the 
characters of which are well ascertained, and five of them repre- 
sented with great skill on the accompanying plates. The different 
views of the head of these species have been made in similar atti- 
tudes’in order to facilitate the comparisons. 

A glance at plate VIII, will show at once the specific oe 
between Phr. cornutum (fig. 1-6) and Phr. coronatum (fig. T-12). 
The profile, the position of the eyes and nostrils, the polygonal 
plates of the head, the direction of the spines, need scarcely to be 
alluded to, to render the differences apparent. The scales of the 
inferior surface of the head (fig. 8 and 9) exhibit a still more 
striking difference of form and arrangement in the two species. 
The femoral pores in Phr. cornutum (fig. 6) are but little conspicu- 
ous; they are much more so in Phr. coronatum (fig. 12), although 
the figure does not represent them as such. ‘The only specimen 
on hand at the time at which this figure was made being in a dried 
state, the fleshy parts had shrunk and the femoral pores were thus 


358 APPENDIX (C.—REPTILES. 


reduced and concealed. On specimens preserved in alcohol they 
are found as distinctly marked as in Phr. douglassi of plate VII, 
fig. 10. 

If we compare now Phir. douglassit with the two preceding ones, 
its rounded and anteriorly truncated head gives to it a peculiarly 
different aspect. The reduced cephalic spines constitute another 
feature quite as striking, especially when combined with the aspect 
of the upper surface (fig. 6). The lower surface of the head being 
covered with uniform scales, there is no possibility of mistaking it 
either for Phr. cornutum, which has a row of larger scales extend- 
ing from the chin to the throat, or for Phr. coronatum, in which the 
dissimilarity in the shape of these scales is still greater. 

Phrynosoma platyrhinos seems at first very similar to Phr. 
douglassiz, on account, no doubt, of the small development of the 
cephalic spines and the uniformity of the scales of the lower sur- 
face of the head. But we need only compare the vertex, the oc- 
cipital plates (fig. 1 and 6), the margin of the lower jaw (fig. 3 
and 9), the profile (fig. 2 and 7), and the position of the nostrils 
(fig. 4 and 9), in order to become satisfied of the specific distinction 
between these two species. 

The chief difference between Phr. modestum and Phr. platyrhinos 
are not to be found prominent in the head, but rather in the gene- 
ral structure of the body and tail. The specimen which we have 
had figured being a young one, these differences might appear too 
trifling. But recently we have received from Colonel J. D. Gra- 
ham, a series of full-grown individuals, by which it can be shown 
that this species, which we had distinguished from the others upon 
an immature specimen, appears still more distinct upon the ex- 
amination of the adults. In comparing attentively the figures 
which we now give of Phr. modestum, the differences will appear 
evident. The vertex is much more inclined forward in Phr. modestum 
(Pl. VII. fig. 2) than in Phr. platyrhinos, (P1.VI. fig. 6). The plates 
which line the margin of both the upper and lower jaws are likewise 
different, as well as the scales of the inferior surface of the head, 
which are proportionally smaller in Phr. modestum than in Phr. 
platyrhinos. | 

Of Phr. orbiculare we have seen only two specimens, and these 
were young individuals. But there is something so striking in its 
features as to enable us at once to distinguish it from its congener. 
Its snout is flattened, and the extremities of the jaws much more 
protruded than in any other species; it therefore differs greatly 


- APPENDIX. C.—REPTILES. 359 


jn that respect from Phr. platyrhinos, modestum, and douglassiz, 
with which it has in common the small and uniform scales on the 
inferior surface of the head. In the mean while, that same charac- 
ter distinguishes it from Phr. cornutumand coronatum. The femoral 
pores are very conspicuous. ‘The plates which line the jaws are 
proportionally much smaller than in any other species. 

The genus Phrynosoma is a truly American type. The distri- 
bution of its species over the continent is as follows :—Phr. orbicu- 
lare in the vicinity of the city of Mexico and in Sonora; Phr. cor- 
nutum, from Texas to Arkansas as far as the Rocky Mountains ; 
Phr. platyrhinos, in the valley of the Great Salt Lake; Phr. mo- 
destwm, in the valley of the Rio Grande del Norte ; and Phr. coro- 
natum and Phr. douglassiz, in Oregon and Caieema: the latter 
species extending as far eastward as the valley of the Grad Salt 
Lake, where it has been observed in company with Phr. platyrhinos. 

The division of phrynosomas into two groups, according to the 
position of the nostrils, whether situated within the internal margin 
of the superciliary ridge or at its extremity, would bring into one 
group Phr. orbiculare, coronatum, and douglassi, and into another 
Phr. cornutum, platyrhinos, and modestum. On the other hand, if we 
subdivide the species according to the shape of the profile, we 
would have on one side: Phr. orbiculare, cornutum, and coronatum, 
and on the other Phr. douglassit, platyrhinos, and modestum. Phr. 
douglassiz is the only species in which the cephalic spines remain in 
an undeveloped state. Phr. coronatum and cornutum have a double 
series of horizontal pyramidal scales on the periphery of the 
abdomen, while in Phr. orbiculare, douglassii, and platyrhinos, 
there is only one series of these, very small already in the latter, 
and totally absent in Phr. modestum. The difficulty of establishing 
subdivisions in this genus is thus plainly evident, and shows how 
natural and well circumscribed it is when considered as a whole by 
itself. 

We give now the diagnostic characters by which the six species 
of Phrynosoma may be distinguished. 

J. PHRYNOSOMA ORBICULARE, Wiegm.—Profile declive, tips of 
of jaws protruded, nostrils situated at the anterior extremity of 
the superciliary ridge. Occipital and temporal spines strong and 
well developed. One row of pyramido-horizontal and abdomino 
peripheral scales. Scales on the inferior surface of head, small, 
of a general uniformity, although irregular in shape. The plates 
on the margin of the jaws are inconspicuous, and very little larger 


360 APPENDIX (C.—REPTILES. 


than the others: the lower jaw is bordered with a double row of 
plates, the small size of which is the most striking feature in this 
species. The scales on the belly are subtriangular, posteriorly 
acute, and slightly keeled in the young individual under considera- 
tion; they are smooth in the adult. Femoral pores well developed. 
Lower surface. of the body maculated with irregularly elongated 
blackish spots. 

II. PHRyNosoMA cornuTUM, Gray.—(PI. VIII, fig 1-6.)—Same 
general profile as preceding. Nostrils situated within the internal 
margin of the superciliary ridge. Occipital and temporal spines. 
longer and more acute than in the preceding species (fig. 1, 2, 
and 4), A double row of pyramido-horizontal and abdomino-peri- 
pheral scales. Scales on the inferior surface of the head small 
and slightly keeled, of a general uniformity, except one row on 
each side somewhat larger, pyramidal, acute, slightly raised and 
directed horizontally outward and backward (fig. 3). <A series 
of very large, inframaxillary plates, sharp on their outer edge, the 
posterior one of which is transformed into a spine. The plates 
lining the margin of the jaws are not prominent. The scales of 
the belly are prop6rtionally small, subquadrangular, posteriorly 
very acute and keeled. Femoral pores undeveloped or else rudi- 
mentary (fig. 6). Inferior surface of the body, unicolor. 

III. PHRYNOSOMA CORONATUM, Blainv.—(Pl. VIII, fig. 7-12).— 
Profile declive; snout protruding; nostrils situated at the extre- 
mity of the superciliary ridge (fig. 7). Occipital and temporal 
spines perhaps less acute, but stouter than in Phr. cornutum (fig. 
8). A double row of pyramido-horizontal and abdomino-periphe- 
ral scales. On the lower surface of the head (fig. 9), there exist 
two double series (four rows) of quite large pyramidal, posteriorly 
acute, slightly raised scales, inside of which, two double series of 
much smaller scales are observed, the two innermost rows the 
smallest. The remaining portion of the lower surface of the head 
is covered with minute polymorphal scales. The plates of the 
inframaxillary row are strong, flattened horizontally, and very 
acute posteriorly: they are very much approximated to the margin 
itself of the jaw. The scales of the belly are of medium size, 
smooth, sometimes subtriangular, pointed posteriorly, while others 
are subquadrangular. Femoral pores very conspicuous, the series 
of either side coming nearly into contact on the middle line of the 
belly, forming there a curve, the convexity of which is turned for- 
ward. Lower surface of body maculated as in Phr. orbiculare. © 


Ackerman Lith 375 Broadway NY 


. CORONATA | Blainv. 


iL’ 


iT IN 
di 


PHRYNOSO 


Fig. 7-12. 


Gray, 


CORNUTA, 


MA 


@) 
U 


- 
« 
at 
* - 7 
t . 
. 
* 
. 
_ ‘ 
‘ 
Ne 
* 
oe 
(— 
~ 
5 


' APPENDIX C.—REPTILES. 361 


IV. Purynosoma poverassil, Gray.—(Pl. VII, fig. 6-10.)— 
Profile subtruncated (fig. 7). Nostril openings situated at the 
extremity of the superciliary ridge (fig. 9). Occipital and temporal 
spines reduced to blunt knobs (fig. 6 and 7). One row only of 
pyramido-horizontal and abdomino-peripheral scales. Lower sur- 
face of head (fig. 8) covered with small and uniform scales. Mar- 
ginal series of scales of the lower jaw large and conspicuous, flat- 
tened, with a sharp horizontal edge. An inframaxillary row 
likewise conspicuous in their size (fig. 7, 8, and 9). Scales of 
the belly smooth, acute posteriorly. Femoral pores very conspi- 
cuous (fig. 10). The series, from either side, are separated on the 
middle region of the belly by a free space. The lower surface of 
body appears as if unicolor, but, when examined carefully, nume- 
rous and crowded dots are seen. 

V. PuRYNOSOMA PLATYRHINOS, Girard.—(Pl. VII, fig. 1-5.) 
—Snout truncated, flattened, concave (fig. 2). Nostrils situated 
within the internal margin of the superciliary ridge (fig. 4). Oc- 
cipital and temporal spines of middle size (fig. 1 and 2). One 
row of pyramido-horizontal and abdomino-peripheral scales, much 
smaller than in the preceding species (compare fig. 1 to 6, on pl. 
VII). Lower surface of head covered with small, nearly uniform 
scales; on both sides, and near the neck, a series appears a little 
more conspicuous, slightly raised and acute. Inframarginal series 
of plates, large, of stout appearance, sharp and acute (fig. 3)y 
above which two rows of small plates are seen lining the margin 
of the lower jaw (fig. 4). Scales of the belly smooth and of mid- 
dle size. Femoral pores very conspicuous, but more apart than in 
the preceding species (fig. 5). A free space is left on the mid- 
dle region between the right and left series. The lower surface 
of the body is unicolor. 

VI. Purynosoma MopEstum, Girard. (Pl. VI, fig. 4-8.)—Snout 
truncated, flattened, but not concave (fig. 6). Nostrils opening within 
the internal margin of the superciliary ridge (fig. 7). Occipital 
and temporal spines of middle development (fig. 4 and 6). No 
pyramido-horizontal scales at the periphery of the abdomen. 
Lower surface of head with minute, generally uniform scales; a 
row lining the inframaxillary plates, distinguishing itself by its 
serial arrangement and its slightly increased size (fig. 5). Row 
of inframarginal plates resembling that of Phr. platyrhinos, 
(fig. 5); above which only one series of smaller plates is observed, 
lining the margin of the lower jaw (fig. 6 and 7). Scales of the 


362 APPENDIX (C.—REPTILES. 


belly proportionally larger than in any other species, smooth, 
quadrangular, and trapezoidal. Femoral pores conspicuous; the 
series from right and left coming into contact on the middle line 
of the belly, and forming a curve, the convexity of which is turned 
backward toward the vent. The lower surface of the body is uni- 
color. 


PHRYNOSOMA DOUGLASSII, Gray. 
Pu. VII. Fia. 6-9. 

Syn. Purynosoma povetassit, Gray, Synops. Rept. in Griff. Anim. Kingd. IX. 
1831, 44.—Waat. Nat. Syst. Amph. 1830, 146.—Wisem. Herp. Mex. 1834, 
54.—Hoxsr. N. Amer. Herp. I. 1842, 101, Pl. XIV.—Gnray, Cat. Brit. Mus. 
1845, 227. 

Agama douglassii, Brut, Trans. Linn. Soc. L. XVI. (1828), 1838, 105, Pl. X.— 
Haru. Med. and Phys. Research. 1835, 141, fig. 3. 


The specimen figured is not the largest of the species, but Piss 
the only one secured at Salt Lake, we took it in preference to any 
of those which we have from Galeria and Oregon. It is about 
three inches and a-half long; the head is a little less than a fifth 
of the entire length. The body, from the neck to the origin of 
the hind legs, is about of the same length as the remaining pos- 
terior portion. The body is suborbicular, being nearly as broad 
as long. These dimensions may vary to some degree among 
different individuals, but, generally speaking, they give to the 
Species its specific character as far as the form and outlines of 
the body are concerned. Viewed from above (fig. 6), the body is 
subtriangular, as broad behind as long; acuminated, but rounded 
in front. The vertex is cordiform, a little depressed in the middle, 
forming over the eyes a prominent carina, composed of four or 
five elongated plates, and posteriorly terminated by a blunt spine 
or acute knob. The occipital region, immediately behind, is de- 
pressed, and provided with three acute knobs, the central one in- 
conspicuous. The temporal region is bordered posteriorly by 
three or four spines or subpyramidal plates. The auditory aper- 
ture is vertically oblong or semilunated. The nostrils are pro- 
portionally very large, and situated at the anterior prolongation 
of the superciliary ridge. The polygonal plates are very minute 
on the nasal region and all along the upper jaw; the marginal 
row, however, being a little more conspicuous. The profile (fig. 7) 
exhibits the characters just alluded to, and gives a perfect picture 
of the physiognomy of this species. The eyes are oval. The same 
figure 7 exhibits the double row of large scales of the lower jaw. 


= 


Joy) 
Hine 
fev, 


= WAVES TE, 
Seesscrresesssst 
Se PORE, 


e) 
ee DO EP. 


Tueumaiszoy 


oS Git BY 


ak -eepe 


APPENDIX C.—REPTILES. 363 


Both of these rows meet in front of the symphysis of the jaw and 
behind the angle of the mouth, but leave between them a narrow 
area covered with very small scales. The inferior surface of the 
head (fig. 8) is very uniform; the scales which cover this region 
are small, some subovate, others subquadrangular. The pyramidal 
scales of the back constitute several irregular rows. One distinct 
and crowded row borders the outline of the abdomen. ‘The scales 
on the belly are smooth, subquadrangular or lozenge-shaped, a 
little larger on the middle than toward the sides. The femoral 
pores are quite conspicuous (fig. 10), from fifteen to seventeen on 
each side. 

The general hue of the specimen before us, preserved in alcohol, 
is olivaceous-green above, lighter below. Behind the occiput, an 
each side of the neck, there is an elongated patch of black, behind 
which a very small patch is seen, and, farther backward three or 
four others in pairs, triangular or quadrangular, the last of which is 
placed above and in advance of the hind legs. Four or five pairs of 
spots, diminishing gradually in size, may be followed along the tail. 
The dorsal patches are bordered with a line of light yellow. The 
abdomen appears unicolor, but, on careful examination, numerous 
brownish dots can be observed. The upper portion of the legs has 
a few small blackish spots. 

The specimen figured is from the valley of the Great Salt Lake, 
where it was collected by Captain Stansbury. 


PHRYNOSOMA PLATYRHINOS, Girard. 


Pu. VII. Fie. 1-5. 


The general form of this species is more elongated than usual, 
the neck not so much contracted, and the head, consequently, more 
detached from the rest of the body. The shape of the head is as 
much a characteristic of this species as its structure. We allude 
to its circular form when viewed from above (fig. 1), and to the 
flattening of the nose, as exhibited in the profile (fig. 2). The 
upper surface exhibits two strong occipital spines, and a row of 
smaller ones on the temporal region, five on each side. The poly- 
gonal plates of the cordiform and flat vertex assume a symmetrical] 
arrangement on both sides, so as to divide the space into two 
ovoidal areas, the outer row of which is composed of larger plates, 
while those enclosed are smaller and more irregular. Thus, the 


j 


364 APPENDIX C.—REPTILES. 


two posterior angles of the vertex, instead of being acute, as in 
Phr. douglassi, are rounded, without any prominent knob or spine. 
Along the superciliary ridge which overlaps the eyes, the plates 
are the largest, at least, five of them, there being two very small 
ones nearly above the middle of the eye. The occipital area 
exhibits three subcircular plates in its centre, larger than the sur- 
rounding ones. A group of large subspinous plates is likewise 
observed at the base and in advance of the occipital spines. The 
plates on the snout and along the upper jaw are small and incon- 
spicuous, except the marginal row, on account of an acute pro- 
jection of each of them. ‘The nose is flattened to excess, slightly 
concave, and the nostrils are situated at the inside of the super- 
ciliary ridge; thus placed in front (fig. 4). The eyes are cir- 
cular. The lower jaw wants the first row of large plates which 
we have seen in PAr. douglassii, but the one which exists is so 
much more developed than in the latter, and composed of six very 
conspicuous plates on each side and two small ones (fig. 2 and 3). 
Between this row of plates and the margin of the jaw there is an 
area, covered, on its anterior portion, with two, and, near the 
angle of the mouth, with three rows of small and polygonal scales. 
- On the lower surface of the head, from the chin to the throat, the 
scales are small and irregular; on the sides, however, and only 
for the posterior half of that distance, a row of from five to seven 
acute-edged scales may be seen. The folds of the neck do not 
exhibit any thing peculiar under the throat, but, on the sides and 
behind the ear, they are surmounted with pyramidal and raised 
scales. The auditory aperture is comparatively small—much 
smaller than in Phr. douglassit. The pyramidal and raised scales 
of the back are but little prominent. There is but one abdominal 
series of the latter, as in Ph. douglassit, originating behind the 
fore legs, but not extending so far backward as in others. The 
plates in the post-anal groove form two rows, the first composed 
~ of six, the second and posterior one of four only. The femoral 
pores are but few, from six to seven on each side, and quite dis- 
tant from each other. 

The head and tail are brown above, the upper part of the body 
ash-coloured ; yellowish and unicolor below. On the sides of the 
neck is a large patch of black, and two similar ones, but narrow 
and undulating, on the back. Faint indications of transverse 
bands of black are observed on the tail. 

Collected by Captain Stansbury about the Great Salt Lake. 


a 


a 


APPENDIX C.—REPTILES. 365 


PHRYNOSOMA MODESTUM, Girard. 
Pi. VI. Fic. 4-8. 


The most striking characters by which this species can be dis- 
tinguished from its congeners consists in the want of the periphe- 
ral abdominal row of pyramidal scales, and also in the slight de- 
velopment of the scales on the upper region of the body, which 
loses considerably that rough appearance so characteristic in the 
other species. The tail affords another character peculiar to Phr. 
modestum ; it tapers more suddenly, and is perfectly cylindrical be- 
yond the dilated portion of its base. The head has the same 
general appearance as in Phr. platyrhinos, although it is not so 
much truncated, and still less so in the adult than in the young 
(fig. 6). The vertex is more circular than in the latter, but the 
occipital and temporal spines do not differ much in the two species. 
The nostrils have the same position within the superciliary ridge, 
(fig. 7). The lower jaw exhibits one row of large plates, with one 
single series of very small ones above, forming the edge of the 
jaw, while we have seen two rows of the latter in Phr. platyrhinos. 
Below the head the scales are exceedingly small—much smaller 
than in Phr. platyrhinos. 'They are uniform among themselves, 
except a row of a little more conspicuous ones forming one series 
along the inside of the maxillary plates (fig. 5). The scales on 
the belly are smooth, subquadrangular, and larger than in any 
other species. The femoral pores are smaller, and even more 
apart than in Phr. platyrhinos, but there is no separation on the 
middle line of the belly, and the series from both sides meet in 
advance of the vent, forming a convex curve turned backward 
(fig. 8). 

The coloration is uniform yellowish-brown above, with two 
lateral patches of black on the sides of the neck. ‘Transversal 
and narrow bands of black are seen on the tail. The black spot 
seen on the left side of fig. 4 is accidental; that is, if exists on 
the specimen figured, but is not found on any others which have 
since come to hand. 

Brought from the Rio Grande, west of San Antonio, by General 
Churchill. A series of adult specimens were collected from San 
Antonio to El Paso, by the party under Colonel J. D. Graham, 

late of the United States and Mexican boundary survey. 


366 . APPENDIX C.—INSECTS. 


INSECTS. 


— 


By PROF. S. 8S. HALDEMAN. 


THERE were but few facilities for collecting, preserving, and 
transporting insects upon the journey, and those which were 
brought home are few in number and in bad condition. Under 
these circumstances, Captain Stansbury has allowed other species 
to be introduced, which have been collected by Lieutenant Horace 
Haldeman, U.S. A., chiefly at Fredericksburg and Fort Gates, on 
the western frontier of Texas; and by Mr. Richard Kern, in a 
journey across the plains to Santa Fé. In the latter case, the 
specimens were thrown into bottles of spirits as collected, so that 
there is nothing to distinguish those which were found upon the 
route from those of Santa Fé; but as the greater part probably 
appertain to the latter locality, this has been used to Mr. Kern’s 
species. 


LEPIDOPTERA. 


PAPILIO ASTERIAS, Cramer. 
A specimen with the patagia yellow, and forming a continuous 
lateral yellow line with the spots upon the thorax and head. © 
CYNTHIA CARDUI, Linnzus. 


A specimen of this species, which is common in Europe and the 
United States, and one of the most widely spread species known, 
occurring in India and Africa. On this continent it has been found 
among the Rocky Mountains and in California. 


PIERIS PROTODICE. 


DEILEPHILA LINEATA, Fabr. 
(—daucus, Cramer). Harris, Am. Journal Sci., vol. 36. 


Picea 


As a 


. APPENDIX (.—INSECTS. 367 


HYMENOPTERA. 


Among the most interesting entomological discoveries in the 
imperfectly explored parts of the United States territory, that of 
the three new species of the genus Labidus must be included. This 
is a tropical form, and almost exclusively confined to Brazil, the 
most northern species hitherto described being from the West In- 
dian Island of Saint Vincent. The three species were discovered 
at Fort Gates by Lieutenant Haldeman. 


LaBipus sad, Hald. 
Pu. IX. Fria. 1-3. 


Luteous, head brown above, and having (with the mandibles and 
basal articulation of the antennz) piliferous punctures; stemmata 
large, and the posterior ones twice as far from each other as from 
the eyes; face excavated below the antenne, with the lateral cari- 
nz sharp or angular, and the medial line impressed; mandibles 
large, tapering slowly, and curved from the base. Thorax convex 
and shining, with numerous piliferous punctures; dorsal line nar- 
row and distinctly impressed from the most prominent part to the 
anterior margin. Peduncle triangular, excised posteriorly with 
well-developed exterior angles. Abdomen indistinctly pubescent, 
with piliferous punctures posteriorly. Feet slender, simple, and 
uniformly coloured; base of the anterior tarsi excavated beneath for 
the tibial spur; ungues bifid; posterior feet extending beyond the 
abdomen; tibial spurs small. Wings with the stigmata long and 
narrow, posterior half indistinctly sanguineous, the nervures of 
the colour of the body, the membrane a pale tint of the same 
colour, and translucent. Length eight and a-half, wing seven and 
a-half, posterior tarsus two and a-half lines. 


LABIDUS HARRISII, Hald. 
Pu. IX. Fia. 4-6. 


Polished and pubescent, above black, abdomen rufous. Head 
black, pubescent; stemmata large, posterior ones three times 
farther from each other than from the eyes; antennz pale fulvous, 
base pilose; mouth rufous; mandibles pilose, robust, curved at the 
base only, the inner edge rectilinear, and the apex acute but not 


368 APPENDIX (C.—INSECTS. » 


incurved. Thorax convex in front, with piliferous punctures, 
brown, and pilose; wings very pale dusky, translucent, nervures 
pale yellowish-brown. eet slender and short, anterior tarsi 
medial line narrow and impressed; sides and feet dark reddish, 
curved at the base, the outer side being parallel with the internal 
excavation ; anterior tibial spine curved, and rather robust, poste- 
rior feet-not reaching the end of the abdomen. Peduncle of the 
abdomen black above, pubescent, transversely quadrangular, the 
anterior angles rounded. Abdomen compressed, thinly clothed 
with fulvous hair. Length five, expanse nine and a-half, posterior 
tarsus one line. 


LABIDUS MELSHAEMERI, Hald. 
Pu. IX. Fia. 7-9. 


Pale polished amber-coloured and pilose; head nearly quadrate, 
with the vertex black; the two posterior stemmata are distant, 
being nearly in contact with the eyes; mandibles pilose, sickle- 
shaped, curved from the base and diminishing rapidly to a slender 
incurved point. Wings very pale dusky, with the nervures pale 
brown; a black point upon the thorax at their insertion. Base of 
the anterior tarsi slightly curved; anterior tibial spine small and 
slightly curved; posterior feet not extending beyond the abdomen. 
Abdominal peduncle pilose, transverse, basal angles strongly 
rounded, apex concave; abdomen compressed. Length about 
three and a-half, wing three lines. 


AMMOPHILA ABERTI, Hald. | 


A large black and rufous species, the head and anterior wings 
of which are wanting in the only specimen collected. Thorax 
black cinereous primrose; patagia and feet (except the coxe and 
trochanters) rufous ; basal half of the posterior femora black, which 
extends in a line toward the apex upon the upper side; posterior 
tibial with the inner side darker than the outer side; posterior 
wings hyaline, nervures rufous. Abdomen rufous, peduncle and a 
blotch upon the apex above, black. Length fourteen, abdomen 
nine, to the constriction four lines. General form of Ammophila 
sabulosa. Named after Colonel Abert, chief of the Topographical 
Bureau, for his efforts toward the development of the natural 
history of the country, under various exploring expeditions. 


o 


iii 


APPENDIX (C.—INSECTS. 369 


HEMIPTERA. 


CrcaDA RE¢, Hald. 
Pu. IX. Fie. 17. 


Yellow, varied with black, sericeous beneath. Head yellow, 
pypostoma brown; the medial line yellow, and unimpressed, and 
the transverse ridges undulate; eyes connected by a broad trans- 
verse band. Pronotum yellow, with a narrow Y-shaped line 
divided to the base, a narrow transverse lateral spot on each 
side posteriorly, and another anteriorly, immediately behind the 
lateral stemmata. Mesonotum black, with a large lateral elon- 
gated yellow spot, and a pair of similarly coloured medial spots 
in the shape of the Hebrew letter resh inverted, and the points 
' converging anteriorly upon the medial line; tergum dark brown. 
Wings with the nervures yellow to beyond the middle, when they 
become dark brown or blackish. The usual W-shaped mark is 
present; beneath and feet yellow; metasternal spines rather large; 
spines of the feet and apex of the tibial tinged with brown. Length 


of the body fourteen, to the end of the upper wings twenty-two 


se “ a4 ate 
pt Pa eS 


lines; width of the prothorax seven lines. <A large and handsome 
species, from the Great Salt Lake Valley. 


CICADA STRIATIPES, Hald. 
Pu. IX. Fie. 16. 


Above black, varied with a little yellow; beneath yellow, more 
or less primrose, particularly beneath. Head black, with a small 
yellow spot above the antenne; hypostoma prominent, with the 
medial line yellow, and strongly impressed. Pronotum black, 
margined with yellow posteriorly, primose, and indistinctly lateral. 
Mesonotum black, with four small yellow spots, two connected 
with the scutel, and two central, one on each side of the medial 
line; lateral margins and scutel yellow, two raised yellow lines 
extending laterally from the latter. Tergum black, with the apex 
and margins of the segments yellow. LHlytra and wings with the 
nervures yellowish-white ; those of the exterior cells blackish; the 
basal portion, whichis doubled beneath in repose, is orange. That 
of the posterior alulet extending half its length and ending in a 


narrow fuscous band; base of the superior wings with a black 
24 


3870 APPENDIX C.—INSECTS. | 


point above. W spot near the apex wanting. Beneath, yellow; 
end of the haustellum a few points near the joints of the feet, and 
a transverse line at the base of the abdomen, black. Medial and 
posterior femora with an impressed stria along the inferior surface. 
Entire length thirteen, of the body nine and a-half, expanse of the 
wings twenty-three lines. This small species seems to be allied 
to C. rimosa of Say. It belongs to the section of C. septemdecim, 
in which the drums are exposed so as to render their action visible 
in the living insect. | : 


ZAITHA RETICULATA, Hald. 


Dark brown, haustellum stout, and curved, scutel longitudinally 
rugose, elytra with distinct raised reticulations; wings white, 
abdomen black, apex beneath paler, pectus varied with yellowish, 
and the external margin of the posterior femora of the same colour. 
Length eight and a-half, breadth four and a-half lines. This 
species is allied to Z. testacea and Z. aurantiaca of Leidy, (who 
described them under the generic name of Perthostoma,) in the 
Journal of Acad. Nat. Sci. of Philadelphia, 1847, p. 60; but the 
colour is deeper, and it is at once distinguished by the raised reticu- 
lations upon the elytra. 


ZAITHA BIFOVEATA, Hald. 
Pu. X. Fia. 1. 


Brownish-yellow, scutel and beneath darker; head much ad- 
vanced in front, haustellum very long, and curved nearly in a 
quadrant; antenne hairy, the three terminal articulations parallel, 
curved, and of equal length, the apical one thickest, but scarcely 
differing in shape. Pronotum punctate, the anterior two-thirds 
finely, and the posterior third more coarsely and confluently; a 
fovea without punctures upon each side, about a line from the 
anterior or lateral margins; scutel punctate, with the disk lon- 
gitudinally rugose. Feet maculate with brown; margin of the 
venter maculate with flavous. Length sixteen, breadth seven, 
head nearly three, haustellum three and a-half lines. Fort Gates, 
Texas. This is the largest species of the genus, but it has the 
characters of the antenne and rostrum, the long anterior coxe, 
slender feet, and terminal nervures of the elytra without anasto- 
moses, which distinguish this genus from Belostoma. 

In Belostoma, the wing when folded has four nervures (omitting 
that of the fold) which reach the margin; in Zaitha there are but 


4 


Mo(aus SURG SictloneAiy tocrddnidy ¢ —-sadtjei02 EpSadipeO 2 —eP]esAdsIQ PUIeE? | 


G vs i 


Aw peta bo Yu] LEW sy 


APPENDIX C.—INSEOTS. 871 


two, the intermediate ones being evanescent. ‘In the former, the 
first and second (from the fold) unite at the margin, but in the 
latter, the evanescent line representing the second reaches the 
margin parallel to the first nervure. 


ORTHOPTERA. 


EPHIPPIGERA TfIVAVENSIS, Hald. 
Pu. X. Fre. 8. 


Robust, dull brown, beneath yellowish; head rough, antenne 
deep set, filiform, shorter than the pronotum, inserted opposite the 
lower canthus of the eyes, upon each side of a double vertical 
frontal carina; labrum transverse, and with the palpi flavous. 
Pronotum ample, coarsely scabrous, blackish, changing to yellowish 
posteriorly; elytra and wings rudimentary, the former gray, 
mottled with black. Inside of the posterior femora and tibie 
black, the former interrupted near the apex. The posterior tibiz 
have a row of spines upon each above. Length fifteen, antennz 
Six, pronotum seven, width four and a-half, elytra four, posterior 
femora seven, and tibiz seven lines: Chihuahua. 


CEDIPODA CORALLIPES, Hald. 
Pu. X. Fie. 2. 

Yellowish-gray, conspicuously varied with brown, mostly in 
blotches, and upon the elytra and exterior side of the posterior 
femora. Vertex and pronotum scabrous and dark brown, the 
latter margined with flavous, its surface nearly flat, and the medial 
line but little raised. Angle of the elytra marked with a yellow 
line; a narrow line upon the internal margin. Wings bright yellow, 
margined with black. Inside of the posterior femora, tibize and 
tarsi, bright vermilion, a paler tint extending to the outside of 
the tarsi and lower half of the tibiz. Length twenty-four lines, 
(two and a-half inches,) pronotum five, posterior femora ten, and 
tibize nine lines. 

This fine large grasshopper is probably the species which has 
been destructive to vegetation in the Valley of the Great Salt 
Lake. It is nearly as large as the destructive dipoda migra- 
torta, (with which it is congeneric.) The last-named species is 
known under the English name of migratory locust. 


372 APPENDIX C.—INSECTS. 


ANABRUS, Hald. 


This new generic name is derived from the Greek abros, with 
the negative prefix an, in allusion to the unprepossessing appear-' 
ance of the insect. This genus has broad articulate tarsi, the 
soles concave, and the third articulation cordate. Itresembles Pha- 
langopsis in general appearance, the form of the head and labrum, 
the high position of the antenne, the narrow sternum, and the po- 
sition and probably the form of the feet. It has, however, a dis- 
tinct selliform pronotum extending over the basal articulation of 
the abdomen, and concealing rudimentary elytra. Excepting the 
tarsi, the posterior feet resemble those of a Phalangopsis and have 
the spines distributed in the same manner. ‘The ovipositor is 
nearly straight, sword-shaped, unlike that of Phalangopsis, and 
it is two-thirds the length of the body. A single specimen was 
brought from the Valley of the Great Salt Lake, and of this the 
antenne, palpi, and anterior and medial feet are wanting. 


ANABRUS SIMPLEX, Hald. 
Pu. X. Fia. 4. 


Dark shining brown, posterior femora with an external and 
internal row of small spines beneath upon the posterior extremity ; 
tibiz angular, with a row of spines upon each side above, and two 
approximate rows beneath with the spines alternating. Length 
fifteen lines, pronotum six, ovipositor twelve, posterior femora and 
tibiz, each eleven, and tarsi three and a-half. This seems to be 
one of the species which is eaten by the aborigines of the Valley 
of the Great Salt Lake. 


STENOPELMATUS Fuscus, Hald. 


Shining dark brown, abdomen darker. One specimen, seven 
lines long, is from Santa Fé; another is eight lines long, and from 
Chihuahua: the latter has the labrum and tibize darker than the 
general colour. 

It is possible that these small specimens may be immature, and 
they would not have been characterized but for the fact that in 
the allied genera the colour remains remarkably uniform during 
the various transformations, which renders it probable that they 
are not the larve of the rufo-testaceous Mexican S. talpa, which 
my brother has brought from Jalapa. ‘The tibial springs of S. 
Fuscus are well developed, a character by which the adult of 


pita ag 
8 
= 


APPENDIX C.—INSECTS. 373 


Phalangopsis can be distinguished from the pupz of the same size 
and general appearance. 
PHALANGOPSIS. 


There is a larva of this genus in the collection from the Valley 
of the Great Salt Lake, which bears a close resemblance to P. 
lapidicola. 


COLEOPTERA. 


CICINDELA AUDUBONII, Le Conte. 


_ Ann. Lyceum, N. Y.. Valley of the Great Salt Lake and 
Sante Fé. 


PANAGAUS DISTINCTUS, Hald. 


Rufous, sparsely hirsute, elytra with a fascia behind the middle, 
interrupted at the suture, apex blackish. Head scabrous and 
rather large, prothorax coarsely punctured, wider anteriorly than 


posteriorly, sides regularly rounded, posterior angles small and 


acute; dorsal line and region of the posterior angles impressed; 
elytra striate, with large impressed punctures. Length, three and 
a-half lines. Colour and general appearance of P. fasciatus, but 
smaller, with a larger head, the prothorax very different in shape, 
and the elytra more strongly marked. Sante Fé. 


CARABUS FINITIMUS, Hald. 


Allied to @. sylvosus, but somewhat wider. Blank slightly 
bluish on the margin. Prothora, less narrowed behind than in 
sylvosus, with the reflex margin and that of the elytra wider. 


_ Elytra more convex, very shining, finely punctured in lines, with 


three rows on each of distant impressed foves. Destitute of the 
scabrous appearance of C. sylvosus. Fort Gates. 


PANGUS CALIGINOSUS. 
DICHLUS SPLENDIDUS. 
CALOSOMA SCRUTATOR. © 
PLOCHIONUS TIMIDUS. 
Occurring at Fort Gates. 


374 APPENDIX C.—INSECTS. 


STETHOXUS TRIANGULARIS, Say. 


At Tampico. 


MELOPS CONSTRICTUS. 
CHLANIUS SERICEUS. 
c VICINUS. 
AGONUM ERYTHROPUM. 
P&CILUS SCITULUS. 


Were collected by Mr. Kern. 


TROCHALUS EXPLANATUS, Le Conte. 


Or a closely allied species, was brought by Captain Stansbury. 


COTALPA GRANICOLLIS, Hald. 
Pu. IX. Fig. 11. 


Hirsute, dark green, elytra reddish castaneous, feet black. 
Head, pronotum, scutel, and pygidium, green, and densely and 
confluently punctured. Elytra each with four indistinct impressed 
striz; surface irregular punctured, punctures confluent toward 
the sutures. Under parts, including the femora, green, and 
densely punctate; tibie and tarsi, black. Length, eight lines. 
The body, above and below, with the elytra and feet, are irregu- 
larly hirsute, with whitish hairs. Valley of the Great Salt Lake. 


EUPHORIA CERNII, Hald. 
Pu. IX. Fie. 10. 


Dark brown elytra, varied with pale flavous. Head and pro- 
notum densely and coarsely punctured; antenne, palpi, and feet 
reddish-brown. Pronotum margined laterally and posteriorly 
with a narrow band of flayous; scutel black, with elongated shal- 
low punctures, mostly toward the margin. LElytra each with a 
wide sutural and medial groove, irxegularly maculated with dark 
brown and flavous in nearly equal proportions, the lateral and 
terminal with three sub-sutural spots; surface slightly punctate 
with longitudinal striz near the suture,.and fine transverse rugo- 
sities laterally. Pygidium faintly rugulose, pectus blackish. 
Length, five lines. Collected by Mr. Kern, after whom it is 
named. 3 | 


' APPENDIX C.—INSECTS. 375 


MELOLONTHA DECEMLINEATA, Say. 
Valley of the Great Salt Lake. 


PELIDNOTA TRIPUNCTATA. 
Fort Gates. 


AREODA LANIGERA. 
Santa Fé. 


APHODIUS STRIGATUS, Say, which is common in the United 
States, occurs in Western Texas and as far south as Jalapa, in 
Mexico. 


Hyprocuus FovEATUS, Hald. 


Silvery-gray, varied with curious reflexions. Head coarsely and 
densely punctate; the largest punctures between the eyes; palpi 
flavous, prothorax nearly quadrangular, widest before, produced 
posteriorly into an obtuse angle; surface punctate with three 
foveze across the middle, the central forming a triangle with two 
others in contact with it posteriorly. Elytra each with ten rows 
of deep dilated punctures. Feet flavous, varied with brown; 
length less than two lines. Fort Gates, Texas. Allied to H. 
scabratus of Mulsant, 1844. Ann. Sci. Phys. Nat. Lyon, vol. vii. 
p. 873. H. gibbosus, Melsheimer. 


STAPHYLINUS VILLOSUS, Cray. 


Mr. Kern brought a specimen of the species, which is common 
in the United States, and is found in Mexico and Cuba. 


PHILONTHUS comptus, Hald. 
Allied to P. eneus and P. harrisit. Polished black. Head 


rather narrower than the prothorax, with foveze in a transverse 
line, the external one orbital and placed before the middle of the 
eye; the next midway between this and the medial one, which is 
more shallow than the others, and in advance of them. There 
are three additional orbital fovese posterior to the first, and several 
upon the posterior angles, which are strongly and sparsely punc- 
tate. Mandibles strong and incurved, with a stout tooth near the 
base, external margin with a groove for about half its length. 
Pronotum with four distant punctures arranged longitudinally 


3876 APPENDIX C.—INSECTS. 


upon each side of the middle, a second line of three exterior to 
these, (two placed opposite the interstices of the posterior three, 
and the third at the posterior margin,) a third row of three or 
four exterior to the latter, followed by a single puncture in the 
lateral angle: there are also several marginal punctures poste- 
riorly. Scutel flat, oblong, triangular, with piliferous punctures. 
Elytra longer than the prothorax, widest posteriorly, and rather 
wider than long; black with piliferous punctures, lateral margin 
yellowish brown, hair yellowish upon the disk, and fulvous upon 
the lateral margin. Wings fuliginous irised. Tergum, under parts 
and feet with piliferous punctures. Length, six lines. A single 
specimen collected by Mr. Kern, of which the antenne are broken. 


NEcROPHORUS OBSCURUS, Kirby. 
Fauna Bor. Amer. p. 97.—Valley of Great Salt Lake. 


ELEODES cogNATA, Hald. 


Colour, size, and markings as in H. extricata, but the punctur- 
ing is much finer, and that of the pronotum more sparse. The 
elytra have distant, minute elevated points, (some of them con- 
nected with the punctures,) which are more evident posteriorly. 
Valley of the Great Salt Lake. 


ELEODES OBSCURA, Say. 


Two specimens in Captain Stansbury’s collection. 


NyctToBATES (IPHTHINUS) INTERMEDIA, Hald. 


Allied to WV. barbata, Knoch (Tenebrio) Neue Beytraege, p. 166, 
fig. (striato-punctatus, Dejean, Catalogue, p. 225) and similarly 
barbate with fulvous hair. Punctures of the head larger and 
more crowded (especially upon the vertex.) Pronotum with the 
sides more rounded and the surface more coarsely punctured. 
Surface of the elytra minutely but more distinctly punctured, and 
the nine lines of punctures less distinct than in WV. barbata. 

This species agrees in size and colour with JV. barbata, and in 
distinctness of the lines of punctures upon the elytra it stands 
between that species and WV. pennsylvanica, Western Texas. 


ZOPHERUS VARIOLOSUS, Sturm. 


Described from Mexican specimens ; is found at Fort Gates. - 


APPENDIX C.—INSEOTS. 877 


Horia STANSBURII, Hald. 


Black, elytra sanguineous. Head pilose, and with the pro- 
thorax scabrous with confluent punctures. Pectus shining, and, 
with the feet, punctate. Abdomen of the female with a small fovéa 
upon the middle of the three terminal segments. Hlytra scabrous, 
with large irregular confluent punctures. Length of the male four, 
of the female six lines. Valley of the Great Salt Lake. 


MELOE PARvus, Hald. 


Black, somewhat shining, head with numerous dilated punctures 
extending to the labrum, eyes uniform, a smooth fovea near them, 
and a smooth pustule between the fovea and the insertion of the 
antenne. Antenne eleven-articulate, filiform, extending a little 
beyond the prothorax, and apparently alike in both sexes; the 
second articulation is the smallest, and the third equal to the two 
following; beyond the third the length gradually increases to the 
apex. Prothorax subquadrate, but the anterior angles are rounded 
and the middle advanced so as to form a neck; posterior angles 
raised in an obtuse pustule; dorsal line interrupted, impressed 
before the middle, and forming a rima posteriorly. Scutel punc- 
tate with the disk impressed. LElytra scabrous and acute. Ab- 
domen above and below with minute piliferous punctures. Feet 
slender. Length from four to five lines. Kern’s collection. 


Henovs, Hald. 


Form of Hpicanta, with the elytra abbreviated, connate, and each 
obtusely rounded. Prothorax subglobular, lengthened anteriorly. 
Abdomen, with the middle part above, coriaceous, and the lateral 
parts membranous; that of the female inflated. Antenne setace- 
ous, third articulation longest; from the third to the sixth slightly 
dilated and compressed in the male. Ungues cleft, with the parts 
equal. 


HENOUS TECHANUS, Hald. 
Pu. IX. Fie. 12-14. 
Black, short, pubescent, minutely granulate. Head, thorax, and 


elytra scabrous, with confluent punctures; labrum and clipeus 
with dilated impressed punctures, more crowded upon the labrum 


378 APPENDIX C.—INSEOCTS. 


and sides of the clipeus. Length of the female ten, breadth three 
and a-half, length of the elytra nearly five lines. 


ELAPHIDION MARILANDICUM. 
CLYTUS IRRORATUS. 

7 FLEXUOSUS. 
CERASPHORUS GARGANICUS. 


Fort Gates, Texas. 


MEGADERUS CORALLIPES, Newm. 
Pu. IX. Fig. 15. 


Was described from an imperfect Mexican specimen by Newman, 
(Charlesworth, Mag. Nat. Hist., vol. iv., p. 195.) It is now figured 
for the first time, from a specimen taken at Fort Gates. 


APPENDIX C.—INSECTS. 379 


LETTER FROM T. R. PEALE, ESQ., UPON THE LARVA OF INSECTS 
FOUND IN THE GREAT SALT LAKE. 


Wasuineton, May 12th, 1852. 


My Dear Si1r:—The exuvie of insects which you have brought 
from the shores of the Great Salt Lake proves, on examination, to 
have been deposited by aquatic diptera. 

In the mass, I can detect fragments of the larve shells of the 
pupa, and small portions of a mature Chironomus and other 
Tipulide. More than nine-tenths of the mass is composed of 
larvee and exuviz of Chironomus, or some species of mosquito— 
probably undescribed; the fragments being too imperfect to deter- 
mine. 

You are best able to determine, first, whether mosquitoes exist 
at any time at the Great Salt Lake in such unparalleled numbers 
as this organic matter indicates; or, secondly, whether the salt of 
the lake water has preserved their exuvise, so that it has accumu- 
lated through a great length of time. 

A few fragments of insects I have been able to determine as 
belonging to the Linnean genus Nepa, which is aquatic, 7 a 
very few others as Hymenopterous, &e. 

In the hope of soon seeing your Report o: on the most interesting 
portion of our continent, 

I remain 
Yours truly, 


T. R. Peas. 


Captain H. STANSBURY, 
Corps Topographical Engineers, 
Washington. 


I am not aware that mosquitoes exist in such unusual abundance 
in the vicinity of the lake; but incline to the opinion of Mr. Peale, 
that the accumulation of the immense masses of these exuviz is to 
be attributed to the preservative qualities of the lake water. 


H. 3: 


a 


4 


K 
. 
ee 
be 
- 
x 
13 
Nhe 


y 


4 


5 
<i 
r 
‘ 
‘ 
fy 
A, 
t 
. 
‘ 
' 
Cah eeihrer eat arcies 


J hie wd 
A : hy ewe mie 


ee hy OR LE 


ei 
é 
dy by aay 
Awe et t 


Me tae 


APPENDIX D. 


BOTANY. 


* CATALOGUE OF PLANTS COLLECTED BY THE EXPEDITION. 


BY PROFESSOR JOHN TORREY. 


P 


7 
-_ 


APPENDIX D. 


BOTANY. 


BY JOHN TORREY. 


CLEMATIS LIGUSTICHFOLIA, Nutt.—East base of the Black Hills. 
In fruit, September 29th. ‘Tails of the carpels more than an inch 
long, and very slender. 


ANEMONE PENNSYLVANICA, Lin.—Great Salt Lake Valley. 


DELPHINIUM AZUREUM, Mich.—With the preceding. Fl. May 
2d—19th. 


BERBERIS (MAHONIA) AQUIFOLIUM, Pursh.—With the preced- 
ing; on the sides of the mountains. Fl. May 19th. 


ARGEMONE HISPIDA, Gray, Plant. Fendl., No. 16.—With the 
preceding. Called the «Thistly plant’? by the inhabitants. In 
fruit May 19th. — 


VIOLA PEDUNCULATA, Torr. and Gray.—Borders of the Salt 
Lake. 


CorYDALIS AURBA, Willd.—Stansbury’s Island, Great Salt Lake. 
Fl. June 26th. 


ERYSIMUM ASPERUM, D. C.—Shore of the Salt Lake and along 
Weber’s River. May—June. 


STREPTANTHUS CRASSICAULIS, Torr. (Sp. nov.)—Glaucus; cdule 
glabro inflato fistuloso; foliis oblongis runcinato-pinnatifidis vel run- 
cinatis longe petiolatis; floribus erecto-patulis; petalis (purpureis) 
linearibus obtusiusculis calyce villoso-lanato duplo longioribus. 

Mountain side, on the east shore of the Salt Lake. Fl. May 80. 
Found also on the tributaries of the Uintah River, Utah Territory, 
by Colonel Fremont. Annual. This species is easily distinguished 


by its inflated hollow stem and very woolly calyx. The leaves are 


383 


384 - APPENDIX D.—BOTANY. 


mostly radical and deeply pinnatified; the terminal lobe much 
larger than the others, and triangular or deltoid. The stem is 
simple, from one to two feet high, more or less inflated toward the 
base, and nearly naked above. ‘The flowers are nearly sessile, in 
a long terminal raceme, erect when first expanded, but finally 
becoming patulous. Calyx about half an inch long, the sepals 
oblong-lanceolate and woolly externally. The petals are dark pur- 
ple, with a pale waved margin. Filaments all free. The siliques 
are not known. 

Plate I. Streptanthus crassicaulis, of the natural size. Fig. 1, 
a sepal, showing the inner face and part of the hairiness on the 
back. Fig. 2, a petal. Fig. 3, the stamens and pistil. Fig. 4, a 
separate stamen. All magnified. 


S. sagittatus, Nutt. in Jour. Acad. Nat. Se. VIL, p- 12; not 
Hook and Arn.—Shore of the Salt Lake, May 6. 


SISYMBRIUM CANESCENS, Nutt.—West shore of Salt Lake. — 


PHYSARIA DIDYMOCARPA, Gray. PI. Illustr. I., p. 162, (in a 
note.) Vesicaria diguclael, Hook.—On Green VEN In fruit, 
September 12th. 


CLEOME LUTEA, Hook. Fl. Bor. Amer. L., p. 70, t. 25. @. pe he 
Nutt Carrington’ s Island, Salt Lake. FI. Tale 18. 

Except in the greater aati of the stipe and the large size of 
the plant, I see nothing to distinguish C. aurea of Nuttall from this 
species. 


«SIDALCEA MALVHFLORA, Gray, mss. SS. orogana, Gray, pl. 
Fendl., p. 20. Stda malveflora, Lindl. WS. orogana, Nutt.— 
Antelope Island, Salt Lake. FI. June 18-30. A white-flowered 
variety Se in the same locality. 


MALVASTRUM COCCINEUM, Gray, Gen. Ill. t. 121, pl. Fendl. p. 
24. Cristarta coccinea, Pursh. Sida coccinea, D. C., Torr. and 
Gr., fl. 1, p. 682. 3 | 


Tae B GROSSULARIHFOLIUM. M. grossularicefolium, Gray, l.c. 
Sida grossularicefolia, Hook. and Arn. —Islands and shore of the 
Salt Lake. May and June. 

Except in the larger size of the plant and in the less divided 
leaves, the var. 3 does not differ from the ordinary form of I. 
coccineum. 


CALLIRRHOE INVOLUCRATA, Gray, Gen. Ill. 2, t. 117; Pl. 


Y, A 


BN 


Adrerman lath 379 BroacdwayNY 


4 vG 


SPREE RANTHUS CRASSICAULIS, Torre Frew. 


—. . | 
- —"Fiicemmnim Lath 079 Broadway t 


PHACA MOLLISSIMA 8. Torr 


APPENDIX D.—BOTANY. 885 


Fendl. p. 16. Malva involucrata, Torr. and Gr. FI. By py 226. 
Upper waters of the Platte. The large tapering root is said to be 
edible. 


VictA AMERICANA, Muhl.—Valley of Salt Lake, June 1. 


CicER ARIETINUM, Lin.—Sandy bottom land in the Valley 
of Salt Lake; probably introduced. This plant has also been 
found by Dr. Pickering on the banks of the Kooskooskee, or Clear 
Water, in Oregon; and I have received it from Southern California, 
where it was doubtless taken by the Spaniards. It is a little 
remarkable that it should now be found apparently wild in the 
interior of Oregon and in the valleys of Utah. 


PHACA MOLLIsSIMA, Nutt. in Torr. and Gr. Fl. 1, p. 350. 
Astragalus purshii, Dougl. in Hook. Fl. Bor., Amer. 1, p. 152. 


Var. @ UTAHENSIS; foliolis. 6-8, jugis, obovatis; pedunculis folio 
longioribus. Shores and islands of the Salt Lake. This plant 
is abundant in the Territory of Utah, and I have not received it 
from any other region. It differs from the ordinary form of P. 
mollissima: and if there were not what appear to be intermediate 
states of it, 1 should consider it-a distinct species. It is less 
branched, and has more numerous leaflets than the var. @. The 
flowers are violet, four to six in number, in a short spiked raceme. 
The nearly mature legume is densely clothed with long woolly 
cream-coloured hairs, and very closely resembles that of P. mol- 
lissima. Our plant has much the appearance of Astragalus gla- 
reosus, Dougl. (A. argophyllus, Dougl.,) and which, I suspect, is 
a Phaca, but the leaves and fruit are different. 

Plate Il. Phaca mollissima, var. utahensis of the natural size. 
Fig. 1, a flower. Fig. 2, the wings and heel. Fig. 3, the 
stamens. Jig. 4, mature fruit of the var. q Fig. 5, cross 
section of the same. Fig. 6, immature fruit of var. utahensis. 


ASTRAGALUS ADSURGENS, Pall. ?—West shore of the Salt Lake, 
in sandy soil. Flowers white, shaded with purple. This plant 
seems intermediate between A. adsurgens and A. striatus, Nutt. 
The legumes were not found. May 1. | 


OXYTROPIS LAMBERTI, Pursh.—Upper waters of the Platte, &c.; 
frequent. 


HEDYSARUM MACKENZII, Richards. App. Frankl. Journ. ed. 2, 
p. 28.—Promontory Range, Utah. FI. May 1. 
20 


386 APPENDIX D.—BOTANY. 


LUPINUS ALBICAULIS, Dougl.?—High grassy land, Antelope 
Island, Salt Lake. Fl. June 30. A suffrutescent species densely 
clothed with short appressed almost silvery hairs. The leaflets 
are mostly in sevens, oblanceolate and acute. The flowers are 
nearly as large as in L. perennis, in rather dense, somewhat ver- 
ticillate spikes; and the upper lip of the calyx is strongly soccate 
or slightly spurred. 


CowANIA STANSBURIANA, Torr. (Plate III.) OC. foliis pin- 
natifido 5—7-lobatis, lobis oblongis; floribus flavis. C. plicata ? 
Torr. in Frém. 2d Report, p. 314; not of Don. Stansbury’s 
Island, Salt Lake. Colonel Frémont collected this plant in the 
mountains of California, along the Virgin River, a tributary of 
the Colorado. It is nearly related to C. mexicana, Don, (in Linn. 
Trans. 14, p. 574, t. 22, f. 1,) which has also yellow flowers; but 
the leaves in that species are three parted, with linear segments, 
and they have a long narrowly cuneate base. 

A third species of this genus, C. plicata, Don, was introduced 
into England from Mexico in 1835, and figured in Sweet’s British 
Flower Garden, (t. 400.) ‘This is clearly the plant afterward de- 
scribed and beautifully figured by Zuccarini in his Plant. Nov. v. 
minus cognit, under the name of Cowania purpurea. It is also 
Greggia rupestris of Englemann, in Wislizenius’s Jour. 

The C. stansburiana is a shrub attaining the height of from 
six to twelve feet. It is much branched, and the young twigs are 
glandular. The leaves grow mostly from short spurs. They are 
ovate in outline, 4-6 lines long, deeply cut into five or seven 
lobes, and whitish tomentose underneath, except the strong green 
midrib, but green and somewhat glabrous above. They are revo- 
lute on the margin, of a coriaceous texture, and sparingly dotted 
with conspicuous glands. The flowers are solitary, terminal, and 
‘on short peduncles. The calyx-tube is turbinate and glandular ; 
the segments are broad and obtuse. Petals sulphur-yellow, broadly © 
obovate, two or three times the length of the calyx-segments. 
Styles persistent, beautifully plumose, and in fruit an inch or more 
m length. Achenium linear-oblong, striate, and clothed with short 
appressed hairs. For further remarks on the genus Cowania, see 
Plante Fremontianze, in the Smithsonian Contributions, vol. 5. 

Plate III. Cowania stansburiana; a branch of the natural 
size. Fig. 1, a leaf of the natural size. Fig. 2, upper surface 
of a leaf magnified. Fig. 3, under surface of the same. Fig. 4, 


vio Broadway We 


o 


n Lith 


~ Ackcermes 


COWANTA 


Acleeriman. Lith 379 Bro adwad NY 


| ppd 4 7 


LV RTO it 


- i= 


ambien T , 
fen j 
Th si ; 
~ Sz wranodn 4% it 3), baw 
3 
. re at 
Wh ig heoR ET vireO boy gtd bance 


¥ 
soma earth. Apia begs 


RECARO 


TA PH RUAIRT } aE 
388 A ‘SHV se > 


an 


eco ae 
. ah 4 eg it wilt Y. A 


ite 
ae 


APPENDIX D.—BOTANY. , 387 


a flower-bud. Fig. 5, a flower laid open. Fig. 6, a petal. Fig. 
7, plan of the flower. Fig. 8, a pistil. Fig. 9, front view of the 
style and stigma. Fig. 10, side view of the same. Fig. 11, a 
carpel of the natural size. Fig. 12, the same magnified. Fig. 13, 
a stamen seen in front. Fig. 14, the same seen from behind. 
Fig. 15, longitudinal section of a ripe carpel, showing the erect 
seed. Fig. 16, transverse section of the same. All the figures 
except No. 1 are more or less magnified. 


SprrEA DuMoSA, Nutt. Mss.; Hook. Lond. Jour. Bot. 6, p. 217; 
Gray, pl. Fendl. p. 40. S&S. discolor, Torr. in Ann. Lyc., N. York, 
2, p. 195; not of Pursh.—Stansbury’s Island, Salt Lake. Fl. 
June 26. 


Plate IV. Spirewa dumosa; a branch of the natural size. Fig. 
1, the fructiferous calyx. Fig. 2, a carpel. Fig. 3, the same 
laid open. 


S. OPULIFOLIA, var. PAUCIFLORA, Torr. and Gr. Fl. 1, p. 414.— 
Summit of a mountain on Stansbury’s Island, Salt Lake. Fl. June 
26. A tall, much branched shrub, with leaves scarcely more than 
half an inch in diameter. 


CHNOTHERA CHSPITOSA, Nutt.—Shore and islands of the Salt 
Lake. May and June. Usually acaulescent, but sometimes throw- 
ing up a branching stem about six inches high. The flower is 
from two to three inches in diameter, white and fragrant. WS. 
montana, of Nuttall, is hardly distinct from this species, and, 
perhaps, @. marginata should be regarded as a variety of the 
same. 


CH. scaporpEA, Nutt. in Torr. and Gr, FI. 1, p. 506.—Western © 


shore of the Salt Lake. Fl. and fr. May. 


CH. ALBICAULIS, Nutt.; Torr. and Gr. Fl. p. 495.—Islands of 
the Salt Lake. Fl. June. Stem about a foot high; the flowers 
small, white, and inodorous. 


GAYOPHYTUM RAMOSISSIMUM, Torr. and Gr. Fl. 1, p. 513.— 
Antelope Island, Salt Lake. Stem about eighteen inches high, 
with very slender branches, and flowers even smaller than in Mr, 
Nuttall’s specimen of this plant. The pedicles are about twice as 
long as the ripe pod. 


MENTZELIA ORNATA, Torr. and Gr., and Gray, Pl. Fendl. p. 47. 
Bartonia ornata, Nutt.—Islands of the Salt Lake. In our speci- 


388 APPENDIX D.—BOTANY. 


mens there are only five petals; and the filaments of the five 
outermost stamens are only a little dilated, while the anthers are 
perfect: but in other specimens, collected by Colonel Frémont, 
there are ten petals, of which five inner ones are rather smaller 
than the others; and so they are described by Mr. Nuttall. Sir 
William Hooker thinks that MZ. levicaulis is not distinct from 
this species; but Dr. Gray states (1. c.) that it differs in its yellow 
flowers, which open in the sunny hours, while in M. ornata they 
are white, and open toward sunset. 


M. ALBICAULIS, Dougl.; Torr. and Gr. 1. c.—Valley of the 
Salt Lake. : 


Eropium cricutariumM, L’Herit.—Islands of the Salt Lake. 
Fl. June. This plant is widely spread over the western part of 
North America, from the Rocky Mountains to the Pacific, and is 
doubtless indigenous. 


HEUCHERA RUBESCENS, Torr. (sp. nov.)—Scapo nudo glabro vel 
scabriusculo; foliis suborbicularibus breviter 5—7-lobatis glabrius- 
culis, lobis crenato-dentatis, dentibus setoso-mucronatis, vel obturis ; 
panicula oblongo; thyrsoidea sublaxa; staminibus exsertis; petalis 
linearibus calyce zquali longioribus. 

Stansbury’s Island, Salt Lake. Fl. June 26. Rhizoma thick 
and somewhat ligneous, clothed with brown vestiges of leaves. 
Leaves an inch or an inch and a-half in diameter, nearly orbicular, 
mostly cordate at the base, somewhat coriaceous, either wholly gla- 
brous or very sparingly strigose-pubescent, moderately 5—T-lobed, 
and the lobes crenate, or broadly toothed. The teeth usually 
mucronate and sometimes ciliolate. Petioles 2-4 inches long. 
Scapes varying from a span to fifteen inches high, entirely naked, 
except a few remote appressed scales. Panicle rather loose and 
few (15-20) flowered. Flowers about one-third larger than in 
H. americana. Blacts lanceolate and often toothed. Calyx pur- 
plish red, campanulate, pubescent; the segments linear-oblong, 
obtuse, and nearly equal. Petals narrowly linear, persistent, about 
as long as the stamens. Styles much exserted. 

This species has the foliage of H. parvifolia, the inflorescence of 
H. hispida, and the calyx of H. americana. 


Plate V. Heuchera rubescens, of the natural size. Fig. 1, a 
flower. Fig. 2, the same laid open. Fig. 3, transverse section 
of a capsule. Fig. 4, a seed. All the figures are magnified. 


f 


a 


Ackerman ath 379 BroadwayN Y. 


PeornigA BUBESUBNGS,, lore. 


bd 


wee) Py Le 
> * 20) 
: 4 
+, 
af 
7 
v 
* 
td 
¥ 
= 
t 4 
. 
Re at “i 
aa 
d 7 ¥ -" 
ub 
‘e de 
- s 
a 


APPENDIX D.—BOTANY. 889 


PHUCEDANUM CITERNATUM, (var.? PLATYCARPUM.)—Fructibus 
obovatis, alis membranaceis disci sesquilatioribus.—With the pre-. 
ceding. Except in the broadly-winged fruit, this plant does not 
appear to differ essentially from P. biternatum, Nutt. 


THASPIUM MONTANUM, Gray.—Fl. Fendl. p.57? On a mountain 
bordering the Salt Lake. Fl. May 25. One specimen has a per- 
ennial root, crowned with several spreading scapiform stems, which 
are (in the flowering state) from five to eight inches long. The 
whole plant is very glabrous and somewhat glaucous. The leaves 
are bi-tripinnatifidly cut, with oblong, acute, entire, or incised 
lobes. The yellow flowers are in dense umbels, with numerous 
rays. There is no involucre, and the involucels consist of T—-9 
linear lanceolate leaflets. The carpels of the young fruit are 
furnished with five broad, undulate wings. The vittz in the in- 
tervals seem to be solitary, or sometimes double. 


ASTER OBLONGIFOLIUS, Nutt.—Stansbury’s Island, Salt Lake, 
June 26. 


ERIGERON CONCINNUM, Torr. and Gray, Fl. 2, p. 174.—Valley 
of Salt Lake, May 30. 


DIETERIA PULVERULENTA, Nutt. in Torr. and Gray, Fl. 1, p. 
101.—Green River, Sept. 12. 


SOLIDAGO MISSOURIENSIS, Nutt.—With the preceding. 


LINOSYRIS SERRULATA, Torr. (nov. sp.)—Ramulis scabriusculis ; 
folus anguste linearibus trinervibus rigidiusculis acutis, margine 
serulatis; capitulis fastigiato-corymbosis subquadrifloris; squamis 
oblongo-lanceolatis glabris subquinquefariam, imbricatis laxiusculis, 
exterioribus multo brevioribus, coroliis glabris.—Valley of the 
Salt Lake. 


GRINDELIA SQUARROSA, Dunal.—Bear River, near the Hot and 
Cold Springs. Fl. May 10. 


STENOTUS CcaspPrITosuS, Nutt. in Torr. and Gray, Fl. 2, p. 238.— 
Valley of the Salt Lake. 


AMBROSIA CORONOPIFOLIA, Torr. and Gray, Fl. 2, p. 291.— 
Table land at the northern extremity of Salt Lake Valley, Sept. 19. 


Monorurix, Torr. (nov. gen.)—Capitulum hemisphericum, radia- 
tum. Involucrum subtriseriale; squamis subzequalibus oblongo-line- 
aribus. Receptaculum, nudum. Flores radii foeminei, uniseriales 


890 APPENDIX D.—BOTANY. 


ligulati; ligula oblonga, apice tridentata. Flores disci hermaphroditi 
4-dentati. Styli rami lineares, appendice elongato-lanceolata ter- 
minati. Achenia radii et disci conformia. Pappus uniaristatus; 
arista scabra corolla breviore; squamule, null, suffrutices e 
basi ramosissimi. Folia opposita, vel alterna, ovata petiolata den- 
tata vel sublobata. Pedunculi terminales, elongati, monocephali. 
Flores lutei. 


M. sTANSBURIANA, Torr.—Crevices of limestone rocks on Stans- 
bury’s Island, Salt Lake. Fl. June 26. 

The lower part of the stem is thick and ligneous, but the branches 
are herbaceous. These are about a span high and are minutely 
glandular-pubescent. The leaves are scarcely half an inch in dia- 
meter, broadly ovate, or almost orbicular in outline, often subcor- 
date at the base, with a few coarse, obtuse teeth, or almost lobed; 
the lower ones mostly opposite, but the upper ones often alternate. 
Heads 6-8 lines in diameter. Scales of the involucre in two 
or three series lanceolate, acute, glandularly puberulous, some- 
what villous at the tip. Rays 6-10; the limb longer than the 
tube, and nearly twice as long as the involucral scales. Disk 
flowers constantly 4-toothed in all my specimens. Achenium obo- 
vate-oblong, compressed, slightly hispid-ciliate on the margin, 
crowned with a single rigid, upwardly scabrous bristle. 

This genus is nearly related to Perityle of Bentham (Bot. Sulph. 
p- 23,) but differs in the absence of squamellz on the achenium; 
the pappus consisting of a single bristle. A second species exists 
in Lindheimer’s Texan collection of 1850, (No. 314.) 


Plate VI. Monothrix stansburiana, of the natural size. Fig. 1, 
a leaf. Fig. 2, A head of flowers. Fig. 3, an involucrum laid 
open, the flowers removed to show the receptacle. Fig. 4, the 
same divided longitudinally. Fig. 5, an inner and an outer scale 
of the involucrum. Fig. 6, a ray flower. Fig. 7, a disk flower. 
Fig. 8, corolla of the disk flower laid open. Fig. 9, branches of 
the style and their appendages. 


CHENACTIS STEVIOIDES, Hook. and Arn.; Torr. and Gray, Fl. 
2, p. 871.—Strong’s Knob, Salt Lake, June 10. Several of the 
ray flowers have the corolla dilated, but the lobes still nearly equal, 
and, as is the pappus, considerably shorter than in the disk flowers. 

C. tenutfolia of Nutt. is scarcely distinct from this species. 


C. ACHILLEHFOLIA, Hook. and Arn.; Torr. and Gray, Fl. 1.¢— _ 
_ Stansbury’s Island, June 20. Stems about a span high, several 


aa ae 


7s 


4 


‘Ackerman. Lith 379 Broadway NY 


MONOTHRE 31 


a 


fpeiareein coerce 


CHENACTIS ACHILLE A\FOLIA Hooke Arn. 2. 


_- 


APPENDIX D.—BOTANY. 391 


from one root. Leaves somewhat fleshy, densely clothed with a 
white tomentum; the lobes very small, obtuse, and much crowded. 
Heads few (8-6) in a terminal corymb. Flowers of the ray and 
disk nearly alike, funnel-form. Pappus of about ten oblong, 
obtuse, denticulate scales; five of which, in the disk flowers, are 
nearly as long as the tube of the corolla, and the five other about 
half as long. Scales im the ray flowers much shorter than the 
corolla tube. 


Plate VII. Chenactis achillecfolia, of the natural size. Fig. 1, 
ahead of flowers. Fig. 2, an exterior scale of the involucrum. 
Fig. 3, an interior scale of the same. Tig. 4, a disk flower. 
Fig. 5, cross section of an achenium. Fig. 6, a ray flower. Fig. 
7, branches of the style and appendages. Fig. 8 and 9, scales of 
the pappus from a disk flower. 


LAYIA GLANDULOSA, Hook. and Arn., Torr. and Gray, P12, 
p- 394.—Valley of the Salt Lake, east side. 


ACHILLEA MILLEFOLIUM, Lin.—Islands of the Salt Lake, June. 

ARTEMISIA TRIDENTATA, Nutt. in Trans. Amer. Phil. Soc. 
(n. ser.) 7, p. 898.—Green River, Sept. 12. Many of the larger 
species of the genus are called «“Sage’’ by the hunters and emi- 
grants. | 

A. FrigipA, Willd.; Torr. and Gray, Fl. 2, p. 424.—With the 
preceding. 


A. Lupovicrana, Nutt., gen. 2, p. 148.—With the preceding. 
A. CANADENSIS, Mich., Fl. 2, p. 129.—With the preceding. 


SENECIO FILIFOLIUS, Nutt. in Trans. Amer. Phil. Soe. (n. ser.) 
7, p- 414.—Green River, September. 


S. HYDROPHILUS, Nutt. 1. c.—Valley of the Salt Lake. 


S. HOOKERI, Torr. and Gray, Fl. 2, p. 488.—Weber River, 
May 16. Scales of the involucre with black villous tips. 
_ TETRADYMIA NUTTALLI, Torr. and Gray. 7. spinosa, Nutt., 
1, c.—Shore of the Salt Lake, May 5. A thorny shrub, about 
four feet high. _ | . 
| CIRSIUM UNDULATUM, Spreng.—Stansbury’s Island, Salt Lake. 
FL June 24. 


STEPHANOMERIA RUNCINATA, Nutt. in Trans. Amer. Phil. Soc. 
‘7, p. 427.—Carrington’s Island, Salt Lake. 


392 APPENDIX D.—BOTANY. 


LYGODESMIA JUNCEA, Don.; Hook., Fl. Bor. Amer. 1, p. 295.— 
Stansbury’s Island, Salt Lake, June 23. The heads in our speci- 
mens are quite as large as in L. grandiflora. Captain Stansbury 
states that the flowers are purple. 


MALACOTHRIX SONCHOIDES, Torr. and Gray, Fl. 2, p.486.—Shore 
of the Salt Lake, and on Carrington’s Island, May 30. The pap- 
pus is decidedly double in this species. The outer series consists 
of five slender, nearly glabrous, and somewhat persistent bristles ; 
the inner of about fifteen scabrous capillary bristles, which are 
caducous, and separate in a ring. I have seen the same character 
in two or three other species. Dr. Gray, in his Plante Fendleri- 
ane, (p. 113, No. 453,) says that he noticed in « M. sonchotdes, 
M. coulteri, and especially in M. californica, that two (opposite) 
bristles of the pappus are naked, instead of barbellate, and rather 
stronger and less deciduous than the others.” In JZ. sonehoides 
I believe the outer series always consists of five bristles; but in 
some species they are variable in number, and in others are entirely 
wanting. 


CREPIS ACUMINATA, Nutt. 1. ¢.; Torr. and Gray, Fl. 2, 489.— 
Stansbury’s Island, Salt Lake, June 23. This is the tallest of our 
indigenous species of crepis. Some of our specimens are about 
three feet high. The radical leaves (including the petioles) are 
more than a foot in length. 


Plate VIII. Crepis acuminata, of the natural size. Fig. 1, a 
separate flower magnified, as are the following. Fig. 2, an ache- 
nium with its pappus. Fig. 3, one of the hairs of the pappus. 


FROXIMON cUSPIDATUM, Pursh. Fl. 2, p. 742.—Valley of the 
Salt Lake. 


CASTILLEJA HISPIDA, Benth. in Hook. Fl. Bor. Amer. 2, p. 105. 
—Shore of the Salt Lake, May. 


C. MINIATA, Dougl. in- Hook. Fl. Bor. Amer. |. c.—With the 
preceding. 


C. SESSILIFLORA, Pursh. Fl. 2, p. 738.—Weber River. 


PENSTEMON GRANDIFLORUM, Nutt. in Fras. Catal., 1813.—On 
the Arkansas River. 


ERITRICHIUM GLOMERATUM, D.C. Prodr. 10, p. 131. Myosotis 
glomerata, Nutt.—Near Salt Lake City. Fl. April 29. 
oe 


Wily 

Ve dy | 

; A WW i: 
Naa 


| 


Aclcermaan, Lith.379 Broadway, NY 


APPENDIX D.—BOTANY. 393 


EcHINOSPERMUM FLORIBUNDUM, Lehm.; Hook. Fl. Bor. Amer. 
2, p. 84.—Valley of the Malade, Sept. 25. Near H. deflecum. 


AMSINCKIA LycopsorpEs, Lehm.; D. ©. Prodr. 10, p. 117.— 
Shore of the Salt Lake. Fl. May 5th. 


MERTENSIA DRUMMONDII, G. Don.; D. C. Prodr. 10, p. 86.— 
Salt Lake Valley. 


LITHOSPERMUM? crRcUMScISsUM, Hook. and Arn., Bot. Beech. 
Voy., suppl. p. 870.—On Green River. In my account of the 
plants collected in California and Oregon by the United States 
Exploring Expedition, I have made this plant the type of a new 
genus, (P%ptocalyz,) allied to Eritrichium, from which it differs in 
its naked corolla and deciduous calyx. 


HYDROPHYLLUM capPITaTUM, Dougl.; Benth. Trans. Lin. Soc. 
17, p. 278.—Ogden’s Pass, May 15. 


EUTOCA HETEROPHYLLA, Torr. (n. sp.)—Erecta, scabro-pubescens; 
foliis oblongo-linearibus subsessilibus, integris vel ad basin utrinque 
unilobatis, lobis oblongis v. linearibus; floribus brevi-pedicellatis ; 
lobis calycinis spathulata linearibus obtusiusculis; corolla patenti- 
campanulata calyce sesquilongiore; placentis multiovulatis.—Val- 
ley of the Salt Lake, on the eastern side. 

Annual; about a foot high. Radical leaves spatulate; the cau- 
line ones broadly linear, 1-14 inch long; either entire or furnished 
on each side at the base (sometimes only on one side) with a 
spreading, narrow, acute lobe, so that the leaves appear somewhat 
halberd-form. Racemes short, terminating the branches. Lobes 
of the calyx about three and a-half lmes long. Corolla widely 
campanulate, almost rotate, about five lines long; the lobes short 
and rounded. Appendages ten, narrow, connivant in pairs between 
the bases of the filaments. Stames nearly equal, a little shorter 
than the corolla. Style somewhat exserted; 2-lobed at the 
summit. Ovary with 15-20 ovules attached to each placenta. 
This species resembles H. phacelioides, Benth., but differs in the 
nearly sessile narrower leaves, the larger and broadly campanulate 
- corolla, many-ovuled placentze, &c. 


Grura (Ipomopsis) PULCHELLA, Dougl. in Hook. Fl. Bor. Amer. 
2, p. 74.—Ogden Pass, May 15. 


COLLOMIA LINEARIS, Nutt. Gen. Amer. pl. 1, p. 126.—W<th the 
preceding. 


394 APPENDIX D.—BOTANY. 


PuHLox HOODII, Richards, in Frankl. Jour. app. ed. 2, P. 6, t. 28. 
—Mountains near the Salt Lake, April and May. | 
P. tonarroiia, Nutt. Jour. Acad. Philad. 7, p. 41.—N BE Site). 
shore of the Salt Take: and near the mouth of Bas River, May 10. 

PHYSALIS LANCEOLATA, Mich.—Salt Lake Valley, June. 


GENTIANA AFFINIS, Griseb., Gent. p. 289.—Moist places, Aug.18. 


ACERATES DECUMBENS, Decaisne in D. C. Prod. 8, p. 522. 
Anantherix decumbens, Nutt.—Mountain on Stansbur'y’s Island, 
Salt Lake, June 26. Stems often assurgent. Calyx and corolla 
green. Crown dark purple. 


CoMANDRA UMBELLATA, Nutt., Gen. 1, p. 157; Hook. Fl. Bor. 
Amer. 2, p. 189 t. 179—-Gitmabairss pare Salt Lake. Fr. 
June 20. 


RuMEX VENOSUS, Pursh. Fl. 2, p.? Green River. Fr. Septem- 
ber 12. 


ERIGONUM UMBELLATUM, Torr. in Annal. Lyc. Nat. Hist. New 
York, 2, p. 241.—Valley of the Salt Lake. 


EK. Fremontil, Torr.—With the preceding. 


SARCOBATUS VERMICULARIS, Torr. in Emory’s Report, p. 149. 
S. maximiliani, Nees. Fremontia vermicularis, Torr. in Fremont’s 
first and second Reports. « Pulpy Thorn” of Lewis and Clark’s 
travels.—Strong’s Knob, Salt Lake, Fl. June 10. 


GRAYIA POLYGONOIDES, Hook. and Arn. Bot. Beech. Voy. suppl. 
p. 338, Hook. ict., 271. G. spinosa, Mog. in D. C. Prodr. 11, p. 
110.—Carrington’s Island, Salt Lake. 


CHENOPODINA LINEARIS, Mog. in D. C. Prodr. 11, p. 164, excel. 
syn. Ell. and Michx.—Mountain on the west shore of the Salt 
Lake. Fl. May 30. This plant attains the height of about three 
feet. The lower part of the stem is stout and shrubby. It differs 
entirely from the C. maritima of the Atlantic States; yet the 
authors who describe it as not shrubby are quoted by Moquin 
under C. linearis. 


_ARTHROCNEMUM FRUCTICOSUM, Mog. Chenop. Enum. p. 111, 
andin D. C. Prodr. 11, p. 151?—North-shore of the Salt Lake. A 
common plant in all the salines of New Mexico and California. 
It is a shrub about one foot high, and much branched. The joints 
of the branches are more or less compressed, and emarginately 


APPENDIX D.—BOTANY. ' Boe 


bifid at the summit. The spikes are cylindrical and are not 
jointed; the flowers being alternate, and immersed in deep exca- 
vations of the rachis. The calyx is quadrangular, and consists 
of four cohering sepals, which are cucullate, spongy at the summit, 
and at length separate from each other. There is but a solitary 
stamen. ‘The seed is loose in the utricle, oblong, and the embryo 
forms about half of an ellipse. 


OBIONE .CANESCENS, Moq. Chenop. p. 74, and O. occidentalis, 
Mogq. in D. C. Prodr. 11, p. 112. Pterochiton occidentale, Torr. and 
Frem., in Frém. second Rep. p. 818. Obione tetraptera, Benth. 
Bot. Voy. Sulph. p. 48.—On Green River. Fr. September 10. 
This is a variable species, especially in the characters of the 
mature fructiferous calyx. Sometimes it is furnished with short, 
irregular-toothed wings, and at other times the wings are very 
broad and nearly entire. 


O. CONFERTIFLORA, Torr. and Frém. 1. c.—With the preceding. 


ABRONIA MELLIFERA, Doug. Miss. Hook. Fl. Bor. Amer. 2, p. 
125, Bot. Mag. 1. 2879.—Strong’s Knob, Salt Lake. Fl. June 10. 
Easily distinguished from A. wmbellata by its broad involucral 
leaves and green flowers. A. micranthus, Torr. in Frémont’s first 
Report, p. 96, and in Emory’s Report, p. 149, seems to be a 
particular state of the plant, in which it bears very small but 
perfect flowers. In those works I noticed the peculiarity of the 
embryo; the inner cotyledon being constantly abortive. The same 
character exists in all the species of this genus: but I have not 
observed it in any other nyctagineous plant. 


SHEPHERDIA ARGENTEA, Nutt. Gen. Amer. Pl. 2.—Black’s 
Fork of the Green River. Fr. September 12. 


EPHEDRA AMERICANA, Willd. Spec. Pl. 4, p. 860? Endl. Synops. 
Conif. p. 254.—Shore of the Salt Lake. A leafless shrub with 
very numerous branches, growing about four feet high. It is very 
doubtful whether it be the same as Willdenow’s plant, which is a 
native of Quito. Although it is not uncommon in the interior of 
California and in New Mexico, I have never received the female 
flower or the fruit. All my specimens are males. H. americana 
is described as moneecious. The Hphedra noticed in Emory’s 
Report under the name of LH. occidentalis, (a mistake for H. amer- 
cana), differs from this species in its three-parted sheaths with long 
subulate points. | 


396 APPENDIX D.—BOTANY. 


TRIGLOCHIN MARITIMUM, Lin.—Pursh. FI. 1, p. 257. —Stans- 
bury’s Island, Salt Lake, June 24. 


POLYGONATUM CANALICULATUM, Pursh. Fl. 1, p. 235.—Valley 
of the Salt Lake? 


AMIANTHIUM NUTTALLII, Gray, Melanth. in Ann. Lye. Nat. 
Hist. N. York, IV., p. 123. Helonias angustifolia, and H. 
paniculata, Nutt.—Valley of the Salt Lake. Fl. May 1. 


AMBLIRION, Rafin. in Journ. de Phys. 89, p. 102; Bernhardi, 
Bot. Zeit. 1835, p. 895? (ex Kth. Enum. 4, p. 255.) Laliwm § 
Amblirion, Endl. gen. sub. No. 1098. Fritillaria § Hucrinum, 
Nutt. 


A. PUDICUM, var. BIFLORUM, Torr. Lilium pudieum, Pursh. 
Fl. 1, p. 228, f. 1.; Schult. Syst. 7, p. 401. Fritillaria pudica, 
Spreng. Syst. 2, p. 64; Nutt. in Journ. Acad. Phil. 7, p. 54. 
Hook. Fl. Bor. Amer. 2, p. 182; Kunth Enum, 1. c.—Promontory 
Range, Valley of Salt Lake. Fl. April 12. 

This rare and interesting plant was long ago proposed as a 
distinct genus by the late Mr. Rafinesque. It is allied both to 
Fritillaria and to Lilium. It differs from both in the want of 
nectaries. Unfortunately the fruit is not known, so that it can- 
not be compared with those genera in an important character. 
Our specimens are all two-flowered. The root is flat, orbicular, 
and toothed round the border, with a cluster of little tubers on the 
upper side at the base of the stem. ‘The leaves are linear, and 
from two to four inches long. The flowers are yellow, nodding, 
about an inch in length, somewhat obconical or funnel-form, and 
entirely destitute of a nectariferous groove. The stigma is simple 
and undivided. 

According to Mr. Nuttall, Fritillaria tulipefolia of Caucasus is 
another species of this genus. I have also specimens of what may 
prove to be a third species, collected by Colonel Frémont on the 
Feather River, California; for the style, though thickened at the 
summit, is undivided, and the nectary is wanting: but there are 
several flowers in a loose racemose panicle. 


Plate IX. <Amblirion pudicum, of the natural size. Fig. 1, a 
sepal magnified, as are all the following. Fig. 2, a stamen show- 
ing the back of the anther. Fig. 3, a front view of the same. 
Fig. 4, the pistil. Fig. 5, a cross section of the ovary. 


oadway, NY 


~ 
2 


Ackerman Lith 379 B 


UM. 


AMBLIRION PUDIC 


APPENDIX D.—BOTANY. 8397 


ALLIUM STELLATUM, Fraser, Bot. Mag. t. 1576.—Weber River, 
May 23. 

A. RETICULATUM, Fraser, Bot. Mag. t. 1840.—Wahsatch Moun- 
tains, June. 


CALOCHORTUS LUTEUS, Nutt. in Jour. Acad. Phil. 7, p. 53; 
probably not of Douglass.—Valley of Salt Lake. The root is called 
<<sego”’ by the natives, and is much esteemed by them as food. 
It is bulbous, and varies in size from that of a pea to that of a 
filbert. Our plant agrees exactly with the description of Nuttall, 
who was probably mistaken as to the colour of the flower. The 
inner sepals seem to be whzte, except at the claw, which is yellow. 
I have not been able to institute a comparison between this plant 
and Douglass’s C. luteus ; but if ours proves to be distinct, it may 
be called C. nuttalli. 

ERYTHRONIUM GRANDIFLORUM, Pursh. Fl. 1, p. 231. Lindl. 
Bot. Reg. t. 1786.—With the preceding. 


TRITELEIA GRANDIFLORA, Lindl. Bot. Reg. fol. 1293. Hook. 
Fl. Bor. Am. 2, p. 186, t. 198, B.—Valley of Salt Lake. FI. 
May. 

JUNCUS BALTICUS, Willd., Hook. Fl. Bor. Amer. 2, p. 189.— 
Antelope Island, Salt Lake, June 1. 

SISYRINCHIUM BERMUDIANA, Lin. S. anceps, Cavan.—Walnut 
Creek. 

Hypoxis EREctTA, Lin.—Upper Arkansas. 

ScCIRPUS TORREYI? Olney.—Gray, Bot. N. States, p. 526?— 
Stansbury’s Island, Salt Lake. Fr. June 26. 

Differs from S. torreyi in its longer and larger spikes, and in 
shorter point of the achenium; but in other respects it agrees. 


Errocoma cuspipata, Nutt. Gen. 1, p. 40.—Antelope Island, 
Salt Lake, June 18. A beautiful grass, which seems to be distinct 
from Stipa. 


_ KokLertia cristata, Pers.—Gray, Gram. and Cyp. 1, No. 45.— 
With the preceding. 


Horpeum JuspatumM, Lin.—Torr. Fl. 1, p. 158.—Antelope 
Island, Salt Lake, June. 


AGROPYRUM REPENS, Gaert.—With the preceding. 
CLymus striatus, Willd.—With the preceding. 


3 / 
. 
m=, 
° 
‘ 
: 
“ 
7 
‘ 
% q 
. 
‘ 
i 
) 
+ 
~ 
t 
* f 
: 
i 
~ ae 
‘ e 
a * 
- ‘ > 
oy 
\ » 
y 
¥ , : 
‘ } 


APPENDIX E. 


LETTER FROM PROFESSOR JAMES HALL, OF NEW YORK, 


a CONTAINING OBSERVATIONS ON THE 


GEOLOGY AND PALZXONTOLOGY 


OF THE 


COUNTRY TRAVERSED BY THE EXPEDITION, 
AND 


NOTES UPON SOME OF THE FOSSILS 
COLLECTED ON THE ROUTE. 


APPENDIX E. 


GEOLOGY AND PALAONTOLOGY. 


ae 


BY PROF. JAS. HALL. 


AuBany, February, 1852. 
CAPTAIN STANSBURY : 

DAR S1r:—I have examined with care the specimens of rocks 
and fossils which you submitted to my inspection. I find them, 
with some few exceptions, to represent very clearly the products 
of four distinct geological periods, as follows:—The older are me- | 
tamorphic rocks of silurian or devonian age, or perhaps both; the 
next in order, and recognisable by their fossil remains, are of the 
carboniferous period; the third are of the cretaceous period; and the 
fourth are of the tertiary. Besides these there are the products 
of ancient volcanic action in the basalts and amygdaloids, with 
some specimens of obsidian. 

After a careful examination of the specimens, and a comparison 
with the notes and journal which you submitted to me, I have 
marked upon the map of your route, and upon the map of the 
Salt Lake region, the different colours indicating the ‘character 
of the geology at the different points where the specimens were 
collected. I am aware that the specimens with the notes, together, 
would have warranted me in colouring in a more extended manner, 
but I have preferred to confine myself to the position and actual 
evidences furnished by specimens. By having the map in this 
condition nothing is hazarded, and every new fact obtained can be 
readily added to it, or it may be filled up to some extent from the 
indications furnished by the topographical features. _ 

It will be the more satisfactory mode to follow your route in the 
remarks I shall make in this connection. 

The first specimens furnished are from the west side of the Mis- 
‘souri River, near and above Fort Leavenworth. These are all 
from limestone of the carboniferous period, and apparently from 
the upper of the two great limestones of this period in the west. 
The most conspicuous fossils are Productus, Terebratula, &c. 

si 401 


402 APPENDIX E.—GEOLOGY. 


The route from the Missouri westward shows a continuation of 
this limestone as far as the Big Blue. 

Here it disappears, judging from specimens and remarks in the 
notes. It is soon succeeded by strata of cretaceous age, which, 
from the specimens preserved, I have been able to recognise as 
extending for a considerable distance on the route between Turkey 
Creek and Big Sandy. 

Among the cretaceous fossils are a species of Pholadomya, and 
the Inoceramus, which is so common and abundant in numerous 
localities in this region. 

It is quite probable that these beds extend much farther, but 
I find no specimens in the collection; and the notes indicate that 
there are heavy deposites of drift, which may have obscured the 
exposure of the formation below. 

This drift formation, (judging from the descriptions given in the 
notes,) or the debris from the immediate geological formation, ap- 
pears to have covered the older stratified deposites, since no men- 
tion is made of them till approaching the forks of Platte River on 
the 25th June. At this point were collected some specimens of 
clays with small marine shells, too imperfect for determination; but 
from the general character, and from the occurrence of bones in the 
same place, it is presumed that they are of tertiary age. Above 
the forks of the Platte River similar bones and shells are noticed, 
and, on the 1st of July, specimens of bones were collected. Nu- 
merous fragments of bones were collected on the 3d of July, ap- 
parently belonging to some mammalia of the herbivorous character. 
These bones are too imperfect for determination beyond their 
general character. From the description of the mode of occurrence, 
and their being imbedded in a matrix of considerable hardness and 
tenacity, one would be led to infer that they were of some tertiary 
deposite. 

Among these specimens is a single ramus of the lower jaw, which 
apparently belonged to some carnivorous animal; but no teeth are 
preserved in it, nor were any teeth of any kind found in the col- 
lection. 

From July 3d to 11th, the notes give no evidence of any thing 
of special interest. On the latter date, bones are memtioned as 
occurring in the locality examined, but no specimens having this 
date are preserved in the collection. 

It would appear that the character of the country from near 
Fort Kearny to near Fort Laramie is uniform, and that no de- 


APPENDIX E.—GEOLOGY. 403 


posite of older date than the tertiary were cbserved. Of the 
specimens collected there is but a single individual indicating the 
character of a marine formation. From the condition of the bones 
it may even be questioned whether the deposite containing them is 
not of post tertiary age. 

The specimens from the vicinity of Fort Laramie are all from 
limestone of the carboniferous period. Some of the fossils are 
identical with species: collected between the Missouri and the Big 
Blue, and we can only suppose, from the great similarity of the 
specimens, that it is a continuation of the same formation. From 
the dates marked upon the specimens, it is evident that this lime- 
stone extends to some distance on the east and west of Fort 
Laramie. 

The specimens bearing date of July 19th, two days’ march 
northwest of Fort Laramie, are a feldspathic granite with little 
quartz or mica. The rocks in this locality are doubtless of me- 
tamorphic origin, probably rocks of silurian age. The specimens 
collected three days’ march in advance of this place, on the North 
Fork of the Platte River, are shaly sandstone and thinly laminated 
sandstones containing fossils. ‘The fossils are some brachiopods, 
with others similar to Monotis, and we may presume from the 
described position of the beds, and from the character of the fos- 
sils, that these beds are of devonian age. In the journal these 
beds are recorded as dipping at the rate of 15° to the north-east. 

The specimens bearing the mark of July 24th, are precisely like 
those collected at Fort Laramie, and contain the same species of 
fossils. On the same date were seen (according to the journal) 
gray and red sandstones. On the following day is recorded a bed 
of coral, three or four feet thick, with Sigillarta and Calamites. 
The specimens of this date sent to me are those of bituminous coal 
and others of soft shale, but I have been unable to distinguish any 
well-marked vegetable remains. 

From the proximity of limestone of the age of the coal, and the 
record of sigillaria and calamites occurring in the same connec- 
tion, it may be presumed that this coal belongs to the true coal 
measures; and this locality is probably an exposure indicating the 
existence of a great basin. ‘This point itself and the surrounding 
country are well worthy of a more extended examination, since 
the discovery of workable beds of coal in this region would be a 
matter of national importance. 

The record of July 27th shows the occurrence of red helen and 


404 APPENDIX E.—GEOLOGY. 


sandstones, which may be of the age of the coal, or beneath that 
formation. 

From July 30th to August 2d, the notes of the Journal and the 
specimens show the existence of compact quartz rock, crystallized 
silicious limestone, and conglomerate. 

From August 3d to August 6th, I have no specimens indicating 
the character of the formations passed over. From this date to 
August 11th, including the distance from the southern extremity 
of the Wind River Mountains to Fort Bridger, the collections are 
all of marine tertiary age, including many specimens of JVautzlus 
and other marine shells. 

From this time nearly all the records and collections pertain to 
the Salt Lake and its vicinity. Near Fort Hall several specimens 
of volcanic rocks were collected, and obsidian and lava about 
the Pannack and the head of the Malade. 

South of Fort Hall the specimens collected are of granular 
sandstone, and of quartz rock resulting from an altered sandstone; 
to the west, and above these, are chert and limestone of carboni- 
ferous age. The limestone in this locality contains fewer shells 
than that in the more easterly localities, but has a large number 
of corals. 

The specimens collected in the islands and shores of the Great 
Salt Lake are sufficient to give one a very good idea of the general 
geological features. The specimens are of metamorphic rocks, 
consisting of talcose and mica slates, hornblende rocks, and a 
few specimens of granitic or sienitic character. 

Some specimens of the latter description occur along the valley 
of Ogden’s River. Antelope Island, Frémont Island, a part of Pro- 
montory Point, and Mud Island, on the east side of the lake, 
judging from the numerous specimens, consist principally of talcose 
and mica slates, with hornblende rock. Carrington Island, Hat 
Island, a point north by west from Hat Island, name not known, 
and a part of Strong’s Knob, consist of similar rocks with some 
of altered sandstone or quartz rock. Im several localities, as at 
Promontory Point and near Mud Island, the metamorphic strata 
appear to be overlaid by a coarse conglomerate, or coarse sand- 
stone, which is partially altered, and assumes the character of a 
quartz rock. 

From all the facts in my possession, it would appear that these 
metamorphic rocks are distinctly stratified and highly inclined, but 
do not attain any great elevation. The direction of the ranges, 


APPENDIX E.—GEOLOGY. 405 


_ corresponding to that of the elevating force, appears to be nearly 
in the direction of north by west and south by east. From the 
form of the lake and the different localities at which rocks of this 
character occur, we may infer that there were two lines of eleva- 
tion, corresponding with the divisions of the lake. 

The more elevated portions of the lake shore, and the mountain 
ranges, consist of carboniferous limestone. In some localities this 
limestone is partially altered, losing its granular character and 
becoming sub-crystalline, or threaded by numerous veins of cal- 
careous spar. In most localities, however, the limestone abounds 
with fossils, particularly corals of the cyathophyllidee. 

From the records in the journal of observation and from speci- 
mens, I have been able to indicate several localities of importance. 
The principal of these is Stansbury’s Island, the summit of which 
is of limestone, and has an elevation of three thousand feet. The 
limestone is said 'to rest on coarse sandstone and conglomerate, 
specimens of which accompany the limestone. Limestone also 
occurs on the mountains of the Spring Valley range, to the south- 
west of the lake. 

Stansbury’s Island, from its position at the southern extremity 
of the lake, and from its isolated and elevated character, has been 
more fully exposed than the localities on the west side. Along the 
western shore, southward of Strong’s Knob, the same limestone 
was examined and noted in three places, and in two it is marked 
as underlaid by sandstone. Limestone also occurs at Strong’s 
Knob with the altered rocks. West of the knob, another point is 
indicated as limestone; and northward of this, Gunnison’s Island is 
of the same rock. Dolphin Island, and also a considerable space 
on the shore west of this island, are indicated as limestone, both 
from specimens and the journal. On the eastern shore, opposite 
Dolphin Island, limestone occurs in close proximity to metamor- 
phic strata. 

It should not be omitted that the same limestone occurs to the 
north of Great Salt Lake City, and is quarried in that neighbour- 
hood. It appears both from the specimens and the notes of ob- 
servation that the limestone overlies a coarse sandstone, or con- 
glomerate, which almost invariably accompanies it. 

Although I have not felt at liberty to colour on the map any 
other points than those indicated both by the notes and by specimens 
examined, yet I can have no doubt but all the elevated ranges on 
the west, south, and north of the Great Salt Lake are capped by 


406 APPENDIX E.—GEOLOGY. 


the carboniferous limestone. Judging from the relative position 
of the limestone, and the metamorphic rocks of Antelope and Fré- 
mont’s Islands, the former occupies the position of low, synclinal 
basins, the valleys between being produced to a large extent, pro- 
bably by erosion along the anticlinal axes, produced by the eleva- 
tion of the metamorphic beds. We may expect, also, that the 
same limestone will be found on the elevated plateaus and moun- 
tains on the east side of the lake. 

It will be seen from these facts that we have very satisfactory 
information that this limestone of the carboniferous period is widely 
distributed in the region around the Great Salt Lake. Its position 
relative to the coal-bed on the North Fork of Platte River has not 
been determined; but since no beds of coal have been observed on 
the slopes of the mountains in the region of the Salt Lake, we are 
left to infer that the coal is to be sought (as elsewhere) above the 
limestone. Since the existence of coal is proved in one point, 
(admitting the evidence in favour of its age being that of the car- 
boniferous period,) we are warranted in the conclusion that it has 
once existed over a much wider area, and can be sought with 
success in the proper situations. The importance of this mineral 
in that distant region cannot be too highly estimated, and the 
geographical position and extent of the beds should be one of the 
first points ascertained in the location of any route of communica- 
tion between the east and the west. 

In comparing the notes and specimens with the map of your route 
and the large map of the Salt Lake and adjacent country, I have 
confined myself to indicating by a colour the kind of rock occurrmg 
at each point, scarcely in any case extending this colouring even 
when the topographical features of the country would warrant the 
conclusion that the same rock existed. Your knowledge of the 
character of the surface and the relative elevations will enable you 
in many instances to determine the limits of those formations 
marked; while, for myself, not fully understanding their features, 
I might fall into some error. 

Hoping to see the investigations you have so well begun carried 
still further, until we can have a good geographical and geological 
inap of this region, 

T remain, 
Very truly and respectfully, 
Your ob’t. serv’t. 
JAMES HALL. 


S40) es 
2 ND 


Ackerman Lith 379 Broadway NY. 


APPENDIX E.—PALAONTOLOGY. 407 


N.B. The colours on the map of the travelled route are— 


Belts ai Carboniferous limestone. 
AFYOOM Gis. oseeas caus Cretaceous formation. 
Yellows. .cscces.cs Tertiary formation. 
(In some places not indicated by colour.) 
CE? aka a Coal-beds. 
Be Misty Sa pid abi inne Metamorphic rock. 


On the Great Salt Lake map are— 


Le EE Aa Metamorphic rock. 

BME eset aes dee saat Limestone. 

a Sandstone and conglomerate beneath the 
limestone. 


NOTES UPON SOME OF THE FOSSILS COLLECTED ON THE ROUTE FROM 
THE MISSOURI RIVER TO THE GREAT SALT LAKE, AND IN THE VICI- 
NITY OF THE LATTER PLACE, BY THE EXPEDITION UNDER THE 
COMMAND OF CAPTAIN HOWARD STANSBURY, T. E. 


THE species described in the following paragraphs are either 
from limestone of the carboniferous period or from strata nearly 
associated, and which, from their character and relations, are 
clearly of the same age. The other fossils of the collection con- 
sist of a few cretaceous species, and of numerous fragments of 
bones from the tertiary formation. . 

The brachiopods were collected to the eastward of the Salt Lake 
region, and the corals are abundant in the limestone to the west 
and north-west of the Salt Lake. The few acephala are from argil~ 
laceous beds between Fort Laramie and the Salt Lake. 


CORALS.—CYaTHOPHYLLIDEA. 
FAVIPHYLLUM? RUGOSUM, (n. sp.) 


Pirate I. Fic. 1a anv 1b. 


Cells deep; structure of the centre unknown; external portion 
cellular, with transverse septz and vertical intermediate dissepi- 
ments, giving a columnar structure. 


408 APPENDIX E.—PALHONTOLOGY. 


The specimens are all silicified, and I have been unable to ex- 
amine the central portion; the exterior, where weathered, presents — 
the appearance of a bundle of the columns of Favosites, except - 
that they are generally four-sided, and the inner sides necessarily 
narrower than the outer ones. 


FAPHRENTIS? MULTILAMELLA . 


Puate I. Fia. 2. 


Coral free, turbinate, somewhat rapidly expanding, cells deep ; 
lamellze numerous, thin; outer portion cellular. 

From the specimens in my possession, it cannot be positively 
determined that this fossil is a true cy phe: but many features 
induce this reference. 


Loc. Cloth Cap and Flat Rock, Cisne Salt Lake. 


FAPHRENTIS STANSBURII, Hall, (n. sp.) 
Puate I. Fie. 3 ab. 


Turbinate, free, or attached only by a pedicel nearly straight 
or but slightly curved; cup rather deep; margin (when entire) 
thin, lamellee numerous, thin, intermediate ones extending from | 
the margin one-third to one-half the semidiameter; fossett distinct. 

Loc. Stanbury’s Island, Cloth Cap, and Flat-rock Poimt, Great 
Salt Lake. 


LITHOSTRONTION 


, (sp. indet.) 
Prats I. Fie. 4ab. 


Coral massive; cells of medium size, deep; lamellz crenulate. 
The specimen is much weathered, and, from the presence of an 
ochreous incrustation, the specific characters cannot be clearly 
described. It differs in the dimensions and other characters of the 
cells from two species of the carboniferous period known to me 
from localities east of the Mississippi River. 
Loc. Top of Stansbury’s Island, Great Salt Lake. 


Platten 


Ackerman Lith $ 


APPENDIX E.—PALEHONTOLOGY. | 409 


BRACHIOPODA. | 


TEREBRATULA SUBTILITA. 
Puate II..Fia. lab, 2ab. 


Gibbous, obovoid; valves nearly equal, convex; beak of the 
dorsal valve elevated, incurved, and perforated at the apex; a 
mesial depression commencing just below the centre and extending 
to the front, which is produced and elevated, filling a broad sinus 
in the ventral valve; surface marked by strong concentric strie, 
or lines of growth, with faint, often almost imperceptible radiating 
strie. 

Fig. 1 a, a young individual. 
Fig. 1 b, an individual of the ordinary size and form. 
Fig. 2 a, a more gibbous form. 
Fig. 2 b, profile view of the preceding. 
Fig. 2 c, front view of the same. 
Loc. Missouri River, near Weston. 


SPIRIFER HEMIPLICATA. 
Priate II. Fiqa. 3ab. 


Shell gibbous; ventral valve more elevated, beaks nearly equal; 
entire surface marked by finely radiating strata; each valve with 
about three plications on each side of the mesial sinus and eleva- 
tion, which plications extend halfway from the base to the beak, 
leaving the upper half of the shell marked only by fine striz; 
anterior portion of the dorsal valve produced in a long extension, 
which fills a deep angular sinus in the ventral valve. 

Fig. 3a, view of the dorsal valve, which, from distortion, shows 
the beak of the ventral valve projecting above it. 

Fig. 3b, front view of the same specimen. 

Loc. Missouri River, near Weston. 


SPIRIFER OCTOPLICATA ? 
Prats II. Fia. 4ab. 


The specimens figured appear to be young individuals, and may 
probably belong to this characteristic carboniferous species. 

Fig. 4a, this specimen has suffered from lateral pressure, caus- 
ing an unnatural extension of the beak. 


410 APPENDIX E.—PALZONTOLOGY. 


Fig. 4b, the dorsal valve of a larger individual, having six pli- 
cations on each side of the mesial sinus. 

The surface is marked by distinct, undulating, concentric striz, 
which are again crossed by finer thread-like elevations, and which 
appear to have been the bases of short spines. 

Loc. Missouri River, near Weston. 


SPIRIFER TRIPLICATA, Hall, (n. sp.) 
PrateE II. Fie. abe. 


Shell subquadrangular; dorsal valve more gibbous than the ven- 
tral; area nearly linear; beak of dorsal valve small, acute, and 
closely incurved; mesial depression shallow in the upper part, but 
becoming deeper and expanded toward the base, and produced in 
front; entire surface, including the mesial sinus and elevation, 
covered with fine plications, which, being simple in their origin, 
soon divide into three, which are continued to the base without 
further division. 

This species differs from the S. striatus, Sowerby, in its form, 
and in the circumstance that the plications are less subdivided 
toward the margin of the shell. 

Fig. 5a, ventral valve, and area of the dorsal valve. 

Fig. 5b, dorsal valve of the same individual. 

Fig. 5¢, profile view of the same. The form is somewhat dis- 
torted by pressure. 

Loc. Missouri River, above Weston. 


CHONETES VARIOLATA, (D’Orb. sp.) De Koninck. 


Puate III. Fic. lab. 


This species bears the essential characteristics of those figured 
and described by De Koninck, though it is larger than most of his 
figures. The broad, scarcely defined mesial depression of the 
dorsal valve gives a straight-or slightly sinuous outline in front. 

This fossil is associated with several of the preceding species, 
near Weston on the Missouri River. 


vy 


Sy 


te 


Pla 


oadway NY 


Ackerman Lith 379 By 


Y 


- - 
- 
re 
‘ 
- 
. 
? 
. 
my 
, 
i ¢ 
i * 
. + . ‘ 
. > p ‘ 
if P 4 ‘ 
, = 
e i —~ id 
t 
* oy 
* a 
- vs 4 ey 
ay _- 
: f } New 
. 
a a eee 
i 5 ae 72 a 
me : 
Sie ay! 5 dint 
y Bae, 
oh Bi ; 
id, A ¥ 2 
ip , Vay ¢ ae t 
‘ : er 
2 ie . 
’ 
y n 
, ~ 
a 
bs 1 
c 
? 
> : 
‘ 
has . 
2 
> % : 
- 4 be 
, ‘eee 
ay i P, 
A ae 


7 a 


APPENDIX E.—PALAONTOLOGY. 411 


PRODUCTUS COSTATUS? 
Prate III. Fie. 2. 


Reference, De Koninck, Recherches sur les Animaux Fossils, premitre partie, 
page 92, pl. VIIL., fig. 3. 

The specimen figured is apparently a young individual of this 
species, in a bad state of preservation. The species is also cited 
by De K., from the Missouri River, from whence the specimen 
figured was obtained. Some other fragments. from the Platte 
River appear to belong to the same species. 


PRODUCTUS SEMIRETICULATUS. 


Prats III. Fig. 3, 5 a and 5 b. 

Reference, De Koninck, Recherches sur les Animaux Fossils, premiére partie, 
page 83, pl. VIIL., IX., and X. 

I refer, with some hesitation, the specimens here figured, to this 
very variable species. One or two of the specimens are very 
narrow and much elongated; the striz are flexuous, sometimes 
preserving the bases of numerous spines, and at other times 
entirely free from these appendages. Other specimens are pro- 
portionally shorter and broader, and present the usual form of this 
species, though none of them are larger than those figured. 

The specimens are all in limestone of a dark gray or brownish- 
gray colour, from near Fort Laramie. Some impressions of the 
same or a similar species occur in limestone from Flat-rock Point, 
and other places in the neighbourhood of the Great Salt Lake. 


PRopuctus ———— (sp. indet.) 
PuateE ITI. Fie. 4. 


This species bears considerable resemblance to P. costatus, in 
the aspect and marking of its surface, but its form is quite different. 
It occurs on the Missouri River, near Weston, associated with 
Ferebratulz and other species of Productus, Spirifer, Kc. 

A species of Productus resembling P. punctatus in the character 
of its surface, occurs on the Big Blue River, in soft shaly limestone ; 
but the specimens are too imperfect for determination. The 
occurrence of this and other species, shows the existence of car- 
boniferous strata at several points after leaving the Missouri River, 
and, in some instances, after crossing tracts of country that are 
probably of cretaceous beds. 


412 APPENDIX E.—PALZONTOLOGY. 


ORTHIS UMBRACULUM? 


Prate III. Fra. 6. 


The specimen corresponds with one from the carboniferous rocks 
of Kentucky and Tennessee, which is referred to this species. It 
is however somewhat distorted by pressure, and other specimens 
are necessary for a full determination. 

Loc. Missouri River, above Fort Leavenworth. 


ACEPHALA. : 


AVICULA? CUSTA. 
Puate IV. Fig. 1 a b. 


Shell obliquely ovoid; hinge line much shorter than the width 
of the shell; beaks prominent, and the upper part of the shell 
gibbous; surface marked by fine even striz. 

This shell occurs, with a few other fossils, in thinly laminated 
gray sandstone, which, judging from the notes of observation, lies 
below the limestone of Fort Laramie. 


TELLINOMYA PROTENSA, Hall, (n. sp.) 
PuaTE IV. Fie. 3. 


Shell elongate-oval; beaks placed about one-third the width of 
the shell from the anterior extremity, somewhat pointed; surface 
marked by fine concentric striz and some stronger lines of growth. 

Loc. near Le Bonte, with the preceding species. 


CYPRICARDIA OCCIDENTALIS. 
Puate IV. Fie. 2. 


Shell rhomboid-ovate; anterior extremity rounded; posterior 
obliquely truncate; posterior slope with a distinct carina extend- 
ing from the back to the posterior basal margin; beak near the 
anterior extremity; surface marked by concentric striz and some 
more elevated ridges or lines of growth. 

This species occurs in a gray argillaceous limestone, associated 
with Productus, Chonetes, ete. 

Loc. from the Big Blue River. 


379 Broadway NY. 


ith 


W 
a 
| 
ad 
(o) 

mee 
re 

<x 


APPENDIX E.—PALAONTOLOGY. 413 


ALLORISMA TERMINALIS, Hall, (n. sp.) 
Puate IV. Fie. 4 ab. 


Shell elongate-ovoid, with the posterior extremity extended and 
sub-acute; beaks anterior, or in a line with the anterior extremity 
of the shell; surface marked by strong concentric ridges, which, 
diverging from the beak, are more prominent on the central portion 
of the shell, and gradually die out on the cardinal line. The shell 
is also quite smooth toward the basal margin, with the exception 
of the rather coarse concentric strive which cover the entire surface. 

Fig. 4 a, view of the left valve. 

Fig. 4 b, anterior view of the shell. 

Loc. from the Big Blue River. 


Nucuna Arata, Hall, (n. sp.) 
Puate IV. Fia. 5 a b. 


Shell oval-ovate, rounded before, and gradually narrowing be- 
hind the beaks (posterior extremity broken off); beaks prominent, 
closely incurved; posterior lunule elongated and distinctly de- 
fined; surface marked by distinct (rather sharp where unworn) 
equal concentric ridges, scarcely so wide as the furrows between 
them. The ridges, when seen in a longitudinal direction, have an 
imbricated appearance. 

Fig. 5 a, the right valve, showing the anterior extremity broken 
off. 

Fig. 5 b, a cardinal view, showing the beaks and defined pos- 
terior lunule. 

This beautiful species occurs with Terebratula, Spirifer, and 
Productus, in a decomposing limestone, on the east side of the 
Missouri River, below Weston. 


GASTEROPODA. 


PLEUROTOMARIA CORONULA, Hall, (un. sp.) 
Prate IV. Fie. 6a bed. 


Depressed trochiform, volutions about five, angular, slightly 
convex on the upper side, and sloping almost uniformly from the 
suture to the margin; lower surface more rounded; aperture sub- 


quadrangular; striz bent abruptly backward on the acute outer 


414 APPENDIX E.—PALZONTOLOGY. 


edge of the last volution, which has scarcely a distinct carina; 
upper margin of the volutions, along the suture, marked by an 
elevated nodulose ridge, giving a beautiful coronate feature to the 
upper part of the shell. Umbilicus, none. 

Fig. 6 a, view of the upper surface of the shell. 

Fig. 6 b, view of the aperture, which is imperfect. 

Fig. 6 ¢, anterior view of the shell. 

Fig. 6 d, base of the shell. 

This beautiful species was found with other carboniferous fossils. 


EUOMPHALUS SUBPLANUS, Hall. (n. sp.) 
PraTE IV. Fig. 7 a b. 


Spiral, convolute; volutions about five, in contact, round, or 
very obtusely sub-angular on the upper outer margin of old speci- 
mens, surface evenly striated. 

In young specimens the apex is depressed, but in the specimen 
figured it is slightly above the outer volution. The specimen fig. 
7 b is the under side of a cast of an individual of apparently the 
same species, the absence of the shell leaving the volutions not in 
contact. 

Fig. T a, view of the upper side of a specimen from limestone 
on the top of Stansbury’s Island, Great Salt Lake. 

Fig. 7 b, a cast in limestone from between the Big and Little 
Blue Rivers. 


APPENDIX F. 


LETTER FROM L. D. GALE, 


WITH 


A CHEMICAL ANALYSIS OF THE WATER OF THE GREAT SALT LAKE, 
AND OTHER MINERAL WATERS AND SALINE SUBSTANCES, 
COLLECTED DURING THE JOURNEY. 


a 


a. Sh oe 
ae 


*< 


pe 


. 
ae 


APPENDIX F. 


CHEMICAL ANALYSES, &c 


BY DR. L. D. GALE. 


ee 


Sir:—I have carefully examined the specimens of water, and 
earthy and saline compounds, from the Valley of the Great Salt 
Lake, which you put into my hands for chemical analysis, and I 
herewith report the results. 

I have inspected and tested all the specimens, and made:a de- 
tailed analysis of such only as I deemed might be of some 
interest to know. ‘Thus, the water of the Great Salt Lake, that 
of the Hot Spring, the Warm Spring, and the native saleratus, 
are all more or less important to the public. 

The first of these is perhaps the most important of all. The 
water of this lake must vary considerably in its strength at different 
seasons of the year. It is important, hence, in stating the strength 
of the water to state the time when the water experimented on was 
collected. That fact, so far as it relates to these experiments, 
will be found, it is presumed, in the body of the work. 

The specimens examined contain full twenty per cent. of pure chlo- 
ride of sodium, and not more than two per cent. of other salts, and is 
one of the purest and most concentrated brines known in the world. 

The strongest brine reported by Professor Beck, on the salines 
of the State of New York, is that of the new well at Syracuse, 
containing 17.35 per cent. of chloride of sodium.—The water of 
the Warm Spring is a sulphurous water, strongly impregnated 
with sulphuretted hydrogen, and has medicinal virtues that may 
render it valuable. 

The native saleratus from Mud Plain, as well as that from the 
banks of the Sweetwater, is a valuable domestic salt. 

Before stating the results of the analyses made, it is proper to 
say that the quantity of water from the several sources was too 


small to enable me to make so critical an analysis as I otherwise 
27 417 


418 APPENDIX F.—CHEMICAL ANALYSES. 


would have done.—That from the Salt Lake being not more than 
about two quarts, and that from the Warm Spring a little more 
than half a pint, while that from the Hot Spring was about a pint 
and a-half. I was compelled, therefore, to use the greatest possible 
economy in the materials, and to confine my attention to the most 
common materials generally found in salines. Besides this, I was 
also obliged to confine myself mainly to the liquid contents of the 
vessels, and neglect, except in one case, (namely, the water of the 
Warm Spring,) the gaseous matters, and that in consequence of 
the sealing of the bottles having been loosened by the severe agi- 
tation in travelling, so that more or less of the contents of each 
vessel had escaped before they arrived in this city. It was there- 
fore useless to make any experiments on the supposed gaseous 
matters as they may have existed at the sources. 

The great importance of the waters of Great Salt Lake rendered 
it justifiable, in my view, to make some experiments of a practical 
character relative to the procuring from it of a good quality of 
salt, even better than that usually found in this section of country, 
and by which it seems to me the water may be a source of revenue 
and convenience. 

As will be seen in the detailed analysis below, the salt water 
yields about twenty per cent. of pure common salt, and about two 
per cent. of foreign salts; most of the objectionable parts of which 
are the chloride of lime and the chloride of magnesia, both of 
which, being very deliquescent, attract moisture from the damp 
atmosphere, which has the effect to moisten and partially dissolve 
the common salt, and then when the mass is exposed to dry air, 
or heat, or both, a hard crust-is formed. I believe I have found 
a remedy for the caking, which is cheap and easily used. It con- 
sists in sprinkling over the salt obtained by the evaporation of the 
water and heaped up in a bin or box containing a porous bottom 
of blankets or other like material, a cold solution of the salt as it 
is concentrated from the lake, till crystals begin to be deposited. 
This concentrated brine, while it will dissolve none of the common 
salt, will dissolve all the chlorides of calcium and magnesium, and 
carry them down through the porous bottom, and thus leave the 
salt purer and better than any now found in our markets. For 
persons who are obliged to prepare temporarily the salt, as travel- 
lers passing through the country, the water of the lake, without 
concentration, may be used for washing out the deliquescent chlo- 
rides, sprinkling the heap of salt by a watering pot, at intervals 


APPENDIX F.—CHEMICAL ANALYSES. 419 


of two or three hours during a single day, and allowing it to drain 
and dry at night, and be spread to the sun an hour or two the 
following morning. 

This experiment is successful on a small scale, and will no doubt 
admit of extended application. 

The water of the lake examined was perfectly clear, and had 
the specific gravity of 1.170, water being 1.000. 

One hundred parts by weight were evaporated to dryness in a 
water-bath below the boiling point, and then heated to about three 
hundred degrees of the thermometer, and retained at that heat till 
the mass ceased to lose any weight. It gave solid contents 
22.422, and consisted of 


PEERAGE SOON cx cnicica sind ipaaeigse seme tdsiseyvan destes(varanedscs 20.196 
SEU EAU OE SO, oncon cas trace nae ces icninnes soso tesiseuee pes eseceoaas 1.834 
OMIOrIGe Of MA SNESIUM hx caeaacneaca dec tonddetecvtestrseveds- set 0.252 
RO RierT rol | COPCTUM cise cinns seas saincoceacselee need yscadoyemens 0. trace. 


The water of the Warm Spring of Salt Lake City is a Harrow- 
gate water, abounding in sulphur. The water is very limpid, 
having a strong smell of sulphuretted hydrogen, and contains the 
gas both absorbed in the water and also combined with bases. 

The specific gravity of the water I found to be 1.0112, and, 
when opened, was highly charged with gas, although the cork had 
allowed much of the gas, and water even, to escape. 

One hundred parts of the water were evaporated to dryness at 
a temperature of about 200° of Fahrenheit, and yielded solid 
matter 1.082000. 

The heat necessary for this also carried off sulphuretted hydrogen 
per cent. 0.087454. 

One hundred parts of the water gave an analysis of the following 


results :— 

Sulphuretted hydrogen absorbed in the water..............++ 0.037454 

ge iM combined with bases*.........ssseseee 0.090728 

Carbonate of lime, precipitated by boiling............sssse000 0.075000 
Carbonate of magnesia, ‘‘ SERN ey corse aca wnceaee 0.022770 
Chloride of calcium .........cs.cseescccenencenccctesceteccas oeseeeses 0.005700 
Sulphate of Soda .......scseeesecceesveeeeese cesses eens sa eeeeeeaveneeens 0.064835 
Chloride of SOdiUM.........s.sescceececee serececneescsesseneneaeeseeans 0.816600 
1.023087 


* Probably combined with some of the bases and decomposed by the heat used 
to separate the water in solidifying the contents, as the gas could hardly be de- 
tected when the contents were dried. 


420 APPENDIX F.—CHEMICAL ANALYSES. 


The water of the Hot Spring was found to have the specific 
gravity of 1.0130, and one hundred parts yielded solid contents 
1.1454. 


Chioride of SOG. is <scc:sainsins snae cea caeds yen ddkgse bi eenenn ee ee 0.8052 
Chloride Of magneRitii.. scan scsn sc .dencesacteesninsponpeecbupnangeeenen 0.0288 
CFRIOFIGG GL GORCTIMIS. 5, 5 cece oncuacuce coasse verweudce meas castene nace 0.1096 
Dulphate Of Pass 1.0 ieinepet vices sleet edie.) Gecucns ce. <deues carat oP 0.0806 
Carbonnte of limes ach 63a. Fe kis he Sh eee 0.0180 
PARTBGIE Satie b sikistg do midid ils wannabes «cman wh le eaves deepak Oe 0.0180 

1.0602 


NATIVE SALZRATUS AND ALUM. 


The specimen labelled Efflorescence from a Saleratus Pond, on 
the Sweetwater River, has been tested, and found to be composed 
of the sesquicarbonate of soda, mixed with sulphate of soda and 
chloride of sodium, and is one of the native salts called Trona, 
found in the Natron Lakes in Hungary, Africa, and other countries. 

Three grammes of this salt in dry powder, cleared of its earthy 
impurities, gave carbonic acid 0.9030 of a gramme, which would 
indicate 1.73239 grammes of the sesquicarbonate. The other salts 
were found to be the muriate and sulphate of soda: the proportions 
were not determined. 

The specimen of alum from Alum Point, Great Salt Lake, is a 
rare and interesting mineral. It is a true alum; but instead of 
being an alum with an alkaline base, as potash or soda alum, it is 
found to be an alum with a base of manganese, differing from all 
other true alums in crystallizing in needle-shaped quadrilateral (?) 
prisms. It is soluble in several times its weight of water. It has 
the taste of ordinary alum, though less strong, from the fact, per- 
haps, that it is less soluble. 

The mineral is an effloresced mass, found on the surface of a slate 
-rock abounding with a sulphuret (as is believed) of manganese, from 
the decomposition of which the sulphur, being oxidized, is converted 
into sulphuric acid, and combining directly with the base, manga- 
nese and the alumina of the slate, forms the alum in question. 

The specimen, as it reached me, had lost nearly all of its water 
of crystallization; and, in order to make.a fair analysis of it as a 
specific salt, a portion of the specimen was dissolved in water and 
recrystallized, and the crystals dried to the first appearance of 
efflorescence on the projecting points, and then a given weight of 


4 


APPENDIX F.—CHEMICAL ANALYSES. 421 


the crystals was heated to drive off the water of crystallization. 
Having previously learned that it was an alum, with the double 
base of manganese and alumina, I made a careful analysis and ob- 
tained the following result from the salt dried by blotting paper :— 


100 grammes of the freshly crystallized salt gave— 


Tae ee iinSina'«cniricasinal casa vacant asic tenlepieuasien ve seas some Cuamioctaenm a Maer tates 73.0 
etenyxide OF MAAN PONEB) casi cis's/tnadsonawhoase vactwnnedsncsacbeidanuasacs> 08.9 
MBE MAUR iene corn vaaschicna once aeninienids do aialnsuaiaua dascisvas deietinlenamaeiaaa peice ane 04.0 
AUG DCAM cwnonn acne nas saskanescansinsainveg hence nate sdenaiaas <nusenss 18.0 


It is not easy to explain the relation of the acid to the two bases 
here in accordance with the usual constitution of alums, as there 
does not seem to be enough acid for the supply of an equivalent to 
each base, nor of alumina for the manganese. But as I have not 
the time to repeat my experiments, they must stand for what they 
are worth. | 

This salt may be substituted for common alum in nearly all its 
various uses for tanning, in combination with the salt brine of the 
Great Salt Lake, in what is called the process of tawing; and 
where tan bark is difficult to obtain, it is a valuable acquisition to 
the arts. And should the locality of alum at any time give out 
from exhaustion, the rock may be blasted and the alum made by 
artificial means, as the alum slates of England are worked. 

The manganese alum is also susceptible of various uses as a 
mordant and a colouring agent in dyeing, where it is not only a 
substitute for common alum, but subserves other and additional 
purposes of communicating various tints of fancy colours to shades 
of red, brown, &c. Thus we have purples, lilacs, browns, and 
many other tints, from the use of manganese. 

For details in this art, see Parnell’s Applied Chemistry—Calico- 
printing. 
| Yours truly, 

L. D. GALE. 
Howarp Sransspury, Hsq. 
Captain Corps of Topographical Engineers, U. S. Army. 


Wasuineton, March 25, 1851. 


APPENDIX G. 


METEOROLOGICAL OBSERVATIONS. 


mM 
A 
) 
KH 
<a 
— 
b> 
fa 
i} 
n 
ra) 
° 
4 
= 
Oo 
| | 
id) 
=) 
Pe 
>) 
am 
ro] 
& 
Hi 
<a 
i 
id) 
4 
oo 
(=) 
A 
S| 
Ay 
Ay 
<q 


424 


(-]) wa0zg 

“M “S 0} LeyIvg— "WIT “AID 

"IBID 

*IV91O 

"M ‘§ OF “PRIZES “ATO —"AB9TD 

"IVIIO 

"IVIIO 

*IVITD 

i) 6) 

“NI 0} SULIVETQ— "WIM “ITD “47 

‘VYSIU UL UIVI—sSpno[D 

“IVIIO 

‘ystur AUTVIT 
‘uiva LAvoy LIOA—'W ‘V Z 0} OT WOdy poqsvy 
WI0}G “M “N 0} SOUT? 4B DBursdda “YT ‘N UWOAz 

dUBOTLINY BV SUIMOTA “44ySIM Sutinp ui103s Lavoy L190 A 
‘JURYSIP JOpunyT, “M ‘N puv 

‘J 0} ‘TG Wo Surposooad ur10yg—'g 0} SutujzySryT 

"M ULAVIIQ =*g WOT Suista “g ul spnojo LAvoypy 
‘a{qIpNeV JepuNnT, 

"M ‘N 0} “O'S UOAy Surissed “Ay “Gg “AA ULAL0}S “WIM “ATO 

"MIND “11D “YI VITO 

‘PUTA OT}}IJT—'g WOIZ UIVI puUB JOpUNnY YIM ‘W109g 
‘DUISIL “G 

wor ‘una LaAvozT ="AUp [IV “Ay “GS Woaz “quays soddy 

"Q 0} ‘WITU “ALD ‘ABOTD 

"| ‘SG WlO1J ULI0}S BV SULMO][OF Urey 


‘oy ‘SCQOTD 


‘3S ‘WN 
1a | 
0 
W'S 
0 
"TULBD 
| i ane 
4 “MON 
“4T AL 'N 
“L°M ON 
“98 "M 'N 
“IT “MTN 
4 HS 


‘qUS10TL 
yeu 
-jxoiddy 


“JOM 


69 
69 
LY 


"O0LT 


“AELTNOW LEH, 


“SNOILVAUNSAQ TVOIDOTOUORALATY 


69 


19 
08 
¥9 


GL 
G9 


*potoe}}V 


GLI6G 
T8666 
C66 66 
10866 
68666 
9ZT 66 
FLG'63 
60666 
S166 
96166 
06686 
[00°66 
666 3G 


¥G6 86 
686 86 
LET'63 


GIL'66 


eeeeee 


L90°6% 
P6166 
OLL'63 


“LOJOULOIVE 


‘Nd @ 
MOY LE Ur OS” 2s 
‘Wd 6 
‘Nd. ¢ 
pe Wee 66 » 
“Wd ) 
‘HN GL 
A VOL BG. as 
‘Way 
“WW CV ih Me 55 
"I °V Ut 9G ” 
‘Wd Q 
AW a SG 3s 


Wd BT 


‘Md 6 
‘Wd J 


‘WGI 
KVL | FG » 
“d 086 
‘W'd g 


“WGI 
‘WV J 


eg Avi 
*IMOyT LET 


“6F8T 


425 


APPENDIX G.—METEOROLOGICAL OBSERVATIONS, 


‘WHI “110 
“TG 0} Suruyysry 
‘IVI 
"IBID 
*IVOIO 
"IVITO 
‘QSpIL JO “4d ySOYSTY MOTE OT49TT © MOT[OF[T—‘ULI0IG 
‘SULIpUsOSe [[T}S Jnq ‘esply—"uINd “WI “AIO 
“MOT}VALOSGO JSVI BOUTS UIA AAVOY PU W10}8-1OpUNT I, 
‘SLOMOYS [VUOISVOOQ—UIIU *11() 
"UIU “ITO 
‘TT 07 Lepunyj—szoMoyg 
‘aSpld MOTEq JeYyyeA AaT[NS ur ‘staMoyg 
“MOT Une eT) 
"IV9[O 
"IV9TD 
“TANOSSTI, OF “WT “N 
SuuuNnd Yo9Id JO poeq avou sdu[d-sursoy—av9[Q 
‘quoTIdUIvOUE SVT JO “MA “N 0} OSpPIY 
‘MS pus 
"M 0} YUNS Spnop»—jysIu Surmmp w10js ou—aVIID 
"M ‘S WoO JUeTIND Jaddq—'|M “Gg 
0} Ausiwdde Suissed -q *N pus “"M "N 0% ULIOG 
"IVI[O 
"IV9TO 
930[du109 
-UI avd JOMOT—OTOIID B Jo $ OJ WOOM puUNoL ‘o[epT 
"IV9TO 
"IVOTO 
*O9TQ JNUTLV AM 
jO OpIs “AA oSpia uo saxojzaenb yusreyIp ur sult0yg 
“UITU “UENO “YT 
‘BULAvI[O AYS—SOSSVUL UL "WIND “110 "YT 
‘porve[> puv “Wd TL “MS 
pus ‘g 0} posuvyo—puIa “QI “WV g YB UIeY— WIN 
“ULL0YG 


PEEEE ci «. 
HudduniindAZaaauis 


eee Rae 
sFER aa 


© is 
ee 


 <, 
7% we 


& 
ical 


roe 
nm 
Eo 
TD U2 


= 
Y Alm 
Ee 


= 6, 
EES 


We ws 


+ 
= 
MW 


~~ 
mn 


eco 


SL 
08 
tL 


¥9 
v9 
69 


"I1e wadgQ 
‘Irv uedg 
‘116 ued 


v9 
69 


661°86 
699 86 
09°86 
PSGL'8S 
9EV'8G 
VEP 8G 
VEV 86 
IVS 86 
66986 
IGL'°86 
1G8°86 
99L°86 
068°86 
198°8@ 
690°66 
090°66 


GEL'6G 
OLT'636 


P81'6z 


[60°66 
01066 
G10 66 


90666 
691°66 
GLO°63 


0¢8°86 
G98°86 
98°86 


610°6G 
SOL'66 


eas 


G 99 


g 9 


| ‘Apurg iq woON—"M '§ 0} Sarm1o0f wp “IT—"AVOTN ‘d "AM ‘N bis * ae LL G99'°8Z ‘Md g 
‘ONG OV] pus Apurg wooMjyog—aAvolM | “I'M “N ae si ov: 99 F89'8S | “HV 6 
‘IV9TD | “UM 'N oe ve 99 09 O19°83 ve PL 8 
“kpuvg opyywyI—'M estuNg 
0} AOIVOTD WOZILOF[—'|T “GF 0} Aopungy—'wiU ywayg di sna a cad T G6LF'8Z Nd 6 
wn ‘UIvE FYSIS—"WIN | “d “TN ¥ > suis oy TZ O88 "8S 4 
A "UU “UIT ‘d “Wy eee eee eee Gl OZE'SZ ‘N ZI 
= "M “§ Juedan9 ign tl yaeq | “dN .s Bo phe TL SLP'8Z N'V6 | SL » 
= ‘dug 
ee yoorg syuvasiag@—" A “Sf 0} SurazysyyT—"wru “winH SH | 96ST $e i SL 6LP'8S ‘Wd 6 
SS ‘yoo 8,y3044q—Avp ]]V SUISIA “YVAYS PUB ‘WIND “ITD ‘d-q‘g a nag pa + Ze LL9°8% ‘W ‘d 
KS “Wa a t@) “4 "N eee eee eee G) 089'83 “W “Vv 6 
BS "90.10 foymnJ.— avo] O ‘ITM CN eee eee eee e9 6F9'RZ W'V 9 ar i 
= “IBID | I'M 'N | G8PT Ne At 08 G69'8G ‘Wd § ° 
< ‘on[d JO "Gg OSpry—'xvoip “41M ae i $e 9 TP9'86 'N GI 
| "M *§ Ul SuLtveTp— "WIN | “98 “MN oe “ oa 6g GOG'83 ‘HN 'V 8 
3 "M ‘S Ul topunyy—"UIN | “38M 'S on o we G9 G9G'8G ih a eT 
= “MN UT SuruqqSy—avopo “A oy IO YT | “deg ss | “0 = OL BhO'RS "W'd 6 
a UT “19 “d "T'S eee eee eee 6L G6L'8G ‘W'd g 
oO wa 4 (@) “d 1 ‘9 eee eee eee OL C8 SS ‘W °V 6 OL 33 
4 ‘ONT OTIITT JO W0}0G—'N PUT “M *N OF “ULM “ATK ‘dW me ae me 89 GGL°8S ‘W'd 6 
a 0 |] “HUN | 6891 ~ oy SL GZ9'8G “"W'd § 
A ‘dumvd 4Svl ooUIs SuIpusosy—avo/) ‘IN ase ass aus PL LL9°8S ‘WH 'V 6 
| “THU 4 tg “41 a | ‘N eee eee eee 9 G99'RG "WH ‘Vv 9 6 ” 
& ‘LINOSSTP, PUL ON[_ U9eAQoG OSpry—‘avo[D | “d “M ‘'N ” a ng 89 8LP'8Z ‘WN 'd 6 
bd "ULI "119 ‘d 'N eee eee eee 08 GPL'SZ Wd g 
H ‘N pus ‘y 07 Surssed ‘copungy qT “uL103g d ‘q ‘N nae SAS bad 08 IPL'8Z "NZL 
= “HN Pus "TOF "98498 “ITD dG ‘N | SPPL fm ers 9L 869°86 Nv g | 8 cunt 
a En | | eas, | in cee || me SSeS 
= “qu S10yT “POM. *OOTT *poyon,}V “ko. 
‘oy ‘sano1y “pura eyeur LOPOUMOTBE 
-rxoaddy 
“MTLIWOWUTH, TL, 


426 


‘PANUNUIO—SNOILVAUASAQ TVOINOTOUOULAT 


METEOROLOGICAL OBSERVATIONS. 427 


APPENDIX G. 


‘Und puv ‘WIN—"M ‘N 0} Jopunyy, 
‘ayo “VI—" ‘N 0} sepunyy, 


"MN put ‘yf CN 09 Surujpysry 

“a 

"JLOJ VAOQV SoTIML G9—‘AVITH 

“MN 0} “WTO pue Surayy sry 

"M PUB “NI OF ‘WIND “777 

"‘JLOJ OAOGV SO[IUL YF—'AID 

“AUIVOY 10,7 OAoqe sop yg—‘aAvo]D 


"N07 “7B.3g 


“a1 “YT 
“ATO “YT 

*IVOIO 

ULB] 

"MIL0YG 

"M ‘S 0} Yavq—'7eaYs ‘uD 

"WIM "ATQ—‘o[vy 

"Ul0}}0G 0348[G—"1v9]9D—Su0.14s A190 A 

"WIM “ILQ— "ov 

"0}}8[q puv on[g woemyoq aspry—unru “AIQ—‘o[eVy 
"IU “IIQ) (Suoa}s L210 4 ) 


“IBID 
‘WV G 3B W10}9G— "MIM “AQ 
‘OUISLI 
Spholo—"M “S 0} SuIuyYysII—"M “Gg PUG “Ay 0} “THIN 
"YVIYS “AIO 


"M ‘N 0} "}ta3s “aI9 
"AM PUB “GT 07 48 “97 

"M OF “ALO “YT 

“ONT OTP JO wi0x30g—-av91_) 

"IVITD 

‘IQ Apueg Sig—"g ‘g 07 ‘mM -IIQ—"aB9]9) 


cs) 


E 
2 


ca 


[cofical<a—icall<opi<a 


SHS easegg 


BOs 
z 


. . . . . 


MNNNNINNNNNNNNNW 


. 


NANNNM 


a 


G6PG 


$9 
‘asTAUNG 

¥8 

bL 


vg 
"eSTLUNG 


GLE °LS 
P6913 


OIG LG 
90G°L6 
€99°L6 
G98" LG 
LI9'LG 
669°L6 


“GEL Lo 


06L°LZ 
688°L6 
666 '°LG 
G80°86 
666°L6 
9F8° LS 
918°L6 
906°L6 
L616 
L68°L6 
800°86 
86186 
E1686 
T90°8a 
90°86 


660°86 


6L1°83 
PLE 8S 
6EP 86 
G6P 86 
66986 
GIL'86 
69°86 
619°83 


¥G 


66 


GG 
IG 
0G 
61 


SI 
LT 


9T 


i! 


”” 


”? 


9? 


9 


9 


9 


»? 


%? 


9? 


9? 


9 


‘yO Yog pus yyON ‘oSpry Jo ysoug—avo[p ‘d ‘MN 68 68693 
“AVOID ‘dM 8 GEe'9z 
O41 
JO SYIOT YON PUB YJNOG UddAJOq OSpIt UQ—"AVdIO ‘dM SL OFE'9Z 
‘YAO T YJNoG Jo Surssoag—"avoyyO ‘dM ‘9 TZ 889'9% 
wo “TOF “WIE Yuvg—"M “§ PUB “| 07 SuTUZYSrT 0 IL 82993 
Al “a's 89 G89'9G 
pea “et pus 
= ‘TN 02 “MN Woay Surazysi, pus ‘uno yuesoad 
b> 4@ {UI OS's 18 ‘| 03 “M‘N WOU 69 ye “q 0} 
a ‘T'S wow $y 48 | 'g OF "N Woay Suissed sm103g "4S ‘WT e) 8919S 
ee “IBOT) "98 “HL “8S 18 88196 
. i) “WIM “ATQ—"M 07 SuruzysryT | -“d-q'g CL 81893 
i "§ 0} Surytom “Mg pus gop win | “d-q'g #8 IFL'9Z 
= aD | ‘d-w’s 9) G00'LZ 
o ‘Boy Aavoyy | IW A'S 19 89 L80°L3 
id) ‘asTIUNG 
3 "aI9 0 G9 02" | ¥16'9z 
a “WT 0} ULL0}g "38 “HN oy 6L 080°LZ 
=) ‘A ‘N 0} “Und “4, puw ‘UITU “AID ‘d ‘a ‘N oe GL 661° L% 
oe “M 0} “ALD “3]—TBOTD 0 oe TL 160°LZ 
= "IBID IL 'N i GL SEL LG 
= "IV219 | ILM IN 69 G9 O81 LZ 
‘ostung 
ub ‘N 0} Sursra “Hy 0} Jopuny? WIA ‘u1103s AAvOTT 4L‘°N a 89 9IT'LS 
3 "M ‘§ 0} Aopuny? YI ‘w40jg—"WIIU “IND | “IT "MN 08 681° LZ 
— “II) “41 ‘HN G8 G1L6°9G 
A "189190 | “FL 'A‘N 89 966'LZ 
<3 | 
*poyoe 
= *qUSIOH poypoui yy 
oy ‘sanorg “pur 9) eUL *TOyOUOATe, 
i af “‘TULANOW UAH, T, "BFS 


428 


*panu27juuojg—SNOILVAUHSAQ TVOINO'TOUOALA IL 


429 


METEOROLOGICAL OBSERVATIONS. 


APPENDIX G. 


"“IV9TO “d 1 NI see 
"IVITO “a aC) 'N soe 
*IVOTO “d “7 ‘NI eo 
aresnac| Sicm ow i 
‘OTUIVIVT JLOG—"WIN | “d-G'N | 6197 
"THIN ‘ad “AA "NI eco 
ON MMe ta 
"Wd § 9OUIS T1038 AAvOF]T—'mIN | “d “MN iy 
‘M ‘S 0} possed wroyg | “4S aS | GoeP 
0 se 
"MN 0 “und “warm “at |'d ‘gq | Egor 
Pa a ee 
0 ae 
“ICITO “FT a “'N eee 
‘Ivo | «dT 'N | 66988 
*IVITO “d a0 ‘g eee 
"YIOT SOUSIMGT—'g 0} SUIIO}S pu JOpunyy, aT “a ‘Ss es 
tm 'N | oo 
ormg | 
‘M'N OF ‘uN | ‘d-q'g | F9RBE 
“MOTION YSV SAoge sopim gZ—"av9TD | IHN |“ 
7 1801) ‘aeg| oo 
MOTION YSV Wor soplm gI—"M “NO}uND | ‘d-qg | ° 
0 re 
eet | -apeaeg | 
Se at ec <a 
“19 “VY sige). 2° 
“‘W*V J 48 posved 
Urey “Ysa T[e Ajrvou urea pure “M “g 07 sopunqy, ‘MS | 9198 
. ‘MOTOH, YSY—SUISTL “AQ “g 07 Aopunty, CNG 2 


ooo 


G9 
&9 


&¢ 
‘OSsTAUNG 


go 
‘OSTIUNG 


G6 


66 
‘Wd T 
Lg 
“OSTAUNG 


69 


oe 
‘esTaung 


GLI'9¢ 
6E1°9G 
G9T'9G 
LGG'9G 
IOL'96 
61E°9G 
VEL'9G 


666 GG 
008°Sz 
G6L°S6 
89L°S6 
91S°SS 
OLG'GG 
L00°96 
G9G'9G 
POT'9G 
9F L9G 
66096 
LLI'96 


TG8°9G 


61°96 
678 °9G 
L6G°9G 
PLG'9G 
689°96 
L199°96 
899°9G 


LLO'9G 
869°96 


PIL'9G 
LEL'9G 


WV 9 


‘Wd 6 
‘Wd @ 
"WN 'V 9 
‘Wd 6 
‘Wd @ 
‘WV 6G 
‘Wd 6 
‘WV 9 
‘Wd g 
‘lV 6 
‘MW 'd 6 
‘W ‘d @ 


"MW 'V 9 


‘Wd @ 
‘Wd @ 
‘WV 9 
Tag 
‘Wd 
‘WV 6 
‘WV 9 


‘Wd 6 
‘IW 'd @ 


‘WV 6 
‘Wd 


&T 


ol 


IL 


Or 


? 


9 


99 


99 


9? 


99 


99 


29 


9) 


APPENDIX G.—METEOROLOGICAL OBSERVATIONS. 


430 


“UOTPOOATP VY} Ul JopunyT—'M *S 


WOdZ SUIMIOD SpnojoO—'M puUB “AA “SG OF ‘TUIU ‘mMND |'F “q AqQ ‘9 


‘yCog OwMVvieyT 
I9AO **\4 0} SpNOTD osueq—"A ‘N 0} “HE “S ‘Zz SpnojO 
“puta 
OY} YIM AOMOT “TY “Gg MOT vyVaWs 1addq—'e ‘spno[D 
"TT WOU dLOU UOTPDIAIG—"G ‘Suryroiq spnojg 
"M “ 0} 10}qS1[T—SuIsvoo uly 
‘UIRA [RUG 
"YOOID 10}}1G—'N 0} Surm1x0j *wIy 
"G6 (N Pus “M “N 0% “AIO 
iia de es = ls ‘8 
“Ured ‘YI— "i 'N 
WOIJ SONUIME MOF VIOJ PUIM Su0G “MN puB 
‘"M—M ‘S'S Wor Suissed Suryysty yA ‘ur1049 
‘Py M Pas *M “N OF Spno[p 
‘SM ‘'N put ‘N 0} spno[g 
"IBID 
"IBITD 
“ULB “YT 
"M ‘S 0} UlIOVg 
“IN, 


- 


"MN 0} Surorsoy “waTyy 


‘T'S 0} W1I0IG 
“IBOTO 


‘ox ‘Sano1p 


~ 
eS 
2) 


Sssiea PE 
ZAAAAAS 


_ 
fo ah 
= 
A mM 


O18 oj 
Eee 


ri 


oo 

-©& 
Ee 
Mn MMA 


“Pura 


‘VUS10H 
0} BUL 
-Txoiddy 


~ OL TZ 
= TL SL 
7 
ahi OL GL 
o 89 OL 
oe 69 G9 
oe 09 99 
saa = 1g 
ue dss g) 
one cs G9 
asa isa 08 
ses aan G8 
nas on e) 
eee ars 8) 
ae nas lL 
om 69 TZ 
ane ae G8 
eee eee Vis 
08 
$8 
9L 
nie 69 PL 
ne G6 98 
al PL 9L 
POM “ood T “‘pompeyy 
UTLTMWON UH T, 


*PAINU1U0) —SNOILVAUASAOQ TVOINOTOWORLAT 


OLP'SS 
LOG'SG 


L09°S6 
8I9°SS 
LI9°SS 
96h SS 
069°96 
69h SG 
LLO'GG 


66P'96 

19°GG 
T69°96 
€8L°96 
G98 °9G 
FE6°9S 
9F8°S6 
L¥8°9G 

¥8°S3 

08°93 

68°96 
GL8°SS 
G06°SG 
066°96 


"LOJVOULOIB ET 


a a eee Se Se ee eee a 
¢ or 


431 


METEOROLOGICAL OBSERVATIONS. 


APPENDIX G. 


‘A “8 07 Surayysty—avop 
"J 0} SUIMUIOZ STUI0}C 
"IVITO 
"IV9[Q 
‘W ‘d ¢ [1}UN oUTYsuNS ponulyUu0D—'gq “a 
0} “M WOd, UOZIAIOY “HM UL SutUyYSt{T—"M UT «veIO 
"‘pesveod UIvI—"\ “N WO Surtvs{O 
"MN UWlOay oto Gory M ‘urey 
‘N SUI0S “| OI—'g Su10S "4 07 wI101g 
‘SUISVEIOUL “Wf “J 04 Spnolg 
“W] Ul PULM YT UrIeI 
‘souvivedde woi1j—'q “g 0} powu0y wa103s LAvoT] 
"M ‘G UT SOY} YIM UOTJOOM 
-109 UI SuI0s spnopo Aq potnosqo AT[eNavd ung 
‘o0¢ SB YOY sv ‘mr Aavoy—"M “Gg 0} Jopunyy, 


"M‘N wou A[ptder Surssed spnopo oy} ropun yyrap 
—'§ put “M ‘g JOpunyiI—"M “N 0} Surtwmaz0f spnoj9 
‘sn jo ‘q °9 
oy} 0} ‘GY puv ‘g Wor Sutssed spnopo-w.10}9— 
'§ 0} pepesdoad 41 sev SuIsvo1oUl pnopo oyI—"M ‘N 
UL peWlIOJ Spnofd Woy Sutpesdoid -g 04 sepunyy, 
"M CN pue ‘Aq 0} “tano yaeq 
"M PUG “MA ‘N 0} Suisvorour ‘wand 
"M ‘N 0} SULULOF “Yeas puv ‘wang 
"IBID 
ad 
BT JO ‘Y Soptm QT duvg—"M ‘N Mor Butrvoela—atg 
‘M‘N Woay quetim9d raddn—avetg 
‘YOON esa0H puofog espry— szopunyy—'g 07 w103¢9 
"‘SYOIID VS.10F] 
pus 107}IG UW9eMJoG BSpIyY—"MA “g 0} spnojo Lavozy] 
ULBA FYBIS—g “UIIN 
a EN 
"a ‘GS pue ‘g 03 Jepunyy, 
6 “M PUB 'M “S OF “MIU “IND 


ai 
Fee 
Zw 


Si 
| 
op) 


= 


. 


1S 


ae 
Soken 
*% wage, 


é 


od 
WW) WD 


Ho 


ie 
cs 
ZA 


Noy 
= ey 
Bz Zz, 


. 


03 
ae 


— 
A a A oe A 


. 
. 
. 


Q 
. 


NAA 


o 


saeaa 
3 
TM 


SF6 FS 
O91 'SS 
609°FG 
VI9'FG 


969°FG 
009°FG 
009°FG 
G09'FG 
969° FG 


VE9 VG 


TE9°PG 
1P9'PG 
189 VG 


9L9°FG 


199'¥G 
689 VG 
P99 1G 
699° PG 
679 VG 


VV9'PG 
LIL VG 
899°VG 


SILPS 
608°GS 
LEV 'SS 
GLY'SS 
T9V'S3 


‘Hd 
‘Wd g 
WV & 
‘astIuNng 


‘Wd G 
‘Wd Q 
‘Wd ) 
‘Wd 9 
‘Wd G 


‘Wd f 


‘Wd @ 
‘Wd Z% 
‘Wd [ 


HW GL 


‘WV [I 
“WV OL 
WV G 
WV 3 
ove 


‘WV 9 
‘Wd 
‘Wd @ 


WV O08 TL 
‘WV 6 
"WV 9 
‘HW ‘d 6G 
‘Wd f 


&G 


GG 


1G 


99 


9 


9? 


APPENDIX G.—METEOROLOGICAL OBSERVATIONS. 


432 


"H “S 0} “TE ‘ung 

"M 'S pUv “g 07 Sursta ‘T “ATO 

"M PAVMO} SuIstt ‘T ‘AIH 

“BUISVILOUL ST “AIO 

"09 ‘esIquns ‘ToJOTIOMIOY, "g 0} [ “AIO 


"IV9TO 


‘puIM ‘WG Sutmoys “ag 
0} SUIvJUNOM WOI} AOT[VA BY} O7FUT BurutoD eyoug 


VITO 


“a “N 9%} 
WOIf 9Z901q GB S9JVOIPUT “NE 0} SOTIUI 19} ynOGR 
‘otuivad oy} ULI Y ‘posnygrpun AoT][Va o[oy A O44 
I9A0 Suissvd oyours oyj—a0}1vNb 4vyy wor 9A0qe 
9Z90IG YSIIf G ST 919} YLT SOPBOIPUT ‘AA “G OTT} 09 
SUIVJUNOW YT} UO ‘SpoOOM OY} UL dy OSIv] W—"AvITO 
"IVITO 
*IVIO 
"M ‘N Woay AT[VUOTSvODO PUTA Su01}s—a1VvITD 
*IVOIO 
"IBID 


‘op ‘sano1g 


‘TMCS 
"e *g re 
‘Tl -T'S = 
OT a 
‘TOL A 
‘7 ON i 
2-“TN bes 
0 ee 
‘2 OCN a 
7 ON a 
‘2 ON ef 
0 a 
‘OM =e 
mM SP 
TOM" oe 


3 

MN 
‘o°M ON = 

MN 


‘QU SIO] 
“put oO} eu 
a“ xo0.ddy 


66°GSS ‘Wd 
G06°SZ NGI 
G16°S3 ae 
F6°S3 “WV OL 
16°G6 EVE AB 9 
L8°S6 ‘N ‘d OT 


L8°SG “Wd 6 
06°96 ‘Wid J 
66 'GG iets 


G6'SS ‘Nd f 

L6°SG “Nd § 

£0°9G “WC 

F6°SZ |: ey, 


‘Wd g 
‘WV 6 
Wd G 
‘Wd @ 
“HOT 
‘W'V 6 


I ‘ydeg 


1g ‘sny 


*INOJ “£eq 


“pomvyyV 


*LopyoULO IVE 


*WAILANONUIHY, “6P8T 


‘NVGOO HL TAOAV IAGA GAYGNAN ATUNGALUOA LAOGV “AMVWT IVS JO AGTIVA 


*“panuywUoj—SNOILVANASAQ TVIINOTOMOMLA TY 


438 


APPENDIX G.—METEOROLOGICAL OBSERVATIONS. 


‘peireder pur wexo1q Weeq P¥Y YOIYA IdjJoMOItG 


-U19}SIO B YIM Woye, ‘oyVp sSIy} Wop ‘sMOTZVAIOSGG seeeeie ae une HO VAS) GO'9% ‘Wd @& Z ‘00q 
“ABITO Ne ane bak GG 8g 16°SG ‘Wd OT 
: “M pUe *g§ 04 “AT ‘T HN a ae G9 99 068 °S3 ‘Wd 6 
“M 0} “ULI “3 puBw “AID ‘Oo ON ee ig L9 GL G06 SS ‘Wd 9 
‘6 “M'N ig if A 9L GC6°GG =| W'd O¢'G 
“MN pUe “AM 0} “AIQ—"g 07 SurysturMIp ‘wn “SoM ie Eee BL G8 GG6°GG ‘Nd PT 
"g 03 AaT[VA JO SepIs Y}0q WO sULeyUNOM UO ‘UND 57 AN NG see “a0 08 98 96°G6Z -| ‘Nae 
Go OM a ne SL G8 86°SZ ‘Wd [ 
“A “§ 0} SUIBJUNOUL pulyeq Surui10j ‘uinD ‘T'S ig =e GL 68 G10°93 ‘WM GT 
hha saa aes §9 OL V6 SG VG 
“TC aks a 6G G9 66°GG a Vas 
“G OL oe ap GG GG 06°SG Vo), 
“Ayoug ae Gh re Se LY 1G 88°GG EA 9 LE oss 
F “UIT NT “6 Ss ws — $9 69 G8°GS “Wd 6 
“G “g 07 possed spno[g—'T “a1p sea bak va hs G8 68 68°GSG ‘Wid § 
“ULOYS ‘T'M‘S = eg 69 Jl G18 "SS “WH GL 
“‘UVIPIIOUL UO pus “A 7B ‘osTe “BUIUILOJ Spnoypo {BG 
0} SUIvJUNOW jo syved uO SurmMAOF ‘IND—'Y 0} & “AID "2'g nay =v LL gL F3°SZ ‘M'V 6 
“MN Pus °§ 0} G Spno[p 6S rity sa 99 L9 61°S6 ‘W*V 9 OL » 
‘§ 0} MOZTLLOY UL LBITD—G “UITN ‘$ 2 'S site ne L9 GL G8°GSG ‘W'd 6 
“M‘S Worf SULLOPS ‘6 M'N ans “ee 98 G6 68'S “Wd § 
‘6 U'S aie a 08 68 06°96 “WV 6 
: SG 'S 0} Spno[p Aster Gl és Bae cL GL L8°SG “WV 9 6 » 
“SUTYBIIG Spno[p "S “GN eS ee G8 §8 68°SG ‘Wd OT 
“SULYVeIq Spno[y 6 AN oa oe 98 68 G8°Gsm ‘W'd 6 
°"G “UIN Sea <a ae 08 68 68°SSG “Wd g 
"M ‘N St ‘oT[VA 04} Jo sopts o71soddo uo ‘surey 
-UNOW 9} Zuo[e pur oy,—"g ‘N pus ‘gg 0} ‘WIN ‘6 “AN sd a 98 §8 68'S ‘Wd p 
“M 0} “UTU *uOny) ‘6 ON sai Soe 98 66 G28'GG | ‘N'dg 
“MN pute “MM OF “OID “4IK) ‘T’M'N “i ad 66 cOL G88°GzG ‘Wd G 
‘6 “M‘'S PUB Gg OF “AID "S “M°N a og &6 GOL G1L8°93 ‘Wd f 
“MS Wor 
‘AS OF So[IUr 19} Ynoqe pum “M‘N OF ‘TE “WIND |"S'M‘'N’M er ao 6 S0L 06°SG ‘Wd § 
‘ysiy spavk goz “MN Wory Surposs 
-O1d SpUIM[IIGM Susty “yg 0} SUISvOTOUL “tUnY) "2" M eas “> 96 FOL CI6'SZ ‘Wd Z 


28 


APPENDIX G.—METEOROLOGICAL OBSERVATIONS. 


434 


“"N 0} ULed “9, GPT “pH “UITN 

: 6 “WIN 

; 6 “WIN 

“SUIBJUNOU OY} UI UIGI 10 MOUS—"g 0} ‘TU Lavo 


‘pS ‘WU | “Gg “g WO’ SuONs Surmolg 
"AvIIO 
"M pus *g 0} Surus0f “mUIN 


: “a0 "FT 
"IID “YT 
*IIIVITO 
"AID "YT 
‘N wou parvoddestp ‘ur “110 "7 
‘LON 0} Surura0y ‘uarra *a10 -yT 
‘Z ‘t19—Suriveddesip spnojg 
WOZILOY 
oq} UO you '| “N pus ‘| 0} Surursoy sum Larvoypy 
‘DS put “q “N 07 Suymaoy “win 
"N 0} Sur0S -urar pue ‘wiu “119 
"N 0} Su103 ‘aru puv wu “a9 
"MA OF} G SCRE “aI 
"§ OF “AID 
"g 03 APsour ‘Zz “119 
"S 04 § “AID 
G “S 0} “AID 
‘N wor Zurrvoddestp ‘g 03 “a9 
‘Bupyuis “Ni 0} g “WIE “AIQ—"T “a1 


‘op ‘sano1g 


*Ppanu2qu0oj—SNOILVAUTSIO TYOIDOTOUORLET 


. 
re 


“PUTA 


“‘9U310H 
oy8 ur 
-rxoiddy 


“POM 


‘UILIWOWWAH T, 


LL'GG 
08°SS 
08°SZ 
SLL°SS 


G6L°SG 
GLLSS 
08°SZ 
LL’GS 
ELLGG 
018°96 
GI8°Ss 
&61°S3 
98°SG 
L8°S6 
98°S6 


G98°SG 
G98°GG 
L8°S6 
L8°S6 
68°96 
06°SG 
16°GSG 
G6 SG 
G0'9G 
00°96 
96°93 


*IOjOMLOIET 


‘WH ZT 
‘KV TT 
‘NV 6 

“OSTAUNG 


‘H GI 
‘Wd 6 
‘Wd Q 
*posung 
‘Wd @ 


W ZT 


‘WV G 

‘estaung 
‘Wd OT 
‘Wd 6 
‘Wd Q 


‘Wd ) 
‘Wd 9 
Wd G 
‘Wd > 
‘Wd @ 
‘Wd % 
‘W ‘dT 
WCL 
‘H'V OT 
WG 
‘WV g 


“6F8T 


Te: 3s 


08 9 


“heq 


| aon 


435 


ial 


APPENDIX G.—METEOROLOGICAL OBSERVATIONS. 


"S WOT “UMIM “AID 


"AI][VA VOY OY} TOAO sozvy 
[PAS] Ul popuszxe ‘aovjins 9A0ge 4OoJ OGL ynoge Soy 
"M “@ WOU Surur09 fe ‘uM “AID 
‘MOZTLOY “G Ul SUIMAOF Spnojo—oor Surzvoly 
‘uns 94} puNnot [VY 
QSIVI—2YZV] JO OOVJANS TOAO JOYOIY} Surm0seq Soy 


"JSOIJ-IVOY YIM PdITBA0D SaTpog 
IV—'T “400 ‘are url Suyvog oor jo sopyavd jeug 
‘aYVl WO UMOP MOT SoJ—avo]O 


*N W101} Soy Lavapy 
"N Wuoay Soy Lavopy 
‘N WLOAy Soy AAVOFT 
"N worl Soy AAV] 
"N wloay Soy Aavop 
"OYVl WOAJ Surm0d Soy -y7 
“Uda[[VJ SVY MOUS Jo J[VY B puv YouT uv ynoqy 
— ‘dx ey} JO odusins uo Boj yySys YM ‘IBID 
"Uvseq JI WIM “AA “GO WOT 
SU01}8 puUIM—siNoY OM} 4Sv] OY} Suramp Surpey Moug 
UIRY 
"m0T}00ATIp ATA UL ‘SUIeJUMOW UO MOUS JO UIVYy 
"G “WIN 
"§ “OLIN 
"G “WEN 
“MOTO 
‘§ Woy -yea38 seddn—"M ‘N Woy Surmo0o spnojg 
*“§ “ULIN 
‘6, WIN 
“TN WlOdy pUuIA YIM ‘u1078s Y 
“A ‘N WlOdy purA YIM ‘o10}s B OG 07 M90S p[nOM 
| OU T—"M "N °F "MG Woy Apyomb Burst “wy 


on 
En 


xi 
= 
Z 


J 
oz 
Zi 


VA A Aa 


an 
Se oO 
i 2 


a 
Sea 
Due 


mae 
joo 


nh DAHA w 


a 


aoe 


g°98 


96 
8G 


63 
06 


GG 


0G 
91 


LG 
Ié 
6S 
08 
66 
GG 


6G 


&& 
o'Lé 
Iv 
L’vv 
bP 
OV 


LE 
LE 
68 
LP 


GOP 


6G 
0g 
&G 


GG 
GP 
LY 


9V 
6P 


GY 
&V 
6G 
g¢ 
GG 
vg 
LY 
6g 
09 


8P 


0g 
GG 
69 
69 
GG 
69 


&9 
vg 
Gg 
Gg 


8P 


GL°GG 
9L°S6 
G8L 9¢ 


08 °SG 
G8° SS 
668 SG 


G6 GG 
90°96 


696 GS 
66 GG 
90°96 
809% 
90°96 
P0'9G 

L90°93 
OL°9G 
L0°96 


610°96 


G98°96 
08°96 
TLL°GG 
GOL'SG 
08°96 
L8L°GG 


OGL'GS 
£08 °S6 
61°96 
LL’GS 


GLL'SG 


"Md ) 
"Wd 9 
*yosung 


‘Wd 
‘Wd [- 
‘W GT 


‘WV TI 
“HV OL 


WV G 
‘ostaung 
“Wd GT 
“Wha 
“qosuNng 
"Wd @ 

“W ZI 
sa sa 
‘HV 6 


-astiung 


‘Wd G 
“yasung 
‘Wd @ 

“WGI 
BATE 
“WV 6 


"OSTIUNG 
“Wd GT 
‘W ‘d 6 
‘yosung 


‘HW ‘dg 


co 


MD 
z 
} 
e 
AH 
=< 
> 
(om 
RA 
mn 
ise) 
fen) 
| 
< 
io) 
et 
ido) 
ro) 
eI 
=) 
pS 
=) 
ie) 
H 
& 
j 
O 
bd 
ke 
A 
A 
2 
i 
Ay 
<j 


436 


*S[VAIOJUL 4B Sjsn3 Lavoyy 
"IID 


"MN Woay Sutstt ‘ura Lavoyxy 

‘TN ‘Barurr0g MOU GIBIIG—"M “N WOAZ Iva] 
‘6 “MS Wosy “wad “ITD 

“M PUs "MN 0F “WIN—"dozoy *g Ul ABETD 
“IBITO 

"g “MIU “IIQ—"g Surree[p 

"MS *§ “WIU “AID 


“MOoug 
“WI10}8 04} JO JUOMOOUOU 
-M09 OY} SOUIS MOT[By Soyour ¥{T ynoqe {Mous 47 
“MOuIg 
“MOoug 
"MOug 
"MOUS “VT 
*g “WIN 
‘MOUS “4/'T 
“MOUS "QT—'G WOT, Buravs[yH 
‘qysIU OY} Sutanp WoT[ey svy Your % qnoqe ‘sous “4'T 


"g “UIIN 


“op ‘sano19 


Nal cs 6 


os 
E 


“PUL 


NNNNNNNMNRM 


‘qIS1I0H 
oy BUr 


-Txoiddy 


GS 


"JOAN “poypseyyV 


‘WaLANOWATH 


"PaNU1}U0) —SNOILVAUHSAQ TVOIDOTOUORLA 


GL°SS 
98°93 


£06°S6 


£68°93 
88°96 
88°96 
98°96 
68°96 
06°96 
88°96 
06°96 
89°SG 
L9°SS 
69°96 


601°93 
089°SZ 
LI9°G3 
OLG'SS 
ELG'GS 

89°96 
009°S3 
89°96 
G19°S¢ 


EL°9G 
GL°SG 
GL°SS 


“LOPOULOLV 


NCL 
‘WM 'd 
‘Wd ¢ 
‘WL 
‘WY TT 
‘NM 'V 6 
‘esTAUNG 
‘NM 'd 6 
‘yasung 
‘Wd @ 
"WV 086 
‘yosung 
Wd @ 


‘NV 6° 


‘OsTAUNG 
‘Wd TT 
‘Wd @ 
‘yosung 
‘Md 
"Wd T 
“NCI 
"W'V 6 
W "Vv G 
‘estaung 
‘Wd [TT 
"N'd 08'6 
“N'd 0§°8 


v 


9 


99 


99 


~ 99 


437 


APPENDIX G.—METEOROLOGICAL OBSERVATIONS. 


‘IOYSIY Spnojo—urerl “VT | “TM ‘N 24 LE 6& Gg GP9'SS “H'd T 
‘OL pus g moe9M40q 
peyoeyo yUoTMAAOM puvaMdn ‘ATpideva Surpey Mou avg 0 sit OF OF $S G69°GZ ‘NZL 
LS) ne G18 8& Gg EGL'SG AY sie 
“QAOGB 499J YOZ WN0Q’ ‘UMOp MOT Surms0F “yVayg TS pol aie Lg ge 9G SIL'SZ ‘N'V OL 
“Tred A4STIA 27e te ie cg Gs 6G 9PL'SS SLVSs | Line 
*S10MOTS UIVy ecocee eco ; eve 9¢ 6g L08°S3 "Wl “a CL 
“ULe yy ecovece eee wee 9¢ 8G 9Z8 SG ba /\8 ‘d {Lik 
09 668°S3 “Wd QT 
UIIN | ~T 2 N ee 98 L& 69 68°SG “Wd 6 
‘urer Apvoys ‘}T ‘T M'N i oat 1g 09 GI8 "GG ‘NW ‘d 8 
‘sy[enbs urey | “% “M‘N a 68 OF 9¢ 61° ‘Wd ) 
G6 M'N or 68 GV 09 LL°GS “Wd 9 
‘0e'F ynoge ‘nN Aq 
‘M WOdJ Sard ‘[[ey JOYS ‘rey pues uleY—"p “WIN 6M — 68 GP 8g 9L°G6 eres 
‘T 0} SUIVJUNOM I9AO pus ‘AA 0} “WIN, PM a OF 9V Gg 9L°SS ‘Nd 7 
"MN pus "(4 07 SurezUNOM to urIey ‘pb 'M sf &P is] a2 LEL'SS ‘Wd 
‘DUISVOLOUL UOZIIOY “AA Ul “7BI}S “UND pM e &P GY vg 66L 96 ‘Wd @ 
regan 
Uv} “4[ SLOW SUTeJUNOM IOAO Spno[gQ—'suoIseI 
zoddn ul Ulivi Io Mous YIM “MN 0} “4es ‘WIND b'M we GOP ie) vg GEL'9G “Wd T 
‘yerjs ‘wanD 7M 'S. ae TP 0g Lg 61°96 “NGL 
*4e1}S ‘Tan Aa § ly G'eP SP 9¢ 08 GZ “HV TT 
"48149 ‘UND 7's be OF SP 9g G6L°SZ ‘HY OT 
"M Juermn. todd q—" 4 
0} Ivel[d “J Ul SUIVJUNOM JEAO SurmI0F 4VI4s ‘WIND Geer gi 1g OP vg 8L°9G “WV 6 
"M Wor ‘e 4vays ‘mang ‘9 "a 2 Lg 68 6g 9L°9G SSMS OL 7 
19 61°93 “Wd GT 
‘T'S ae 68 GV 09 9L°SS "W'd 6 
‘M WOT, 910M SuTUI0D spno[Q—‘10}sIOM ,7 oTIBIOq 
A[mappus ItV—"M‘N pue “M UL Lavoy “yeays ‘ump ‘T‘'M'S oes GP GP v9 GIL'GS eens 
"M 0} WOZTIOY Ul SuIMIIOJ g “4RI4S “IID yO wns CP SP 8g 00L°96 ‘Wd g 
8 ‘8 Pe 68 GP vg L69°SG | NV OS OT 
“M ‘N 04 
Suludoy ‘uND—"M “g wos A[pIder Surssed ‘wre “a19 b'S a OF 9F 69 689°SG te ME 
‘9 ‘WIN pg eee 88 ep 0g - 89'9G ‘OSTAUNG | G ” 


“SUIVJUNOM IOAO “KT *§ 04 g ‘WIND 4g GS GP goog «| ostaung | FT 5, 
‘y “WIN on 99 LQ"SG He TT 
"S (N 0} ‘Watt “ATO - 69 ¥G°GG “Wd 6 
"N 0} Ivo]Q— "| “GS pus ‘g 07 ‘wu Lavoqy i GG GG’GZ ‘yosung 
Lie | a ie ‘Wid § 
wa ai 99°S6 Sc 
Zi ae 6G°Gz Nd T 
a ort 19°S6 WEL 
= * £9'°9% «| 'N'V O8'IT 
= a 69°96 ewe Ok 8 
ce ‘Wd Z MoGe “47 “A ‘N 0} poSueyd pur 
wa 84 “WOlj}eIIp vy} UI W8eS SUA ‘MOUS TIT ‘TOTU 
a AAvoy pus ‘'g 0} UAALIP 910M SpnoyoO oY, ‘“MOTAq 
}PF SBA FI BLOJoG OUT} oMOS suOIsoa aoddn oy} 
= ur [fvaoid 03 porvodde pura ‘A ‘N O4—'N wWoaz 
te SULIve[D “YSIY JOYVI “ATO ‘WIND YA “Stids0w uy sae + )'9G ‘NM ‘d 6 
g e ese $3°SZ “WW OST G 
= 67 ¥8°9G Hd § 
2 “SUIVJUNOW IOAO SUIMIAIOF “yVIYS ‘UND ‘ae ae 13°93 ‘H ZL 
=) “IB9T) FG &G 688°96 Bona) 
2 “voi 1% +P cgog | “ostaumg | Zt 
=e 6g G69°S6 “Read OT 
A 02'8 9B pUIM Jo sTrenbs 
AAvoHT “UdT[CZ SvyY YUL ouo ynoqe—pesveo Moug oe vA) 089°SZ ‘NM ‘d 6 
6 ‘MOug Sn 9 899°S3 "Wd g 
bd “MOUS om &g GPI'GG ‘K'd 9 
H “MOUS - 0g 669°96 ‘posung 
= ‘OST SUN—AMous 47 rs 0g OL9°SG ‘Wd g | TT “Ue 
S Eee Lee | Se eee | ere Ree | Pls eer 
a "qyStOT] bikes et sete ee 
‘oR ‘sano1g o7eUL *TOPOULOLE 
-rxoiddy 
‘UALAMOW UAH, “OS8T 
co 
a *PanurzzUoO—SNOILVAUASAQ TVOIDOTOUORLATL 


METEOROLOGICAL OBSERVATIONS. 439 


APPENDIX @G. 


"IBID 
"IV9IO 


‘S ‘AN SGM pUIM of} JosUNS preMO}, ! UOZTIOY “A 
0} Joqaenb 4vy} Wo; porjor AT[eNpVas spnopo oy 
“MN Wosy Surreejo yng “Z “| “g PUIM oY} “Wa g 
‘posted MOUS pus ‘Jet ey}? puIM ‘fziey vw pus 
InOoY UB YNOGe UWOTeEIIP SIy} UI MITq IT “499TS 
preg JO 1az0vAvYO OY} 9LOUL Suruinsse penuryu09 
MOUS OUT, “G"M‘N PABY MOT 03 UvZoeq 4 “Wa 
08'S yHoge yy ‘A[LAVOY SuTT[VJ Mous ‘wjVo UoTy 
“MS 0} potooa AT[VNPIS puIA oy} {FG pUIM 
oq} YIM “WV [][ Jnoge psdouUeNIMO0D wWI0}s MOuG 
"Gd “SG 0} esuop ysour ‘G “WIN 
‘YSU 94} Suranp spews & MoTq 
‘JUSIUpIUL 10438 UOOS “| ‘Gg Moy pavy MoOTq 0} ULZOg 
aN, 


"IVO] 
“OOVJAINS DAOGB 499J ONOT ‘oaoqe ‘yosuns 1097% : 
MOOS “SUIMAOJ Spnopd peyyrayg “g 04 “48 “AID ITVaTD 
"N 0} 1B9ID 
" *M 0} IB2TO 
‘MOUS “YT 
"MOUS “YJ 

“MN S2[TUr Og Urey 

-UN0W YSIYy v MoI Zurpssoo1d anodea jo umnjoo 
% oq 0} savodde uoezjo “MQ “N 0} Sutma0g F ‘wd “4g 


MW 


. 


Ae waa 
TH U2 


ica 
a wD 


Aw, Z, 


PEEP 


E 
Zi 


&T 
9T 


66 
9% 
G'9G 
GS 
GG 
0€ 


GGG 


TS 
06 


LG 
9G 


VG 


86 
0& 
08 
LG 


8G 


GG 
67 
&G 
vg 
8V 
6F 
IV 
GCP 
0g 
GV 
&V 


8EL'GS 
6EL°96 
GL°SG 
0L°S6 
99°SS 
¥9°S6 
09°S6 
89°SS 
89°G6 
8S°G6 
69°G6 


O19°SG 
L09°S6 


TL°S6 


GL°GG. 


IL°S6 
VL°S6 


91°96 
LL’SG 
OGL SG 
GPL 9S 
61°96 


STL'Ss 


"WN ‘d 6 
Md Q 
‘Wd J 
‘Wad @ 
‘Wd f 
‘Wd @ 
‘W'dZ 
‘Wd T 

WGI 
aT Ty 
“NV OT 


‘WV G 
‘N'V 8 


‘Wd ZT 
‘Wd TI 
‘Wd OT 
‘W 'd 6 


‘W ‘d Q 
‘yosung 
‘NW 'd 2 
‘Wd T 
weil Le 


‘MN 'V 6 


ST 


9? 


‘ouy AIA MOTI 

‘aug A19A MOG 

‘aug AIVA MOUG 

‘aug A19A MOUG 

‘ouy AIBA MOTI 

“g EN 

"M ‘N 0} Sururs0g “wry 

"IVITD 

"M 0} SuLIve[p 

hy EIN 

"Sg 0} ‘ulIU ‘*NY 0} “98238 “wd ‘yZTUOZ UO *AI9 
‘aSnVd JUIT 

-edde Aue ynoyq}IM ‘ornjeroduie, Ul UOoVeIAIvA WONT 

“M 'N 

‘pf ‘WIM EG “4VI}S “IID *N 0} {OA0qe PUIM "Ay “N 
"PUM *§ YIM APUOpIAS 

‘g 0} SUIGJUNOW UO ‘wMIU ‘UOZIIOY “A “N UO SuLtvelyO 


‘Ud][BJ SVY SOUL g YnOGe :pesved Moug 
“ATIAVOY SuLMOUg 
“MOUS *9/T 
7 aN 
"MN OF “M YSnory} ATC 
-npvid possed pura {oe', 78 peouemM0s Mous “V7 
"p “UIIM “IID 
yee a) 
‘SUIBJUNOUL UO MOUG—'|] “G 0} F *uNIU “AID 
"IBID 


mM 
A 
(2) 
HH 
AH 
<x 
b 
oe} 
ea) 
n 
Q 
co) 
a 
< 
>) 
KH 
id) 
=) 
PS 
=) 
iam 
(=) 
is) 
aH 
ia 
T 
ido} 
4 
Lal 
QA 
ZA 
isa 
Ay 
Ay 
<< 


“oO” ‘sano1D 


440 


ana 
See 
iz, 


nD NNN ANDn Annm 


a= 8] 


5 
coc] 


re re 
. . 


*PurM 


‘qq S101 
07 BUI 
-Txoiddy 


"JOM 


96 6g 
9% GP 
a ZG 
LE 6G 
6G SP 
L1G VP 
“s 60 
LT 6G 
0G GG 
LG 0g 
86 GG 
8G 6G 
0$ §gG 
0G 0g 
GG GP 

§G 
8% eq 
&%@ $3 
PG GG 
GG v3 
8Z SP 
63 67 
GS 0g 
LI 69 
6 66 

‘very =| “poqony}y 
"WALANOWUAH I, 


“SNOILVAUASAO TVOIDOTOWORLAYA 


CEP SS 
80F'SS 
SOP SS 
9GP GS 
CEP GG 
66h GG 
LE9'GS 
099°S3 
OGL°SS 
608 °SZ 


_ 08°Sz 


968°S3 
968°GG 
G08°SS 


GOL'SS 
699°GS 
8G9°GG 
089°CZ 
009°SZ 
089°S6 


689'°GS 
O19°SS 
€99°SS 
00L°SG 


‘$69°96 


“IOJOULOIVET 


‘Wd G 
‘Wd g 
‘Wd T 
‘WV TL 
‘WV 6 
‘WV Q 
‘Hd OT 
‘Wd 6 
‘Wd G 
‘Wd @ 
‘Md % 


‘WV OLIL 


MY TT 
‘WV 6 


‘asTAUNG 
‘Hd GI 
‘Wd TT 
‘Wd 6 
‘Wd Q 
‘Nd G 


‘Wd 
"Wd T 
ADVE TL 
MV G 
‘asTUNg 


Sl » 


2 


APPENDIX @.—_ METEOROLOGICAL OBSERVATIONS. 


441 


‘Tyeus Ata {MOUS “ArT 

"N 0} YIVp °° 0} rvepo uozi0y { Mous “4°7 

"M ‘N 0} “98 “AIO 

“SSOT “M “N 0} “98 “aT 

"ae9[d ATOezred *g pus yyIUEZ {°M “N 0} “ATO “VT 
‘PM 'N }Uorano z9ddp 

"M “N Woy woo szeAo A[prdea Sutpesd0ad ‘uu “110 
"M ‘N 0} Joyarup ‘wiu 4] ‘ perzeaoo Axg 

G YS “AID 

"N 03 “998 ‘and ! 1BETD 

"N 03 9S ‘mInd favelO 

"N 07 “98 “wand 

*N 0} SurITvepo { “NT WOT BuTUIOD aAoqe “AID 
9A0QB Sulywog ‘adi Jo sepotyaed 10 ‘Mous +47 
“SUISVOLOUL “NT OF ‘WIM MOT { MOUS “47 


"P02J OOST 
qnoqe ‘surejunom jo opis uo dn ySiy Sutims0j 49 
"MH ‘N Wody CACM YOIYM 4S “110 Aq poras0d AG 
"N 0} “UUIN, 
‘SUIVJUNOU UO UMOP MOT SUIUIIOZ “YC 
*YBIYS “IID PUB “TIO 
“MOUS "YT 


“Ayureyd uses ‘oom Mojeq poyeanyes oroydsom 
“38 24991] Moped osngtp Aq popunosans ‘a[qrstA 
Ajureyd st uoom ey, ‘A[IAvey os you 4nq ‘BurMoug 
“MOUG 
‘BUl}jIIp ‘ouy Moug 
‘SuljIIp ‘ouy Moug 
‘Sujjiap ‘oug Moug 
‘Suljjiap ‘sug Moug 
‘Sulyjlup ‘euy Moug 
gn 
*MOug 
ye 


16 <H 


lcallca—ico—icalicaplc 


NHintnna 


MMMM 


eae Ss 
ne) 


nHNA 


a ae 
aeons 
BAe SaZeeeee 


is 
Pens Sic 
7 el j 


OO SH SH SHH 16 16 09 
EE 


EEE 


ra 


: 


EEE 


= 
CA A a A A A 


O1S°SZ 
OLS°SZ 
00L°SZ 
86L°SG 
09L°SZ 
BLL°SG 
O8L°SG 
G8L°SG 
68L°S6 
GG 
LL’GG 
LL’GS 
08L°SZ 
OLL'SS 
GLL'GS 
GLL'SS 


L6L°SG 
861°96 
OFL'SS 
SIL'GG 

89°GG 
099°SG 
09°93 


G9G°GG 
OSh SS 
OGP'SS 
168°SZ 
088 °SG 
0&8 "SS 
008 °SZ 
00F SS 
O&h SZ 
SEP SS 


‘Wd g 


‘Wd ) 
‘W ‘dG 
Wa @ 
‘Wd T 
HV TT 

Ot V. OGEG 

"WV 08° 
Hed TT 
Wd & 
‘Wd J 


02 


61 


99 


99 


APPENDIX G.—METEOROLOGICAL OBSERVATIONS. 


449 


~ "IBOTO 


“M 'N 0} “IN 

‘Pp ‘TS WoIy Surmoo ‘7 ‘wu “119 
“O[QISIA WOOL {4S TOMO] UL MOUS “4/7 
“MOUS “}'T 


‘und “A. JO uONdooxXS WIM 
‘gpno[o 4sn3 WoOzZT10Y "M—'p “WIM “IID "Gg 03 ABIIO 


"MOUS "9'T 

‘morpoysed a[qIsIA £¥}VI4S TOMO] “IID { MOUS *y'T 
“MOUS "9'T 

‘MOUS "9/T 

“MOUgG 

‘PM ‘“UId ‘O86 ¥—'G “WIN 

‘p “WIN 

‘7 “WIN 

‘p ‘WIN 


“MOUS "4'T 


“oy ‘sano1p 


*Panu2}U0)O—SNOILVAUASAQ TYOINOTOUOALAT 


“Pur 


‘qq510H 
oyRUL 
-Txoiddy 


G'9G 


“MTLIMOWUAH YT, 


68 
og 
0g 
GP 
0g 
Gg 


Tg 
GP 
bP 
SP 
LY 
6P 
0g 
SV 
GP 
OF 
8v 
69 
Gg 
Gg 
0g 
6P 
tg 
0g 
~8P 


“poqonyyy 


GLESS 


069°S3. 


G6G'SG 
00S°SZ 
8IS°Ss 
LOG"GS 


eercee 


009°SZ 
GV'GS 
GP'GS 
VY 'GG 
9F°SS 
LUGS 
S9P SS 
OFF SS 
SOP SG 
068°S6 
GLESS 
098°SZ 
098°96 
68°96 
91h GG 
GV GG 
SP'GG 
9F'SG 
06P'SG 


“LOJOULOLE ET 


WVOSL | 86 
‘Nd GT 
‘Wd TT 
‘WM ‘dG 
Wd 
‘Wd G 
‘Nd 06 P 


‘Wd f 
‘Wd @ 
‘W'dZ 
‘W ‘d [ 
‘KGL 
‘HV TL 
‘HV CT OL 
‘WV OL 
‘NM 'V 6 
‘WV Q 
UAVEE 61 GG» 
"W'd TT 
‘Wd @ 
‘Wd ) 
‘Wd G 
‘Wd g 
W'dZ 
‘Wd T 
"W GL 
HV IT | 1g “wer 


“IMO; “keg 


“OS8T 


Coe ee 


448 


APPENDIX G.—METEOROLOGICAL OBSERVATIONS. 


‘[ereze]1Ipenh “epg ‘eSavy soyep—Aous “47 
pee 
ew 

2359) 
"S “M WOTF “AT 
"Wd G 48 ‘MOUS “47 ITM “MA ‘G THOT syenbs {F ‘wIN 
‘9 0} Iv9e[D UOZIIOY :e “MIU “IID 
"M‘N 0} “4S ‘umd Se “My WOAy Surum0d “wu “IID 
"G “UATW “AT 
G “WIM “AID 
“§ “UN 
"M ‘S UL Suarsra 
“q 0} duissed ‘mmo Lavoy {°g wWouy LOMO] “IU “ATO 
"MOUS OULT 
"PM Yaer1mo asddn ‘wu 119 
‘9 “WIN 
‘aug. ATAA MOoug 
‘oug AIOA MOTIG 
‘@ THIN 
"G."OATN 
"GROEN 
"GTHIN 
"Gg “WAIN 
"N 0} ‘watm Lavozy 
“MN 0} f “UdTO “aI 
*S “UIU “21D 
‘g “UIT "aD 
"MOZLLOY 
punor [[@ Sururzoy 448 ‘und fe “MQ 03 Bur0F 47 “arg 


> 


"IBOTO 


“§ 07 ‘877 Jo doz wo ‘warm £°4Q *N 09 “48 ‘AND {28919 


ai 
OB NN of ad cd 
nee uiww 


- 
Sig 
Fae 0 


>OOH SOSH 


1S 


We) 
ND RNR RMN ANNNDBNADDA 


1G 19 19 15 


st 0 


‘$ “H'S 
vy a's 


GS 
LT 


69°SG 
O0L9°SG 
0L9°SG 
OF9°SG 
OPL'SS 
OTL'SS 
099°SG 
089°SG 
GOG'GS 
OSG°GS 
OLG'SS 
089°S6 


0S°SZ 
OFF SS 
OLE'SS 
068 °SZ 
8GP'GG 
09F SZ 
S8P'S6 
O1S'SS 
8G9'SS 
0&9°SG 
069°SZ 
O&S°GS 
0&9°SZ 
GTS'SZ 
O87 SS 


PLY GS 
GPSS 
8h SG 

06h 'SZ 

069°S6 
089°S3 
819°Gs 


‘KGL 
‘WV [I 
‘N‘V 6 
‘WV g 
‘Wd [I 
‘Wd 6 
‘Wd ) 
Wad 
‘Wd @ 
‘Wd 
BESET 
"H'V 6 


‘H'V Q 
‘Wd [T 
‘MW ‘d 
"Wd 
‘Wd g 
‘Wd @ 
‘Wd T 
x ee i 
“WV OT 
‘M'V 6 
‘N'V Q 
“Wd TT 
"Wd 6 
‘Wd ) 
‘Wd @ 


"M'd @ 
M'd ZG 
‘Wd T 
“WEL 
‘W'v TT 
‘HV OT 
WV 6 


93 


GG 


VG 


29 


93 


99 


16'S ‘Nd > 
86°93 ‘N ‘d § 
F6'9S "W 'd 
96°S ‘N ‘d T 
‘ 16°93 ‘W ZI 
Bs 00°92 MATL 
iS) "I "98 “aI9 ‘TM €60°9 ‘WV OT 
a "IB91Q) SM £0'9 "NH 'V 6 
= "VITO ‘TM 00°92 WV8 | 82 » 
6a 10°93 ‘Wd 
Ll "IBITO ‘IN 80°93 ‘Nd TT 
fa "IVITD ‘l‘’M 10°93 "Nd 
24 "IR9TO 0 10°9% Wd) 
> “IV9TD) 0 G0'9% "Wd G 
=) “IB9TD E49 £093 ‘Nd g 
S "IB9TO Ts £0°93 ‘Wd T 
2 “IID “}T "TS G0°9 WW TT 
) "IBID 0 86°93 ‘N'V 6 
a *S “WIN IID | “TM ‘N G6'9G ‘W'V 8 
I ‘TM LI8°Sz GET || LS. 2 
es *sunoy VG SVT UdT[BI oABT SoqouUr © }noqGB ‘Moug PAK OL8'SZ ‘Wd ZL 
SI | ‘moug | ‘Z'M‘N FLL GS ‘Wd TT 
| "M UW1OIZ % “WAIU ITD ‘T'M OIL'SZ "Nd & 
‘y}iuez ur rvoo LAYS {sexe ore, ‘Moug os ae 089°Sz ‘Wd ) 
od *M UT “IID 0 0S9°9z "Wd G 
bd 19°SZ ‘Wd § 
a 19°93 ‘Nd % 
zi 19°93 ‘Wd T | 93 “UBe 
Ay za 
Ay 
<j *InoyI “keg 


0” ‘sano1p “pul *ojomMoIEg 


“AALANWONUATHT, 


ditt 


*PanuryUu0j—SNOILVAUASAO TVOIDOTOUORLATY 


445 


APPENDIX G.—_METEOROLOGICAL OBSERVATIONS. 


“IB9TN ‘L's ae GI GT 0g 026'9Z ‘MV. 6 
"IB9[) ‘LS eh oL oL GP 80Z'9% WV | IL “40a 
09 091'9Z ‘Wd ZI 
“IBID L'a oh “all SL 0g 091 9% ‘Wd T[ 
"IVOTD 0 cra ial ial YAS OSL'9Z ‘Wd 6 
“AOT[SA UI UMOP MOT “4S “ILD “47 1 a be LI SI ZG LEL'9% ‘Wd / 
"IV9TO 0 . SI 61 0g PEL'9Z "W ‘dG 
"N98 “AIO “YT ‘SM ie 8% ¥G 0g GIL'93 NM 'd g 
"IVITD | “SM 'N ue 8% ¥Z LY ZL'9 ‘Wd T 
= “AYOWS LOY}VA AS ! AvETD Toe on SI SL GP O9L'9Z ‘WV QT 
"IBOTD os cae 9T of ep OSt'9g | ‘N'V6 
‘qsoujavoy ! 1velp 0 a iE)E IL GP O81'9Z Wage NTS! 9s 
"IBOTD £4995 OL ‘TItY JO Opis WO UMOp MOT “4S “47 ‘0 bh 61 61 ras) 840'92 ‘Wd TT 
"IBOTD 0 = GS ¥Z gg G00'92 ‘Wd 6 
SM ‘N 0} “9S “AID + 1BaTO ‘o°M st 9% = UG gg 096°Sz "Wd / 
"IB21D 0 < 1G 6G AS 006°&z “Wd G 
"IBD 0 rae 08 1g cS ZL8°SS ‘Wd fF 
"IB21D 0 we Gg 98 Zq 0S8°Sz ‘Nd g 
Gs ks 1g 18°Sz “Wid G 
ce 18 0g 9L°SS “Nd T 
g'es r8 0g LSS “NGL 
"IV9T). “lM ‘ 6% 08 0g OLL'SZ “NV 6 
"IBID ‘T°M wt BG. 6G GF FL9°SS ‘N°Vg | 08 » 
‘Pp WIN | ‘Z'a'S res Gg 98 1¢ 0Z9°Sz ‘Wd TT 
"M ‘N 0} ToxIep “F “WIN Pe Be a cM &8 ifs 3g 89°93 AG 
‘p ‘WIN b's et 25 98 0g 089°9z "Wii aSy 
"N 0} JoyIep “UIU “47 AS ME 25 ie 0g 0S9°Sz EG 
"g “48 “JID ‘eS 2s Gg 68 SP 09°&% ‘Wd g 
‘p48 “IID "2'a'S St OF +P OF 999°Sz ‘Wd 
‘p y8 “dIQ 2 a's ree Ze 18 GF 8019S “WV TT 
A rte 6G 0g 0g PLaZ "N'Y 6 
‘p48 ‘IID Lets be ie 8G OF OL Sz “RVR 166 9 
‘l'M ont OL II +P 088°Sz ‘Wd TT 
"g “4S “AID 2 ee a oL eT GF S88 SZ "W'd 6 
"M 0} ‘Z “98 “AID ‘T°M ae FI 91 1g 06°92 ‘Wd ) 


G16 GGG oF 06°96 _ “Wid G 


yrs WIG 


"mpeo ATIVON 
‘£pnop ‘{wyeo AprvoN 


°‘S “MS 0} “YS “ATO 
*IBITO 

“M 0} SULUTLOF “4R.148 “TIO 
*IV9TO 

*IV9[D 

*IV9[O 


"IV9TO 
"IVITO 
*IVOTO 
“AVIIO 
*IVITO 
"IVIIO 
“IVITO 
“VOTO 
“IVI[O 
“IBID 
“IVITO 


*AVITO 
"“IVITO 
‘SUIVJUNOU JO Opis UO ‘BdVJANS IveM “ys “4T f.Av919 
“IVITD 


APPENDIX G.—METEOROLOGICAL OBSERVATIONS. 


‘ox ‘sanorp 


446 


eases 
DNR NM 


0 
0 
0 
0 
0 
0 
0 
0 
0 
0 


*Panuijuoy—SNOILVAUASTO 


16°S3 

16°S3 

T6°S6 

16°SG 

06°S6 

06°SG 

00°96 
0S0°96 
06196 
S9L°9G 
LPL'96 
L61°9% 
061°9G 
PLT'9G 
09°96 
68696 
O0LG'9G 
096'96 
OTE "96 
LLG'9G 
096 '°96 
61696 
096'°96 
GPG 9G 
L0G°96 
016'9% 
LIG°9G 
O&6'96 


“AALTMONUAHT, 


TVOIDO TOPO LA JY 


‘LopoULOIV 


‘NM 'd G 
‘Wd fF 
‘W'd g 
‘Nd ZG 
Ve |B Ws 
Wd TT 
"Wd ) 
‘NW ‘d G 
‘Wd g 
‘WOT 
“WV OL 
‘NV 6 
WVYQ |€ » 
‘Nd TT 
‘Nd 6 
‘d 08'S 
‘Wd 
“WV OT 
‘N'Y 6 
WVQ |Z » 
‘Wd [T 
‘W 'd G 
‘Wd ) 
‘N'd G 
‘NW 'd 
‘Wd T 
Ew TE 
‘N'Y OT 


447 


APPENDIX G.—METEOROLOGICAL OBSERVATIONS. 


“4SBOTOAO f ye 
“pornosqo suvyg 
“VYSTpAByS WIG 


“YSBILOAQ 
“IeoTo § ySsolq 
“ABOTD ¢ YSolT 


PUA st 


“sot 
-£pno1y 
“ALCY 


“APHOTO 
“AZVFT 
‘eurysuns wap { Azeqy 


“£pnoyy 


- *4SROIOACO 

"MVD £4SBdI0AG 
“eIQIStA A[pavy s.ivzg 
*IB21O 


‘mee ApIvON 


eee 


-“2n 


66°36 
6 °G6 
G6'G6 
68°SG 
68°SS 
68°S3 
GL'SG 
GL°GG 
TL°SG 
GL'SG 
VL’SS 
BL°SS 
08°S6 
68°S6 
€8°SS 
08°S3 
06°96 
06°SG 
16°SG 
68°S6 
L8°SG 
L8°SG 
L8°SG 
L8°SG 
88°S6 
88°96 
06°96 
66°96 
68°96 
68°96 
06°SZ 
06°SZ 
66°S6 
06°S6 
16°SG 
T6°GSG 


‘WV 9 
‘MV ¢ 
W'VZ 
‘MV 1 
“Wd GT 
‘Wd [[ 
"Wd 
‘Wd G 
‘Nd f 
Wd ZG 
‘Wd [ 
‘KGL 
ee cae 
OW OL 
WV 6G 
‘I'v Q 
Mt 'd LT 
“H 'd OT 
‘Wd 6 
"W‘d Q 
‘Wd 9g 
‘W ‘ad G 
‘Wd 
‘Wd @ 
‘Wd G 
"Wd T 
“H GT 
‘WV TT 
“HV OT 
‘WV 6 
WV Q 
“HW 'd GT 
‘Wd [T 
"i ‘d 
‘Wd Q 
‘NW ‘d J 


9? 


9? 


9? 


. 


Iv SP 99 60°96 KCL 


“Arey AYSYS Sourysung ol hie 68 If eg 80°93 ‘WV TT 
98 8g gg F0'9% ‘N'Y OT 
“£pnoyo ‘L's og G's Gg SF $0'9% Ve Gs 
"qSVILIA() I ‘a's sa $8 8 Lt 90°93 |'Wagte |’ - 
wa 8 LP 90°9% ‘Wd g 
A "YItp $4svo19AQ 0 4s G's 29 oF 80'9% "Wd ) 
gg Ge 9g 80°9% ‘Wd Q 
<4 98 1g 6F 90°93 ‘Wd f 
E "4SBOIOAG) ‘I ‘N a Ks IP 1¢ 10°92 ‘W'd g 
is “MOUS FUSS ‘ourysuns wig ‘L ‘'N he oe 68 Hy 10°93 "Wd Z 
2 0 ca 9g 68 6h 90°96 ‘Nd 
O ‘qSBOLOAQ Tt a's es gg G'gg IF 10°9% MGT. 
a gg g'9g SP OL'9% ‘HY TT 
=< 1g gg 9F 80°93 =| "NV 0201 
= “*4SVOIDAO £ ay}ues £19 A “Mm ‘Sg ci ene (is 0g OL'9Z “WV 6 8 ” 
4 LG eh 10°93 =|"N "4086 
qa 8z Sh 90'9% ‘Wd g 
a “Ivapo “qs ‘a ‘N Ge 8% 6% 6F 90°93 = | "Wd 0Z'L 
3 *"N ur Lpnop 0 ‘Y 6% Ig Gg 90°93 ‘Nd G 
a OGe \ uee gg 80'°9% [Wd > 
es ‘pomnosqo ung | ‘LM ‘N - 08 9g 0g E19 ‘Wd g 
= 0g 98 0g 20'9% ‘Wd Z 
| : "IB9T) ‘LM a 0g Gg sg 80°92 ‘Wd T 
6 1 Oe +8 0g Z0'9G ‘N ZT 
bd ras gg 0g 0'9G)=s |‘ H'VGI'IT 
a "oUIYSUNS FYSIAIg 0 eee 6% Ze 1g Z0'9S ‘WV OT 
= ‘qystaq ‘aveID | “T‘M ‘N “ 08 $8 0g 809% wivg | L 4a 
je pe a es ee eee 
jam 
q “4 S10] "90M ‘ool «=| *poqonyyy “moy] “hed 
‘op ‘gan01g “putAA : ae . “IOOULOIVE 
? “UTLATWOWUTH,Y, “OS8T 


448 


"PANUNU0) —SNOILVAUASAG 'IVOINOTONORLATY 


APPENDIX G.—METEOROLOGICAL OBSERVATIONS. 


“VID 


"IBID 

‘"HOZILOY “WY ‘N UL Spnojid £ ava[O 
‘IVO[D WOZLLOFT 

*L9A0 [[G Spnoypo Ad00B f arvalg 


“AZCTT 
*‘pBoyIIAO Spnopo Adse,q 


‘Ivapo AT[VIWAV 
‘aozLIoy ut Apnoyo {pvayaeao r1vaj9 


"SBOIOAQC) 


‘£pnoyy 

‘BUIMOUS paseey 

“A YSIS Surmoug 

“48¥j SUIMOUS  pavy SuLMo[g 

"MOUS "4/T 

‘Wd [ ysed-j[ey Surmous poousmu0d { ysouz 03M? 


“Ysoly 

‘pnopo A]Zursves9uy 
"Uez0djJ punois { ysorg 
"M pure “Ni WO AveI{Q 
“O[QISIAUL S188 /4SvOIOAGQ 
12) (6) 


"1B31N 


“ALB 


EE 
Aw 2 
Ee 


et ite 
EHS 
ne 


Bile fe 
Zo oF £- 
= a! 


inte 
i oe 
Ewe 


oO ;OREEEZ 
nan AZAAAZz 


ri 
3 
w 


o 


Z 
AS 


86 
GE 
GS 
6& 
&& 
6& 
cs 
vs 
VE 
0& 
62 
0& 
O& 
08 
0$ 
08 
1g 
1g 
6S 
68 
98 
96 
8E 
8& 
88 


te) 
Go 


Vé 
96 
Gs 
cs 
cs 
96 
68 
Iv 
GPP 
&P 


89% 
91°96 
E196 
G1L'9G 
IL°96 
GL'9G 
SL°9Z 
61°96 
0196 
80°93 
G0°9G 
10°96 
65°SG 
00°92 
66°SS 
86°96 
L6°SG 
96°S3 
G6°GG 
88°SG 
88°9G 
L8°GG 
L8°GG 
06°SZ 
L8°GG 
98°SZ 
L8°GZ 
88°SG 
66'GZ 
€6'9G 
96°96 
96°93 
96°GS 
L6°S6 
L6°9G 
00°96 


t 


‘Wid J 
‘Md Q 
‘Wag 
‘Wd 

Hd QS 
"Wd 

‘KGL 
NEE 
‘HV OL 
‘WV @ 
‘WM ‘d ZI 
‘Wd [I 
‘Wd OT 
‘Wd 6 
‘Wd g 
“We'd 

“Wd G9 

‘Wd OL'G 

‘Wd OL'P 
‘Wd 9 
‘Wd Z 
Wd T 

WL 
a 
“WV OT 
‘WV 6 
‘Wd ZI 
‘Wd TT 
‘Wd & 

‘Nd 01'S 
"Wd ) 

Wd QT'9 
‘Wd G 

‘Wd 0Z'P 

Wd OLE 
‘Wid 


TE 


OT 


99 


39 


29 


“spnop> Lovep “77 
"IBID 
*“IVOT[O 
IBID 
“IBID 


*IV31D 


“ICID 


"IBID 
“IVI[O 
*IB91D 
“yySsTIVys {we 


APPENDIX G.—METEOROLOGICAL OBSERVATIONS. 


‘op ‘sanomp 


450 


“panunuoy 


0 eee 
a “M ‘N eee 


“JOM 


‘qQ310H 

“puta 0} VUE 
-xoiddy 
“MIT LIMOWUAH,Y, 


—SNOILVAUASAO TVOIDOIOMOULATY 


91°93 
L193 
61°96 
61°96 
61°9Z 
16°96 
1G'9G 
06°96 
06°96 
66°93 
66 9G 
66'9G 
6696 
6G 9G 
6696 
GG 9G 
16°96 
F296 
&2'96 
G6°96 
0G 96 
¥G 96 
96°96 
04°96 
06°96 
16°96 
02°96 
61°93 


*lojomoIeg 


‘Nd f 
‘W'd OTS 
"We'd 08'S 

‘Wd [ 
‘Wd OS'ZI 

WV T] 

"WV OI 

‘WV 6 

‘WV Q 

"M'd ZT 

‘Wd TT 
"Wd 086 

‘Wd Q 

‘Wd Q 
"Wd O8'F 
‘W'd GTS 
‘Wd GIG 

‘Wd [ 

"WI 

WV TT 

WV OT 

‘WV 6G 

‘WV Q 

‘Wd G1 

‘Wd TT 

“Wd OT 

‘Wd 6 
‘W'd GT'S 


*IN0FT 


“OS8T 


451 


METEOROLOGICAL OBSERVATIONS. 


APPENDIX G. 


PZ 8h G6'9% "K'd OT 
reel) 0 a Fs 1g 6F G6'GS | "Wd 08'6 
oo a a = oe =A TarGe ue 
LG 8% 66'S Wd 9 
“18919 0 aa orgy Ze rg 00°92 ‘Wd G 
"18919 0 oF gg | ge 2g 66°96 ‘Wid f 
qeeI) 0 = zs 98 OF 66'SG Wd G 
08 ge GF 0°9S |W "4 OF'ZI 
08 ai 90°9Z ‘WV TL 
1 | Gog 90°92 WV OL 
Ivs]O (4) coo eoe G'ez aa 90°92 "WV 6 
a a ee arate | a1. as 
GG SF 80'9% ‘Wd ZT | | 
VG ¥9 60°92 Wd OT 
ite OF 80:9Z ‘W'd 6 
Ze 0g 60°92 ‘Wd 8 
8% 1g 80°92 Wd ) 
8% 1g 80°92 ‘Wd g 
oe 1g 60°92 ‘Wd G 
‘AOI a if: 8q 01°93 ‘Nd f 
AB9T) pu A Ze re 09 O1'9% |" 0z's 
are “YT 2 ie Ze ie: 9g 0198 |" 08'S 
MBSE) Tg | oe 08 Ze 8h 60°92 Wed T 
G'Te 68 80°93 |'W'd OS'ZI 
02 62 88 019% ‘WV TT 
Lz 6P Z1'9G "WV OT 
je)! 05 tage ol | = GB 88 0193 ‘W'V 6 
at iy ee sae urtttgd aa iy 
Gq LT'9G Wd ZT 
PZ G% 8P 919% Wd TT 
18910) 0 a es 9% Lg 819% WN 'd OT 
"aval 0 a $s 9% 6F 9193 ‘Wd 6 
A GG 9% #9 L193 Wd g 
‘IaI9 0 = 9% Lz 0g 91°93 ‘Wd | 
‘wozl1oy UO Spno[D |G T'M'S’M) — ** Lo 62 #9 81°93 |'W'agI'9 


bE Pe. ST°96 “Wd g 


APPENDIX G.—METEOROLOGICAL OBSERVATIONS. 


452 


‘NV 6 
‘Wd OT 
2 ‘W'd & 
“IB9TD 0 eee “Ww ‘d 8 
Wd L 
‘NWN ‘d 9 
‘Wd F 
‘6 “MN ‘Wid § 
“IB9[O ye ak eee Wd ZG 
TCM CN a ‘Wd T 
“ICID 7 'N ese ‘KH ZI 
"18919 0 eee ‘WY IL 
"IBID I T ‘g eee WCV OL 
"W'V 086 
‘Wd TT 
*£pnojo 0 eee We ‘d OL 
“poinosqo Le Cole) Ay ‘N‘“d 0G'6 
*FSBVILIAQ 0) eee ‘HW ‘d 8 
T'S ‘W'd 9 
ae ‘WM 'd G 
‘£pnopp whe oS ‘Wd GZ 
° ‘Wd T 
‘£pno[g G "M ‘Nl eee ‘KH ZI 
: ‘NV TI 
“£pno[p 7 ‘N eee "WW °V OL 
‘W'V 6 
‘£pno[g sy “M "g eee ‘WV 8 
"VITO oT, "M eee "W ‘a Il 


ooo 
re | en | ee 


‘qq5IeH 
‘om ‘SaQOTO *‘pULAA o4BUL 
-Txoiddy 


*IoyOULOIV 


“AALATWON ATA, 


"Panu2jquojo—SNOILVAUASAQ TVOINOTIOVOA LET 


453 


APPENDIX G.—METEOROLOGICAL OBSERVATIONS. 


"NW ut Apnopd pu ‘Azey 
"IC21O 


"g WOE ITY 


"IR9TD 
‘UIVSR OFIGM “BYU {4 SIU SULINp MOUS Jo [[eJ I4SITY 
"UleY 

‘BUIUIVI POOUSTIULOD YEG 

‘UIVI SuUITyUIAIdG 

‘MN pue "M Ur F4S1Ig 

‘TIBY 

“UV Y 


"M ‘N Wd YSoTJ SUIMOTG poousmM0S Qe’, ye £Apno[O 


‘£pnoyy 
‘ATVYSI[S UlVs 07 sulIsog 


"U10}S G OY] SYOOT ‘Apnopo f4ysIU Surmnp pura ‘yy 
“£pno[p 


"YjIU9ez 07 "MM wWosz Apnojy 


oF a1) 18) 
"IB9TO 
"ICID 


Ha seeeeeoro 


63 i 
a 


BABA Z 


Mona 


69°S3 
09°96 
09°S6 
09°SZ 
19°G¢ 
69°G6 
L9°GG 
69°S6 
01°96 
89°SS 
¥9'SG 
19°GS 
69°96 
99°96 
bS°GG 
69 'GG 
GG GG 
&9°GG 
1G°GG 
69°GG 
69'GG 
GGG 
69°GG 
69°SG 
69°96 
L9°GG 
89°G6 
PL’SG 
PLS 
GL°SS 
LL’ 
61°96 
08°96 
L8°GG 
06°S6 
66°96 


MW 'd Q 
‘Wa G 
Wd fT 
‘Wd g 
‘Wd Z 
Gia wef 
BEG) 
“HV GG TL 
AOE 
‘WV 6 
‘Wv ) 
‘W ‘d OT 
‘Wd & 
2 ane a 
‘yosung 
‘Md G 
‘Wd > 
‘Wd 
‘Wd Z 
‘Wd T 
“W CT 
‘lM °V TT 
HEV OG OL 
‘I'v 6 
‘HV g 
‘Wd TT 
‘Wd OE OT 
‘Wd ) 
‘Wd 069 
‘W ‘dG 
‘Nd f 
‘W'd 0G" 
"Wd 08'S 
‘WGL 
HEV [el 
‘WH 'V QL 


06 


61 


99 


9? 


454 


APPENDIX G.—METEOROLOGICAL OBSERVATIONS. 


*SpNO]D SUIIE}}VOS YIM “av9TO 
“£¥s ontq yTM ‘pnojy 


‘QULYSUNS JO SUIVOIS [BUOISVIDO {4SvOIOAG 
‘<pnojy—s][enbg 
‘pavy Surmoug 


‘JY OY} SuLInp Mous Jo [ey “4, { Apnojg 


‘£¥S ontq jo soyoyed—Apnojy 
“spno[p 
*£YS ONG JO survdpy 
‘ySUj SurMoug 
"M ‘N 0} pozTys pura “Og'e 1V 
‘Apnolyp 
“MOUS 
pus urler !*g wor sy[enbs ox0A0s ‘Qe'T ye {spno[p 
“SUIUIVYT 
‘£pnoly 
‘Ayenbs ‘pnojg 
"4SVOIBAO 
‘W[eo [ogg ‘4001S pus ‘mous ‘uIvI qIIM ‘s}sny 
‘£pnojy 


“£pno[p 
“£pnolg 


*o” ‘sanoig 


‘Panunuoyj—SNOILVAUGSAQ TVOINOTOUOALAY 


riod 


Nex) 
ea 
Bed 
RM NNR MMR 


A ao 
fee 
+ Bae ; 


19 1919 


MN W 


i) 


‘pa 


<2 
TW 


K 


pQ 


oo 


‘Pum 


“qq S10 
oyBUr 
-Txoiddy 


“POM 


“UTLTMOWUTH T, 


“1ojomoIeg, 


‘W'd g 
‘N'd Z 
Nd T 

‘K GT 
‘NV TT 
"HV OT 
‘N'V 6G 

‘N'Y 088 
‘Wd TT 
‘N'd 6 
‘NM 'd Q 
‘W'd G 
‘Nd fF: 
‘Md @ 
‘W'd ZG 


‘Wd [ 

‘Wd 0Z'ST 
‘WV TT 
WV 6 


‘WV g 
"WV 08") 

‘Wd TT 
‘Wd 08 'OT 

‘Wd 6 
‘W'd 08'S 
‘Wd ) 


‘O98T 


GG 


? 


455 


APPENDIX G.—METEOROLOGICAL OBSERVATIONS. 


‘UOOU Td] “QOS 
‘N00 ABOTD “QYUstag 
‘WYSIIq “IweTD 


‘gy S11q ‘“TeeTD 


"UOZILOY OY} WO Spnopo *47 


"IBID 
‘£pno[p 


“£pno[g 
‘£pnolp 
“SUIMOUS posoUsTIMOK 


“AUTO puw Mea Suns ou {spno[g 


‘A]ju03 Sutmous { eurysuns wig 

‘OUIYSUNS WIp—Su1es[p 

‘SULMOUG 

‘J4S1U SJuLMp Mous seyour ¢ noqy 
‘BUIMOUS PSOUaMIULOD “Wd [][ 7B {4svdr10AGQ 


‘O[QISIA S1¥4S BIOS pue HOOT 
‘£pnojg 

“£pno[p 

‘£pnoj) 


eeoeee 


eee 


GCE 


1L°GG 
GL GG 
EL°SS 
VL'GG 
6L1°SS 
08°SS 
18°S3 
98°96 
06°S6 
GS6°SG 
16°SG 
G6'SG 
68°S6 
88°S3 
16°SG 
16°96 
68°SZ 
68°S6 
98°S6 
G8°G¢ 
98°96 
98°S6 
08°93 
68°S6 
98°S3 
G8°GG 
G8°S3 
G8°GS 
9L°GG 
69°GG 
8S°GG 
99°96 
G9'GG 
69'GS 
09°96 
6h SG 


‘Wd Q 
‘Md ) 
‘Wd 9g 
‘Wd G 
‘Nd F 
‘Wd ¢ 
‘W'dZ 
‘Wd [ 
“HK CT 
"HV IL 
AY. OF 
‘H'V 6 
‘N'V 9 
"Wd TI 
“Wd QT 
‘W'd @ 
‘W'd Q 
‘Wd ) 
‘Wd 
‘Wd Gg 
‘Nd fF 
‘W'd ¢@ 


Wa Z 


“w 


‘Wd T 
"NZI 
‘N'Y IL 
‘NV OL 
‘WV G 
‘WV ) 
‘Wd OL 
‘4d 086 
‘IM ‘d Q 
‘Wd p 
‘Wd 9 
‘NW ‘d @ 
‘Nd f 


VG 


86 


9? 


9 


METEOROLOGICAL OBSERVATIONS. 


APPENDIX G. 


6 


‘qy Su OY} Sutmmp [oJ soyout z { ATIAvoy Sutmoug 
"M “SG pues ‘g SUIeyUNOU UO ‘UND ‘4yeA—4g 
‘£Avoy §G "MM “G Woz ‘yYTMOZ UO ‘7 “4S 
“ITD f°M “S Woay ‘yZIUeZ UO ‘*g ‘zg “AID fg ‘Z “48 “GID 
"YPIUOZ WO "WIM “10 “YT 
*g 0} AoT[VA UL W10}S LAVOTT 
*‘SUIZJUNOM IOAO sossvd 
41 8B MOUS pUB Spno][o Burursof ‘Ay “Gg quorand raddq 
‘p ‘wind “yrayg 
‘F “wd “4ywayg 
"M °S Woay ‘fF ‘tIND “yRAYVg 


*M Sq “Gg woz ‘pF ‘tUND “yrIYVC 

*SUIVJUNOU WO MOUS {°AA “G WOAS F “WIND “ITD 
‘UND “AID $"H *M 'G JUOIIND raddg 

‘S]SNS UI YSory SuIMolg 

“Apnopo {szsny 


‘£pnojg 


‘£pnopo {sjsnyg 
‘£pnolg 
“£pnolyg 


‘oy ‘sanoig 


“panurquoy 


e5 
A 


‘pura 


DANN RNRNNN NMnM DW 


. 


‘qq S101] 
oy BU 
Txoiddy 


“OAL WaT AY, 


‘poqoryyy 


—SNOILVAUASAIO TVOIDOTOUOELE 


9z'97 
eZE"GZ 


098 °SG 
968 °S6 
€0P SS 


V6E°GS 
60 96 
OF'S6 
966 °S6 
OF'SG 
SIP SS 
&GE°GG 
98°SS 
8E°SG 
LEGG 
OF'SG 
GPGG 
GPSS 
PP SS 
9F'SS 
LV'SS 
09°G6 
19°96 
€9°GG 
1G°SG 
IL°S3 


*IojouULOIeg 


‘NV ) 
‘Hd TL 
‘W ‘d 6 

Wd 0G") 

"Wd 0G"9 


‘Wt ‘ad G 
‘Wl ‘ad @ 
‘Wd Z 
"KI 
‘WV TT 
‘WV 6 
orieevn): 
‘NM ‘d OL 
‘Wd 6 
‘IW ‘d Q 
‘Wd g 
‘Wd f 
‘Wd @ 
Wd Z 
‘W ‘dT 
‘WT 
sii Cone 
WY OL 
‘WV 6G 
‘WV Q 
‘MW ‘d 6 


*ImoyL 


16» 


9G » 


GZ» 
PG “W987 


“keg 


457 


APPENDIX G.—METEOROLOGICAL OBSERVATIONS. 


‘p yea cS a eee £3 ag EE9°GG WV Te es 
“IvOTD UOZAOY “Gg ATWO $F 4eNG | ‘Pa 's is oie VG eG G9L°GS "Wd & 
SM 'N ns si a ra OLL SS "Wid J 
"N 0} ‘WIM 4eG | “ZAM ON ae 8% GS 7G 061° "Wd G 
“9 | “SM oN ey ¥G 9% a7 06L°S ‘Wd g 
"© is 1) || “SN sh GS ie #9 €68°Sz ‘Nd T 
"e498 “a1 | “TM 7S ps We 6% 9¢ 88'S ‘NZI 
"IVOTO ‘I'g an aes 9% 7g 098°S3 "HV TL 
0 ci 02 a6 Ze 0L8°SZ "W'V 6 
"ICITO eaceas =s ue 0Z 0g 0L8°SS ‘NV g 
*£OT[BA jo opis “M Uo 910 : SUIGJUNOU uo “48S Hala f “I ach ‘Sg Sone ee CT. TV G98 'GG "NV jh if “ITN 
SAG age - 8 al Le Giece ea 
"IB9TO ZO = phy 61 7g 126'&S ‘Wd [1 
"MOZIIOY “AA WO 4S ‘Nd f AvETO aie a ag = 1 eg 0169 "WM ‘dG 
"L 48 ‘aIQ aN = we 9G ZG G98°S "Wd 
‘g ‘UINd “49 ‘IN oie 9% 6% Ze 08'S ‘Wd G 
°e “18 “119 : SUIBJUNOW m0 SULULLOF "49 5G “M “N od WG GS 0g GCOS GG “I “d g 
“SUIVJUNOW UO SUIMMIOZ “4S “IIQ | “FM 'N age en. oe ZG CF CLO8"SZ "Wd T 
“Gute atl °G MEN sty 0g Ze SP O18°Sz "N ZI 
ras che ae jie 183°Sz ‘WV TL 
GAS “O° |) “TAN of 63 ce *g LEE Gz "WV 6 
he 1¢ SPL 'GS ‘NV 8 
‘J4S1u Sulmnp Aous your # ynoge {¢ “4ea}8 “AID oT ss < er GS LP S0L°SS ‘WV. | 83 » 
“MOUS “YT im 1Z 0g IG 08¢°Sz ‘Wd TT 
Z's ge? 1Z 0g 9g GOSS ‘Wd OT 
9 tind 4 ad 4g ag Ue ie aeiee eee 
‘P ‘Jeg a a He 0g 9¢ ShP Ss ‘MW 'd g 
"2 “yeI4g "e°9 oy 6% gg LG LE 9% "Wd g 
"M WOIs ‘g “yerys “WIND ‘9 oe 6Z gg 9g |) —Se-ez ‘Wd G 
Ks Cle SG "Wd g 
‘MOZIIOY *M “G UO “4Ba}S "MIMD f°M “GY MOIS F ‘WIN ‘9'9 pe re 68 6g Gor oz “W'd G 
"Gg ‘WIN SUS Be “ eP gg Zac "Wd T 
‘SOYVB osIv] {4svj Surmoug | °Z"M‘N a oye 1g 6g Che Se WGI 
"g ‘WIN 0 se oP LP 1g PEGS ‘WV TT 
008°Sz ‘WV OL 
‘ *MOug C'S fag Gg v8 gg 808°S3 "W'V 6 


APPENDIX G.—METEOROLOGICAL OBSERVATIONS. 


458 


“IBID ‘TM 9% 8g 098°S3 ‘NW “d 6 
“T 48 ‘S M'N 83 6g O18 "SS ‘Wd ) 
; “UND “48 “YT 3M 125) 9 $1893 ‘Wd G 
“MOT “"M “N WlOuy ‘und “48 “YT ‘S cs 8g 61°S3 ‘Wd § 
&$ 19 08°SS ‘Wd T 
(as 19 68°SG ‘NGL 
te 89 G8'°SS ge Ga 
‘9YC[ JO oVJans uO pues ‘MoT AIOA Surua0y 4g "Fs 1% tG. ‘| So8'sz ‘N'V 6G 
“AVITO "T “MN 1Z OF O8L°SZ | NV CTL Zee 
“IBITD ae | 9% 69 868 '°SS ‘Wd TT 
“IBITD 6S 83 09 818°Ss ‘Wd 6 
“M'N put “M 9} F 99 ‘TM 08 £9 O18 'SZ “Nd J 
‘§ 07% “YS ‘mNd f*NT OF 4S “4T ‘T ‘M‘'N 98 09 OLL'SS ‘W'd G 
"€ ‘M PUB ‘N 4B.9 5 "Gg 0} g -yeays ‘UND ‘6 MS LP 09 091°93 ‘Wd § 
‘G “WT *ITf) ‘TN oP 09 89L'°SS ‘W'd T 
*G “Utd *AIT/) ‘TN 68 09 G9L'SS ey By 
*g “UTM *ATf) 0 98 19 OLL°SS ‘N'Y OL 
"M OF YIVp $G “ULM “IID LS Gg 8g 09L°SZ ‘WV 6 
"N 0} aBapo “gg 07 “WIN 0 0g ¥G LEL‘SS ‘NV 8 
"8 “M PUT N ‘fg 4Ua}s “M19 “yRIzg las 12 oF COLES ‘WV ) | » 
ane tee ES LB 89. | G8L°9% ‘Wd TT 
"MOUS OULT ‘T ‘MS 8S 9g O€L'Ss wd 6 
‘SUIVJUNOU UO MOUS {PF “WIM “AID 0 0g Z9 GZL 93 ‘Wd ) 
‘p “Wid “a1 0 18 Pg 089°9Z ‘W 'd G 
6 2S oe rae 089°Sz ‘N'd g . 
‘SUIVJUNOU WO UMOP MO] Surur10j “yvayg ob ag 92 0g 069'°SZ ‘Wd G 
“Pp “FBAYS “ATH | 18 8G GOL'SZ ‘WV TT 
as 18 og 069°SZ NVG | 3 “AVN 
“4 St0yT JOM A f *‘poqonyy Vv “Moy “kod 
“PUTA oye Ur “TopoMLOIV 
-txoaddy 
‘UALINON WAHT, “OS8T 


‘PANUNUIO—SNOILVAUASA() TVOIDOTOUOULA 


459 


APPENDIX G.—METEOROLOGICAL OBSERVATIONS. 


sey Mous soqoUr zg ynoqe 


“SIU OY} Suramp uaTTey 
>"M 0} JOTAvoy ‘Pp “WINN 
"MN 0} “98 LavozyT 


ani N woay Ajueppns dn Sunads puv ‘pesuvyo pur, 


"MN pus‘ ‘wtu 4s AavoqyT 

"M PUB “MAN 0} Gg “UMD +yrT 

"AN puv -M 0} “4s “tunD 

"G “FS “AIO 

"M "S Wor PULA 5 °K 07 SopTUL g ‘T “48 “AID 
‘FM'S “3 48 “AIO 

‘PD MS Wor g “AID 


"que1INd 


"M ‘§ Woy ‘surejzunoU jo syved uo Suruz0j ‘wind “47 
‘Pp M'S Woy ‘yZIUeZ MO “AID S UOZILOY “Nf UL ‘mINd *yT 


"MN 0} 48 ¢ tveTD 

‘UOZILOY ‘Nl UI “ys ‘wand * IVOIO 

"UOZTIOY “Nf UL “4S { rBeIO 

"%, “9a cat) 

‘surejunoM Jo syvod uo ‘und “48 4] [48 “ATO 
‘T ‘38 “410 

“SUIVJUNOM I9AO 4SOUL “YSIY Suro10J ‘[ 48 “AIG 
"IBID 

"IBID 

“BS “YT 

"N OF “98 "YT 

"uInd “40 

"M PUG "MN OF T “Und 4g 

"IV91Q 

Pp “YeITS 

"qsty ‘[ 49 ‘AID WLOAZ “tN “48 “4-7 

"S “MS Worf Z “48 “AID 

GS “FBALS 


*M 0} SUIeJUNOM UO APYSIIq auyaTys ung 
“UMOD OT ‘g 49 


"M'N UT ‘Q8"6 jNogs Uoes SUA IOOJOM yULTTIIAG B {4BeT) 


- 19 19 


i 


xi 
| 


a 


See 


i = Hoo 
; = 


Noooce 


E 


CO SH Ho 


Moto ooen HOO 


ae 


WABBBEBEZ EE 


° ° 


RMNNRRNNNNNNNM 


coop ion) 


— 


G19°SG 
8S9°GS 
0SG'SZ 
STS-GG 
GES°GS 
0&9°SG 
069°SG 
669 SG 
0L9°SZ 
OOL'SG 


069°SG 
089°SG 
OGL°SS 
09L°SG 
GGL'GS 
OFL'SS 
09L°SS 
661°S3 
O018°SS 
GE8°SG 
098°SG 
698 °G6 
LG8°SS 
606°SG 
006°&6 
006°SZ 
G16 GG 
616°SS 
GT6°SG 
026 °SZ 
668°GG 
0L8°S6 
08°S3 
LG8°SG 


‘astIung 
‘Wd TE 

‘W'd 06'6 
Wd 6G 
Wd J 
‘Wd ¢ 
‘Wd @ 
‘Wd [ 
yh 
‘H‘V 6 


‘WV Q 
‘astaung 
‘Nd OT 
‘Wd 6 
‘Wd ) 
‘yosung 
W'd G 
"Wd [ 
co) oe aaa | 
REV OL 
‘WV 6 
‘W'V Q 
‘asTIung 
“Wd QT 
"Md 
‘Wd ) 
‘Wd g 
‘W'd @ 
‘Wd [ 
“WV. TT 
‘WV G 
M'V g 
‘asTUnNg 


W'd OT 


9? 


2? 


9? 


99 


—_— —— i - a 


G “AID ‘T'S a 13 62 99 090°93 "H'V 8 


GSS 8h | sI096 |N'V 089 
= “B8T) 6 HS — ¢ &% tP €00°93 HVG | OL » 
"T8919 aE N a 8 GG +9 €F0°9Z "W'd OT 
“IB9TD 6 N Bs = LZ 69 080°92 "WN 'd 6 
w "1Ba[D 0 we cj . 08 +9 LG0°9G "Wd ), 
A : “MS 07 T “98 “anh 0 ie GS 98 €9 020°92 "Wd G 
FE GS L8 99 GL0°9G ‘Nd % 
—< && 8s 99 010°9G ‘Wd § 
2 v8 Lg G9 G10'93 |'N“d O8'T 
= 63 ee te) 69 660°93 ‘NV TT 
3 ‘x91 < Me 18 eg 100°93 | ‘N'V OL 
° “L$ kia GZ LG ¥g G00'96 "W'V 6 
a "M Woy YStq ‘T “AID Ta ¥ 1G &% 09 066°96 cS 
a *oGZ'QT SUM JoJoUIOUE 
i -10Y} OY} ‘astIUNS e10Joq Moy uv Jey ‘¢g 4s8 “AID ‘Ta “ x 8T OF 896'S6 “OSstIUNS | 6» 
S ‘g ‘ayo pu 4g co) = tH 9% £9 09696 ‘Wd QT 
a ‘Wd B48 ‘g “GY Mosy A[Bu0I}s MoTq f*N WOIyg “4g Ta er 8% OL 096°93 ‘N'd 6 
3 ‘UOZIIOY "Gg PUB "AA WO “44S {AvE1D ge | ‘we ay &% 69 686 G96 ‘Wid 1 
oS *g 0} “4S ‘Und {UOZILOY “A UO ‘WIND “4S “4/7 0 sor 1% 1g OL 086°SZ ‘Wid g 
a GS $9 606°S3 ‘Wid § 
GS 89 668°96 ‘Wid @ 
= PS 69 898°96 ‘WN GT 
| 18 gg 91893 | ‘NV IL 
ob G& gg OLL'GG ‘HK 'V OT 
‘SurMous { sure, 
4 -UNOW UO ‘UMOP MOT Sspno[d foyVl uo w40j3s fF 49 "7M a2 1% OFL'SS “H'V 6 
‘7 WIN : SS ees 4 Z06°SS N'vg |g “Ie 
Be 
ps jen "40.4, *INO_ “hed 
“OY ‘gano19 “pula 9} eur “IOJOULOTV ET 
-rxoiddy 
‘WALTON WAHT, “OSST 


460 


*PINUNUI)D—SNOILVAUASAQ TVOIDOTOUOTLAT 


461 


py LLL aman nieentanmimmmeamneemrrieernememmemnerareeenmmeeseeeereeree ee a ee ee ee eee 


APPENDIX G.—METEOROLOGICAL OBSERVATIONS. 


"7097S TIA “AA pue -g WMO; sy[enbs yuvysu0g 


"pM Wory g “ys ‘uN 

"M Wort *% 99 

"M WoT Surreapo {7 4s ‘wy 
"M WOT Surtvapo fz 4s ‘uND 


"Z -48 ‘uO 
"S[BALOUI 4B ‘MQ WOT syTenbs fe 4s ‘mND 

*g “YS “WIN 

‘p “YS “WIN 

“MOG 

“MOG 

“MOoug 

"JYSIU [[@ prey Suramoyq {soyepy oSavy ur Surmoug 
‘so0vjd ur avalos Ays—sossvut AAvoy UI ‘und WC 
Ge Ss 

6 4S 

"$"M ‘S Wouz “g “48 “AID 

‘yp “Wit “48 


‘P49 
"p ‘qs1y 49 
*IB9[O 
“1B9[D 


"2 “MN Wouy ‘Z “aI9 
“mOZII0Y “A UO ‘Z ‘WIND “47 
" “M WOIf YSIY ‘T “48 “ID 


1 


na 


ana 


mie 


EEO 


a 
oD) 


Le | 


Ataats 


. 


Ee elf | 


rer 


oo°oo 


. 


OM MN 


wo EuZ 


DR ANNNM 


. 


DRM MM 


| 


O06P GG 

LY'SS 
OSP SS 
GS GS 
GOP GG 
GOV GS 
LOV SG 
O8h SZ 
CGP GG 
OPP SZ 
OZP SG 
G68 GG 
0&6 GZ 
0ZE GS 
0G8 SG 
GLE SS 
OCP GZ 
OSH GS 
OS GZ 
089°SG 
OF9 SG 
OGL°SG 
86L°GG 
&GL°S6 
LEL'GS 
OF6 GZ 
096 °S3 
096 GG 
096 '9G 
086 °SZ 
L96°SG 
01093 
&80°9G 
840 96 
690'9Z 
€L0°96 


‘M if 
‘Wd @ 
‘Wd [| 
Yh 
‘W'V 6 
‘I ‘VQ 
‘ostaung 
‘N'd g 
‘Wd J 
‘Md G 
‘Wd @ 
‘Wd [ 
‘MV IT 
WV G 
‘NV Q 

‘WV 0Z'9 
‘Wd QT 
“i dG 
‘Wd Q 
‘Wd @ 
"Wd [ 
‘WV TT 
‘WV OT 
‘WV Q 

“W'v G19 
‘Wd OT 
‘Wd G 
‘NW ‘dQ 
‘Wi 'd 9 
‘Wd G 
‘Wd 7 
‘Wd Z 
‘Wd T 

WCI 

WV OT TL 

‘WV 6 


él 


GL 


i 


3 


3? 


99 


METEOROLOGICAL OBSERVATIONS. 


APPENDIX G. 


462 


‘MOUS SOUL F 


qnoge YyIA “4QystU Zurinp *M “N Wor oped ‘4a0[s “VT ee o ms 6Z tF 0¢9°9z ‘estiung | g— 55 
19 069°S3 ‘Wd OT 
“M‘N 0} “Wim Sf avalD bi ae) = A es 19 069°S3 ‘Wd 6 
"MN 0} “Ulta ! 1B9]1O 0 “ea We cg 69 029°SZ ‘Wd / 
“MOZILOY “M WO “Wind “YBVIjZs pus ‘Z “49 ‘Tl a2’s Es LE 68 o9 0¢9°SZ ‘Wd G 
‘qySlig ee pes 8g IP 69 889°9Z ‘NM ‘d g 
SP G9 0G1°93 ‘Wd T 
8s IP 19 0Z2°93 ‘WV [I 
"e 49 0 He 8g IP 09 SI LSS ‘HV 6 
"e "49 0 7 las 88 19 G69°93 ‘NV 8 
‘SIn0y FZ JSC] SULINp WoT VF 
sey MOUS SoU J ynoqe fF ‘mNd 4vIys puB “PeIyg ‘B'S os s 83 0g 099°SZ WV9 | GT » 
‘MN pus "My WOdZ puIM Jo s[[unbs [euoIsvoD0 f Moug | “T *M ‘N ey ™ 0g 6g G69'SS ‘N ‘d OT 
“MOUS 0 ee S 1g 09 F69'°SS ‘Wd 6 
"MOUS LAvoyy ‘9 AN 'N coe eee ze #S €19°SZ We ‘d L 
“MOUS “YT 0 es Ge 96 99 699 GS ‘W ‘d G 
‘S[I@F JI SV SurMevyy ‘mous Aavay "TE “Ale aa §§ 9¢ €89°SZ “Wd § 
‘MOUS “YT | “LM ‘N = 88 GP 6G 899°GZ We di 
"p “98 “AID 0 > GP 6P 8g 889°SZ ‘WV TT 
"UMOP MOT “F “9S “AID 0 rs GF L¥ €9 ZOL'SS “Hl 'V G 
“Spnold MO[—g “4S “AID 0 ve & Ss 0g 899°GZ ‘N'Y 8 
“SITLOU ¥G SV] 
oq} U9 |[CF Sty MOUS jo soqoUl G Hoge ‘mous VT 0) Pra 8B 8Z 0g OF9 SS “osTaUNG FL 9 
‘MOUS | ‘FM ‘'N ‘oe ee 63 99 G69°SG "Wd OT 
*“MOUG ‘TT M'N eer TF 6G 8g €69°GZ “Wd 6 
“MOUS "PT | “€°M ‘'N = Pid 18 8g 0°¢S'°GZ ‘Wd ) 
‘py “M‘S§ Woay spnofs Moug | “fF °M ‘'N > ey g8 09 089°93 ‘Wd G | ST ‘ABI 
‘4 SIOH “JOM "OOL] “poyoniyV “keg 
‘oy ‘sanotD “pura 04 Cul *1ejouloIeg 


-rxoiddy 
“‘ATLANWOWATH I, 


“panunuogjg—SNOILVAWASAQ 'IVOINOTOMOMLATY 


463 


APPENDIX G.—METEOROLOGICAL OBSERVATIONS. 


‘HOOU 9Y} pUNOL OTVY pouyop [JOM 
vy 48 

ae 

‘PY FT 

ce Oe | 

"@ “4VI4S “LID 

G 48 “AID 

G YS “AID 

"M 0} ‘WIND “ys LAVOTT 

“M “SO WOIT “MM “N pus “MQ 09 ‘wIND “ys LAvaTT 
S “M “§ Woay G “WIND “49 

"§ 07 ‘g “mnd “4g 


"§ 07 “98 “AID 

"§ 07 “Vays ‘T “AIO 

*M Ul LeIAvey ‘e “4%I98 “AIO puv “yv4179 
"IBID 

"IVITO 

"M ‘N 04 G “98 “AID 

“IBID 


“IBITD 

“IVITO 

“IVOTO 

“IBID 

"AL “N 0} “aT “97 
*IB9TO 

"8 “M Woaz “g “48 “IID 
S48 


‘y AM Wosy 


‘48 ‘umd {'w ‘y OL jogs Surmojq poousmumoy 


*M WAZ ‘UN “3S “4/7 


"M WOIJ Surveys ‘e 4s ‘wNngO | 


ai 
AZZ 


“9° 
S 


COMEE 
= 


° 
. 


Se 
Bw 


SB =) 


EE 


N 
Z 


. 


= 


neeEe 


jrmrooooon 


ne 


Nai 
TEE goO 
a al 


OnN1S 
EE 


eee 


ece 


G'GS 


8EL°SS 
OLL’SS 
00L°SS 
G69 SZ 
OGL'SS 
OGL°SG 
00L°SZ 
OF9°SS 
899°GG 
8ES'GS 
OLS'SS 
OFS'°SS 
699°GS 
T09°SZ 
099°SZ 
099°SZ 
899°SZ 
018 °SZ 
G06 °SG 
O16°GS 
S66 SG 
LG6°GZ 
096°SG 
TL0°93 
&80°9G 
LG0°9G 
866°SG 
080°9Z 
060°96 
066°SG 
0&6°SZ 
618°S6 


818°SS 
OFL°SS 
002°S3 


‘Wd J 
‘Md G 
Wd 
‘Md 
ewer 
‘HV 6 
‘WV 9 

‘OSTAUNG 
‘Wd OT 
‘Wd 6 
‘Wd J 
‘Wd G 
‘Wd T 
SE a: 
‘WV 6 
‘WV Q 
‘aSLIUNG 
SSE 
‘Wd 6 
‘Wd J 
‘Wd G 
‘Wd @ 
‘Wd T 
EN UL 
"WV 6 
‘WV Q 
“OsTIUNG 
‘Wd G 
‘Wd J 
‘W ‘dG 
‘Wd 
"Wd [| 


‘NV TT 
‘WV 6 
‘WV g 


61 


SI 


LT 


37 


39 


39 


METEOROLOGICAL OBSERVATIONS. 


APPENDIX G. 


464 


“M 0} JeyIep “F-49 

"S “M Wor “YSty ‘g “4S “AID 
"g¢48 ‘VT 

"FYI 

i 9g -M9 “VT 

"g (q31q) 4s “410 “77 

*IVITD 

*IVIIO 

‘IBID 


"IVITO 
“IBO[O 
"IBID 
"IBID 
"IVIIO 


‘SIB]s pus moOOuT 
998 0} Ysnoud AvoTD o1oYdsomye oY} ynq ‘MoUs ‘yT 
"MOZILOY ‘A UL “4S YAvp Jopun avojo {Zursved Mmoug 
"MOTIg 
“‘mOoug 
*MOoug 
"WH 'V OL 38 UBSeq fsoyey [[vuIS ‘mous +47 
‘UMOP MOT ‘fF "39 
‘OYV[ JO OOVJANS UO UMOP MOT “F “4VAIG 
"MN pus ‘| 03 Lavog “F “49 
MS Wor *T “AID 
‘SM ‘S Wor “T “AID 


“oy ‘sanorg 


rere 


. 


;OoOoOooO 


coop] 
MUM 


. 
. 


cooocoo ooo 


7 
. 


7 2, 


Z 


cs 


0 
M 
M 
M 
0 
0 
M 
0 
0 
G 
0 


‘W310 
oy RUL 
-rxoiddy 


PM 


“O01, 


“ALTAMONT, 


“‘poyeezy Vy 


‘Panurjvoj—SNOILVAUASAQ) TVIIDOTOUOALTT 


OF0'96 
080°96 
GS0'9G 
090°93 
00193 
OSL'9Z 
G9OL'9G 
OGT'9Z 
OS1L°9% 
60G'96 
00G°96 
O81°9% 
OLL'9S 
GIL'96 
60193 
66696 


OF6°SS 
006°96 
LESS 
098°G93 
168°96 
008°96 
OLL*9G 
PLL'GS 
06L°SS 
89L°S3 
O09L'SS 


“LOjOULOIVE, 


‘Wd 6 

Wa ) 

“WW *d G 

‘Wd ¢ 

“Nd T 

WY TT 

‘NV 6 

‘NV Q 

“OSLIUNG GG ” 
“WN “d OL 

‘W ‘d G 

"Nd / 

“W “d G 

oe 

a vs), 

‘N‘V § IZ » 


"W 'd 

‘Wd J 

‘Nd G 

‘Wd Q 

‘NM ‘dT 

‘WV IL 

‘W'V 6 

‘N'V Q 

‘OstAUNG | OZ » 
“Wd OT 

‘Wd g§ | 6T ICA 


“NOL “AVG 


‘OS8T. 


465 


APPENDIX G.—-METEOROLOGICAL OBSERVATIONS. 


HIBS yi oo M "N Woy syjenbs [VUOTSVO0 OD Co M'N pats eae ST Le 818 °Ss ‘Wid J | 
“UIvE “YT | *§ “MN a tP 8h 9 BZ8'GS "Wd g 
‘"gmnd YT | “8 "MN ae = 0g 8g 008°SZ ‘Wd T 
‘Uret yy | SZ MN a cy LP 19 028 "SZ WV T] 
ea 6 OM a ae ag Z9 018°SS "NV 6 
“UIRY 0 ies ss SF 19 0089 ‘N'V 8 
“UIVY 0 ay i 187 8g OLL'SS MV Q | 96 9 
SP g9 OFL'SS ‘Nd QT 
“arer Apvoys Ta = e SY 39 O8L'Ss W'dG 
‘ayer Apeayg 0 33 me ia 89 8hL°GS Wid J, 
Urey ‘ta oA GP oF 9 9PL GS ‘Wd G 
‘0 «gg 09 Z9 961°G3 ‘Wd [ 
‘p UIT “VT 0 ca Gg 19 69 8189s ‘WGI 
‘p WTe “YT Ta = oF LY 19 LE8°SS "WV 6 
“ULeL “YT oO a GPP q'GP 39 S18'Gs WV g 
“ULBI YT oO oe GOP GIP 9¢ LLL'9G ‘ostamng | GZ» 
"g “WU “47 0 e IP Z9 S8L'GS ‘Wd OL 
‘p “Wa “yT 0 oe ae GP 9 Z6L'GS ‘Wd 6 
TD si = oP g9 08143 |'Wd Og"L 
‘UIBI FF “49 ‘To 0g 99 | 008°Sz "Wd G 
“M 0} UIBI SF “49 0 ae LY ¢ 39 LI8 "9 ‘Nd g 
"$49 oN ‘i 8h gg 89 028°9Z Wd T 
Lb 3g 9 0989 ee IL 
4S | ‘las 3 oP LP Z9 G89 "WV 6 
S 1p 09 B98 GS "HV g 
‘7qSTa oy} Surmp Urey a 2 x 98 GG OF8 "SZ ‘OSTIUNG | FZ» 
‘PMN put “A WorZ spnops +g 4s 41D | ‘I H's xg “ig ag Z9 81693 ‘Nd QT 
p48 0 i gg 98 9 096 GZ Wd 6 
"g "38 “AID 0 wi og 88 29 OF6 "SZ "Wd ) 
"g “98 “JID 0 ie OF LY 99 016°SG "Wd G 
OF § 186°93 ‘Wd g 
8P 99 166°9G |" Wd OPT 
SP 89 62093 wy 
% “98 “IID 0 os Gg Ip 09 180'9G "WV 6 
B ¥8 “AIO 0 oe g8 88 19 B80'9% “HV 8 
eID} La 


'§ eas 86 G'6G 6G 066°S96 ern goo 79 


30 


eS |) ee ee eee = ~~ we A 


¥g G68°SS ‘Wd g 


‘ZN wen OF aca O16 "ez Md T 
49 “UND “y] ‘TN ne LP we G26°GZ MV TI 
“IB9TD 0 ives SP 09 0L6°S% ‘WV 6 
“IBID 0 i GP 69 FS6 SG ‘NV 8 
i “SUIVJUNOM UO “ys “4T f ABOTD 0 owe Ge 09 €86'SS OSTAUNG | GZ > 
S "MOOuL oy} punog oyey ‘ e "49 T “M eee . eee 99 Z66'SS ‘Wd OL 
5 "2-49 TOM nate aw 19 66'S Md 6 
> *$ “18 0 oy a 99 686 SG ‘Wd ) 
ea] “S$ “MN = bP G9 L96 GZ ‘Wd 9 
a G “3S “uns ‘$"M *N i OF $9 0G6°SS ‘Nd f 
S ‘F “MN ‘N Woay ‘T “und “19 GM fond > 0g 060 °SZ ‘WV TT 
"€ “98 “AIO 0 Ke IP $9 GE0'9G ‘WV 6 
| *€ “38 “AIO 0 8 68 09 800°96 ‘N'V 8 
Oo ‘SO (OF “ALD “YT 0 a ors 9¢ G86 SZ ‘OStUUNS | 8G » 
7 ‘IBO[O "% of «ee eee rere) Z66'SS W *d OL 
S "ABITD lane. a ei 99 186°93 “We'd 6 
Oo “AVIT)D be ie ne OF 99 GF6 GG ‘Wid J 
& 4 ; ‘% “YS “mnO ‘SM 'N oe SP €9 666 SG ‘W 'd G 
3 a 4q ‘N woay ‘sureyunom uo Sururs0j “ys “unD "2 'M 'N sisi GF e9 068°Sz ‘NW ‘d f 
= ‘1°38 "ump | ‘e'M‘N Sar i 9 816°S3 "Wd 
si G6 “FS “uny) ‘S °“M 'N ye Sik 09 806°S3 ‘WY TT 
| ‘g “M ‘N "nN wO.1y 4 "48 “ung 7, ™M ‘N eee CP e9 106°SZ "TA °V 6 
. *T “tind 49 Go MN me GV 69 £68°SG ‘W'V 8 
tb ‘g ‘und "YT ay Mi see eee #3 eC69'9 ‘OstaUuNng 1G 99 
s ‘SINOY FZ JSV OY} Sutmnp uay[ey : 
5 Svy UlvI JO POUL ouo ynoqe ‘uoTZE[NO[Vo YSnor Lg | “[ “MN = wie 19 OF8'SS ‘Wd QT 
a ‘e "M 4q ‘N Woay g ‘wmo 4s f ured “y7 SM 'N a wae 09 E8'SS ‘Wd 6 | 93 “AVIA 
= Ae | he Rae 
™~ qusiog | FAM ‘ears | “poypeyyy ‘mnoH] ‘kv 
"oy ‘sano1p “pul oy BUL | roqomorngy 
-Txoiddy 


“WILTON ATH I, “OS8T 


466 


: ‘PANU2}U0JD—SNOILVAUTSAQ “IVOINOTOUOMLTTY 


Sw 


METEOROLOGICAL OBSERVATIONS. 467 


APPENDIX G. 


*§ “3N). 
*§ “38 “ATO 
76 UN 
°G “WIN 


§ “WIN 


“ATjenbs 
*M Wor Surstt tenbg 
"M Wor Sulsv9e[9>—Zuisveo w101g9 
“MOug 
"N Aq *M prey MeaTq f° “N UOgy 
‘WV £ 98 Uvsaq “oojs pu ‘{iey ‘urea Jo wuI0Ig 
"M 0} Avsy eg ‘und “47 
"IB9T)) 


‘e “4s ‘ung 


‘EG “wind “4g 
Z 49 


‘M ULF 98 ‘ung 


"M WOIy F 4s ‘waNnD 
"$"M ‘N Wouy “g--48 “tanD 
"M WOdf Sativayo 
—ueTey sey your uve jo F qnoqe { Zursvoo uy 
"J29F 0O8—AMO] Surum10j spnop>—uivy 
"WV 0¢'9 #8 UBZaq "MM WoT UIvA AAvoTy 
G4 
*Y}aeZ puw UOZIAI0OY “Gg UO ABATO . ag 
: "M Ur Afsomr ‘% “4g 


ee 
—— en) 
wn 


sH 
Zi 


OmnN 


oi 
E 


= vo eEEE 


. 


om} 
Of 6° 
Ew 


NAAN oH 


7 IZ, I, 


EEERO 


an 
Ze 2 


OSL°SZ 
O08L°SG 
LOL“GS 
GSL°GG 
COL'GG 
OFL'SS 
GOL GS 
069°SG 
G99 °GG 
G99°GG 
GE9°GS 


669 '°GG 
069°GG 
69°96 
865°S6 
669 GG 
869°GG 
&19°S6 
869 GZ 
GE9°9S 
669° GG 
819°GS 
609 SG 
009°S6 
G09°GS 
GT9°G6 


899 °SG 


OTL'SS 
O€L°SZ 
0GL°GS 
00L°SS 
GE8 GS 
08°96 
0L8°SZ 


YO 
‘WV 6 
‘WV 9 
“WV G 
‘Wd & 
"Wd ) 
‘Wad G 
‘Wd @ 
Sia a 
‘W°V IT 
‘WV G 


‘WV Q 
Ee 8 
"Wd 6 
“Wd J 
"Wd G 
‘W ‘d @ 
‘Wd 0E'T 
‘WV IL 
“WV 6G 
“WV 
‘OSTAUNG 
"Wd 6 
"Wd ) 
‘Wd G 
‘Wd ¢ 
‘WW ‘d [ 


ESE. 
"HV 6 
WV Q 
“WV QP'G 
‘Wd 
‘Wd J 
‘Wd G 


[ ‘ady 


T& 


0€ 


9? 


29 


APPENDIX G.—METEOROLOGICAL OBSERVATIONS. 


468 


"MN put *M 0} JoxIep ‘¢ 4g Fe ey O18"SZ ‘Wid } 
- ‘y 48 ‘6 °S oie G68°SG ‘Wd § 
"§ 91 “ge 49 ake " G06°SG ‘Nd T 
P “S 98 “ITD 6S = 696°S3 FEL 
"IV9]Q- 6 se F66°SZ NEG 
"N Wody BuLqwopo $] “A109 “47 0 eS 020°9Z “W'V 8 
°G FS “ALO “YT GC H'S oy G66°9g | NVOS'G 1G» 
‘> ON eee $90'°9Z ‘Nd 6 
"M PUG 'N 0} “ATO “YT 0 “ GP0'93 ‘Wd J 
‘a Aq “N Wmoay ‘T ys “uND 5 ig eee 8F0'9Z ‘Wd g 
6 “MN aA 80°93 ‘NM 'd 1 
‘PON Woy “g “48 ‘ung "S$ “MS ‘e 890°9G Wee BL. 
‘mous AAvaT "AN “a L¥0°9Z ‘WV 6 
‘MOUS “PT 0 wee 60°93 ‘W'V 8 
"M pus ‘N 07 ‘ua LAvoqyT 7 Aart enn L96°9g «= NW OS'G FP. » 
"W'd Q 4ev MoUs “47 { ATTenbg "eS “TN ase 016°SZ Md 6 
"MN pue "MM Wor Suistr “F “WIN 0 s G68°SG Nid J 
0 eee §88°S% eae 
"Z °48 ‘many aN “ne G06°SZ ‘Wd T 
GS “YS “uNnYD 6 M'N oe G06 °SZ as a a 
‘G “78 “UND 0 ce G68°SG ‘WV 6 
0 eee 188° ho): 
‘gqSru oy} Suymp [jenbs Mous ‘g ys "uM | “T “M‘N ae Ors'ss |WvosG | & » 
"N 07 § “9S “UaM) "LC MN a GGL’ SS “Wd 6 
“IBID 0 = 8GL°SG “Wed ) 
*¢ “4S ‘und i OGL'SZ | ‘Wd G 
.'W'd 08'S “MN Woy Tenbs Moug woe FIL'SZ ‘W ‘dg 
"g 07 ULL0}g sez ceLgs -| ‘Way | ‘Ady 
quien “Moy Aeg. 
om” ‘sanorp oyB Ur “IojomMOIe 
-rxoiddy 
% *TALTNOWUIH ‘OSST 


*Panu2z}U0)—SNOILVAUESAD TVOINOTOUORLAYY 


469 


METEOROLOGICAL OBSERVATIONS. 


APPENDIX G. 


“18019 +7 ong oF - eieae ‘may 
"IBOTO as) a 8g 6P = FI8'SS ‘Wd ¢ 
"IV9[D ‘bp M'N ae 68 1g eS G98 °SG ‘Wd § 
ea | ‘ZN i i 0g as O1e°ez Wd | 
*spnojo 
OU + SUI[VO THM SuyeUTIE}[e ‘pur Jo stTenbs ‘avetQ | “FM ‘N # GP 8h a LG8°SZ Te LT 
"IV81D | °S “M'N Pes 6g GF as 088° "I 'V 6 
ETO | 1G MIN ae vs 88 8g 698°SS Ny, 
‘Nv FM ‘N Wo pur jo Tpenbs Aavoy { avo[Q =] oN oun see ee 0g L08°GZ |'W'VOg'¢e 
"IB9[D 0 zs res Lg BS 0Z8°3 ‘Hd § 
‘6 GAG ‘N ‘GN ‘g “nd Wg TO he 68 IP sf OLL'SS ‘Wid J 
46 "Ni $ a0 | "N ‘oe 19 7 "M eco eco GP eco FL1°S3 aN ai G 
"p “yeayg ‘eM os Ip oP sit Q1L°SS ‘Wd g 
"M WOIJ SUIIVITO f UOTLLJ SvYy soypour F{[—Suisvoo urey ain pes ase os ge sis O8L°SZ ‘Wd [ 
‘UIvI puv MOTE "FM ON see se 9¢ eee ZLL'GS ‘MY TI 
UICL Lavoyy ‘p “M 'N ese eco 98 eco 29143 “HV G 
“UIVI KAvo yy es “M N coe oce 1e eco GPL'SS WV 8 
"9OBJANS OAOGV 499J OOS ‘WIM fulvs Lavo “Tt ASN ee Oye IP 09 Ove GG "HY 9 
-ULey ae a “4 ea ae F0LeS" Wa OT 
UIVY oO tee eee CP eae L0L°SS ‘Wd 6G 
‘Urer WIM ‘Tpenbs Kavoy | “G “MN os < OF ne 069°S3 ‘Wed J 
‘p “WI “49 0 oy LP eg ns 189° "Wd G 
"M ‘S WOT purM Jo sTjenbs [euosvoog | “TMS os 6P Gg ar 0021'S "Wd g 
‘Ayenbg | “eg "M'S ree 6h rg rs ZOL'SS ‘Wd | 
Secs 8 0 ts 8h rae it BGL'SS ‘WV 6 
pes 0 sf GPP GP sae 0GL°93 mvp 
‘VYSIU UI UIVL fe "MN SGN ‘fF “und “4g 0 ss IP GP a CELE | Mee 
"7 4S 0 S32 sf cP 69 019°SS ‘Wd 6 
‘e489 “T 8 ssf oP LP = G19°S3 ‘Wd p 
‘zumo4g | “zs oM"S ah Bb eq vai 19°62 "dG 
‘bMS ae ‘und "40 ‘bMS wee OF $G oes VEO IGE ‘Wd @ 
‘UIVA FYSYS $F “WIN P'S soe a og a 0¢69°SS ‘WI 
‘p ‘WIN i 8 ne sf 6P 2 099°SS "H'V 6 
PM 'S "MM SON “p TE YT oO st 187 8h a 089'93 ‘N'V 8 
‘reap Atreryjaed Ays {-g oy 4s fp WIN | ‘ga 'S tf 68 P &g 029° |'K Vv 0S"a 
‘§ 0} ‘SZ 1OAO snoutmny Ay¥s fq ‘N WOaz Sutave[yO a 


ut ot oP o9 | v6ugz | ‘wag 


‘mwolg | °Z “M'N e76'0g | Nad 
*T “410 “9'T — R OF6 SS ‘Wd g 
‘Tt VT 3 'N oc6'ez | “N'ag 

‘IWOI9 TO 96693 | “Nd 

xvOT) TD 100°98 | ‘NVI 
wi "18910 0 G66°93 | ‘K'V6 
Zi ‘away | ‘TT's 08693 | “NV L 
: ‘wig | "SA'S os6'9g | NVQ [ZL » 
> ‘moi | 3 “H'S £16'9% Wd OT 
E 90} | “1 “1s 896°9Z ‘W'd 6 
= ‘Surrvoddesip spno[g ‘Sc ‘A 'N 026°SZ ‘Wd 2 
mA *‘£OT[CA jo 9epis “oO 110 SULBJUNOU I9AO SuUIuIIOF mbeguer 14 | 4) 096°SZ “I *d G : 
° "T9810 | “SM ‘N geez | ‘Nag 
3 *g 'N WOT a! “49 “IIO "9 “Mm "N F00°9% "Wd I 
= SM 'N 10°92 WV TT 
ee ‘6 “MN LC0°9% lame 
S "IVITO 0 080°9Z instal” 
= “IB9[O 0 08693. |H'vos's | IT » 
a "N09 98 “97 0 G96'9G "N'd 6 
2 ‘N 0} 48 “97 0 06'S "Wd J 
3 "VOTO 0 006°9z ‘Wad g 
S ‘meet | “ST ‘N 19693 ‘W'd g 
rs "1v91) | “LM ‘N E6'SS ‘Wd | 
| ‘eag | ‘TL ‘a's SF6'SS ‘WV TT 
o "Iva[O 0 ZE6'SS 
# “IVOTO) 0 866 'SS 
S ‘IvOIQ 0 0062 
i, “IBOTO 0 60666 
es) 
Ay 
rm ‘oon | “poqpeyyy 

"on ‘sanomg *purm 0) eUL | ‘0 }OTIOIE ST 


-rxoiddy 


UALTMOWUTH J, 5 


470 


*PINuy}UIYD—SNOILVAUESAO TVOINOTOWOALATL 


471 


APPENDIX G.—METEOROLOGICAL OBSERVATIONS. 


4 


“A[pidea Sutstr 1oyoWM0IVg—*M ‘N 0% 
SUIIAIA PUB "M “G MOI, Sutousurumioo ‘{~enbs Aavopy 


"2 "AM °G WOIZ spnozpo { ATTenbg 
‘HS PUB “M “SG Woy ‘OS ZI OF OS LL Mou UleY 
‘SUIvJUNOW UO SuLUIBA {°g 03 AAVOY “WIM “4g 
Ts Ue I 2 
"M Wor, SULIvI{O 
‘'p “9S ‘und 
6 4S 
aya Maps 
“M ‘S PUTA ITA “GT "g 78 UIer pus repunyy ‘ ATTenbg 
‘g 03 AAvoy FP WIU 4g 
a 200) 
ses 349) 
vy 4S 
"M Ԥ 07 Lavoy g 49 


‘9017 49 
‘SOL 49 


*S pus “M “§ 04 G “98 “UIN) 
“"S pus “M “§ OF “48 “UND 
"IRITO) 

"IBID 

"IVIIO 

"IV9IO 

“IBID 

"LT “9S “AID 

‘L ys ‘wing 

“M “N 07 T “98 “tan 

“M 0} 4S ‘UND “YT 
“ICID 

"IVITO 

*IVIIO 

"IV9[D 


6 
E 
z 


ie! 
Ae 
DW 


A 
Ee aS) 
eh 


Mn WM 


° 
. 


Seo 
BAe aN 


<i 
<a 
ZAZA ww 


ONG 


(on) 
Sw 


BS 
vt 


x Man) 


Be 


ai 
Ara aa a 


ean) 
ooFE 


a 
aes) 


6 
| 
Z 


aoe 


OTS"SG 
O9F GG 
8IG°SS 
0S9°GG 
069°S6 
GCI GG 
0&9°S6 
809°S6 
029°SG 


8L19°GG 
099°SZ 
F09'SS 
E79 SS 
0§9°SZ 
089°SZ 
OT LSS 
OGL°SG 
PLL GS 
OFL'SS 
PLL GG 


G6L°GG 


818°S6 
08°96 
L6L'GG 
€61°96 
008°9% 
PI8 Ss 
§18°96 
898°96 
088°96 
606 °S6 
006°9¢ 
818°96 
086°96 


“Wd 08°9 


‘Nd G 
‘Nd @ 
‘Hd [I 
‘HV TI 
‘HV G 
"WV J 

‘estIong 
‘Nd 6 


‘Wade 
‘Wd @ 
‘Wd T 
‘WV 6 
HV pL 
"NV 08 

‘Wd QT 
“Wd 'g 
‘Wd ) 
‘Wd G 
‘Wd 
‘Wd T 
‘WV § 
"BE Wop 
VOSS 
‘Wd G 
‘Woda ) 
‘Wd G 
‘Nd @ 
‘Wd T 

REV 
WV 6G 
‘WV ) 
“KV O'S 
‘Wd 6 


91 


GT 


iat 


&I 


9? 


bP) 


9 


bP] 


APPENDIX G.—METEOROLOGICAL OBSERVATIONS. 


‘ayn’ 9Y} 
fo faains ay) uo fjyund ay) fo aouasqn ayy burunp ‘hrp 
ayD'T 7105 ur uayn} asam mozjof Joy)? suoynasasgo ay, 

UII “YT 

"M ‘'G WOAZ Spnoj>a—Aa19Moyg 

"Q. “UREN 

‘qYSIu OY} Sutunp urer fF 49 

‘Wd OG') VOUIS SUIMOTY ‘pP ‘URIU “47 
‘N pus °M 0} § “UTTO “YS 

°Z “48 ‘wuNnDg 

"MS PUB *S 07 § “TUNH 

“MN 9} T “98 “tang 

IBITO 

*T ‘wind “119 

‘pM Aq ‘g WOU ‘eg ‘mind “aA19 
"IVIIO 

IBI[D 

"IB9I9 

"1V9IO 

"IVITO 

‘muozZItOYy “g TO ‘wind AAvoy 4ydooxe ‘avalD 
‘Wd Q[ JOG pasved oped { Mous AAvoT] 
‘mous LAvayyT 

“mous AARVOPT 

‘spavf 909—S]ITY WO UMOP MOT “WIN 


‘oy ‘sano1p 


an 
wi 


Lic | 


i 
Part 


oo N16 
aa 


. 
— 

— 
- 


. fom) oi 


- 
ao 


Ana 2azaawe 


ae 
mo 


Neg 
Ann 


19 3.15 
= 


see 
ZZ, 


*‘purM 


‘qs 
oy vUr 
-Ixoiddy 


“AI COWOW MATH T, 


“peqoeyyV 


‘PANUNUOJ—SNOTLVAUASAQ TVOINOTOUORLA 


06°SZ 
16°96 
16°96 


619°SG 
OLL'GS 
O08 1°93 
00L°GG 
019°SZ 
OF9'GS 
099°96 
8P9'GG 
0€9'G6 
699 GG 
69°96 
OPL'SS 
GEL'GG 
SLL°SG 
OF8°SG 
GI8°GG 
098°96 
GO8"GG 
069°S6 
LG9'GE 
G69°GG 
G99°GG 


*LOJOULOIVE, 


‘Wd [ 
‘WV TT 


‘yosung 
‘Wd @ 
‘HV 6G 
astiung 
‘Wd 6 
‘Wid ) 
‘Wd @ 
“Bee ET 
‘WV 6 
Te Ve) 
‘asTAUNG 
‘Wd 6 
‘Nd ) 
‘Wd @ 
‘WV TT 
‘WV 6 
‘WV J 
‘OsTIUNgG 
"Wd 6 
‘MW ‘dQ 
Wd 0¢*L 
‘Wid ) 


*“Ino}yT 


‘OS8T 


66 


6L » 


SL » 


i ee) 


gT ‘ady 


“Av 


478 


APPENDIX G.—METEOROLOGICAL OBSERVATIONS. 


“ABST 

“AOU 

en) 

"IVIIO 

“IVOTO 

aay) 

"IV9TD 

"IBID 

‘spnopo Surf{ 7 

‘£pnoj9 

‘Kpnory 

‘T'S £ Pp 4e surseq—Surmoug 
(‘peutea ‘y9079,0 g) ‘spnozo KAvOTT 
‘yno syverq uns—Apnoq]9 
“‘SUIUIVIT 

"Spnoypo JIT M 

“iD 

"IB2TO 


G6°9G 
16°SG 
F6'SG 
G0'9G 
10°92 
GO'9G 
60°96 
L0°96 
16°SG 
F8°S6 
68°SG 
18°96 
91°96 
61°96 
96°96 
8L°SS 
L8°SG 
06°96 
GL'SG 
GL SG 
06°SZ 
96°96 
86°S6 
66°SG 
96°S6 
88°GG 
G8°GG 
18°GG 
9L°SG 
69°96 
69°96 
69°S6 
08°96 
18°SG 
16°96 


‘osTaUNg 
‘Wd G 
Wag 
‘I 'V 6 
‘aSLIUNg 
Wd & 
‘Wd @ 
Ti Wag 
‘ostrung 
‘Wd G 
‘Wd 
"WV G 
‘ostiung 
‘Nd 6 
‘Wd ¢ 
‘WV 6 
“9sTIUNG 
W'd g 
‘HW 'V 6 
"Wi 'd 
‘WV 6G 
‘Wd [ 
‘WV 6 
‘Wd 
“WGI 
‘WV 8 
‘Wd J 
‘Wd f 
‘HCI 
‘H'V 6 
‘Wd 08’ 
‘Wd T 
‘ll ‘VG 
‘WM ‘dG 
SHuid fi 
‘WV IT 


| G 


LG 


96 
GG 


VG 


39 


99 


9? 


99 


99 


9? 


2? 


9? 


99 


99 


99 


APPENDIX G.—METEOROLOGICAL OBSERVATIONS. 


474 


OL GL 
06 Ld 
92 GL 
vL G9 

GL 
CL PL 
vg $9 
49 OL 
8L OL 
09 PL 
PL GL 
v9 91 
I G9 
09 b9 
TL GL 
8g 89 
OP LY 
69 LL 
GG gg 
OL 69 
0g gg 
69 §L 
¥9 SL 
vg €L 
g9 18 
99 GL 
0g G9 

‘901 ‘pore 
‘UILAWOWAAH YT, 


oL'GS 
T9°SS 
99°96 
TL’SS 
69°96 
61°SZ 
PL'SS 
1L°SG 
8°93 
LL’GS 
16°S3 
L6°SG 
66'SG 
16°96 
69'SS 
86°S% 
66°96 
PL'SS 
GL'SS 
OL°SS 
91°SS 
6L°SZ 
91°93 
PLSS 
GL’SG 
¥9°SS 
09°96 


*IoJOULOLE 


‘astIung 
‘W'dG 
‘Wd @ 
‘NW'V 6 

‘astIuNg 
‘Wd G 
WV 6G 

‘astaung 
‘Wd 6 
‘N'V 6 
‘Wd G 
‘Wade 
‘WV 6 

‘asTIUNG 
‘Hag 
‘Wd @ 
‘K'V 6G 

‘asTauNng 
‘WV 6G 

‘esTaung 
"WV 6 

‘osTIuNG 
‘Wd 
‘H'V G 
‘Wd 6 
‘Wd g 
‘H‘'V 6G 

‘osTIUNG 


*moyyT 


*PANUHUIO—SNOILVATASIQ TVOINOTOUONLYE YL 


c eune 
1g 2) 
0& ” 
63 ” 
83 ” 
KG ” 
9G ” 
GG ” 
ine ” 
eg 0 At 

“hed 

“OS8T 


1g ¢9 
09 SL 
9 SL 
19 L9 
¥G 08 
09 GL 
tP 69 
0S Ly 
FL 92 
69 09 
9 L9 
PL TL 
69 69 
¢9 89 
68 G8 
GL BL 
99 9 
TL PL 
L9 69 
PL 09 
L9 PL 
PL SL 
GL SL 
Lg 89 
8g pA 
We TL 
9 OL 
‘0011 “poyoeyyV 
“‘TAMNOWUAH J, 


66°G6 
88°SS 
G6°S6 
F693 
16°93 
G8°GG 
18°GZ 
GL’GG 
PL°GS 
69°SS 
€9°S3 
99°9G 
9L°SG 
GL°SG 
TL°S3 
89°SS 
68°96 
18°S3 
VLG 
16°SG 
9L°SS 
TL°SG 
99°S6 
89°GG 
99°GS 
PL'SS 
18°S3 
L8°S3 


‘IojoULOIE 


‘asTaung 
‘W'd 6 
‘WV 6G 

‘estaung 
‘Wd G 
‘W‘V 6 

‘asTruNng 
‘W'd 6 
‘Wd g 
WV ¢ 

‘astIung 
‘W'd Q 
‘NM 'V 6G 

‘ostIung 
‘Wd G 
‘Wd @ 
‘WV 6 

‘esTAUNG 
‘Wd G 
‘WV 6 

‘asTruNg 
‘Wd G 
‘'W 'd g 
‘WV 6 

‘esting 
‘W'd 6 
‘Wl ‘d 
‘ll 'V 6 


“INO 


$I 


cL 


Tt 


OT 


99 


99 


99 


99 


9 


9? 


99 


9 


475 


APPENDIX G.—METEOROLOGICAL OBSERVATIONS. 


98°S6 


 ¥8°GG 


v8°SG 
81°96 
18°SG 
61°SG 
¥8°S6 
LL’GG 
b8°SG 
98°96 
G8°GG 
G6 GG 
16°96 
F6°S6 
66°96 
6 °SG 
16°S6 
16°96 
G6'SG 
16'S 
G8°SG 
66 SG 
68°SG 
68°S6 
16°SZ 
68°GG 
61°93 
IL°SG 
69°SZ 
69°S3 
GL'SS 
VL’SS 
GL'GSG 
16°SZ 
66°96 
6L°SG 


Nd @ 
‘W ‘V6 
“9sTung 
"W ‘dG 
“Wl ‘Vv 6 
‘estrung 
‘Wd & 
‘Wd 
"Wd “VG 
‘Wd & 
Wi ‘dad @ 
‘WV 6 
‘Wd & 
Wd @ 
WV 6G 
‘astLuNng 
"W'd 6 
‘Wd @ 
nv 1G 
‘astiung 
‘H'd 6 
"lV 6 
‘Wd 6 
‘NW ‘dg 
‘WV 6 
‘estiung 
‘Wd 
Wd ¢ 
‘HV 6 
‘astaung 
‘Wd Gg 
‘HV 6 
‘OSIIUNG 
‘Wd G 
‘Md g 
‘HN ‘V 6 


IL 


OL 


99 


3? 


99 


99 


9 


99 


9? 


19°GG 
69°GG 
¥9'GS 
$L°GG 
68 °GG 
IL°SG 
GL’GG 
6L°SG 
16°GG 
88°SZ 
6L1°SG 
68°SG 
16°96 
16°GG 
66°SZ 
66°SG 
P8'SG 
86°SS 
98°SS 
66°SZ 
16°GG 
V6 °SG 
66°SZ 
66 °SZ 
66°SZ 
G6°SG 
¥6°SG 
10°96 
V6 'SG 
66 G6 
16°GG 
L6°SG 
16°SG 
L8°SG 
16°SG 
96°96 


‘NV G 
‘astrung 
Wig 
Wd @ 
‘WV 6 
‘ostaung 
‘Wd 6 
‘Wide 
‘WV 6 
‘aslIUNg 
Wag 
NV 6 
‘osTaung 
‘Wd G 
‘Wd @ 
‘HW “V6 
‘esTIuNng 
"Wd G 
WW ‘d @ 
‘V6 
‘asTIUNg 
‘Wd & 
‘Wd 
uve 
‘estung 
‘Wd G 
‘Wd @ 
‘WV 6 
‘osTauNg 
‘NM ‘d 6 
‘Wd 
WV 6 
‘asTIUNG 
"W'd G 
‘NW ‘d @ 
‘WV 6 


GG 


1G 


0G 


61 


ST 


LT 


91 


Si 


vL 


99 


9? 


3? 


2? 


99 


29 


3? 


9? 


APPENDIX G.—METEOROLOGICAL OBSERVATIONS. 


476 


‘ALLINONWAH, 


88°93 
86°9Z 
66°SG 
L9°SS 
68°93 
F8°SG 
18'S 
8L°SG 
FL°SS 
F9°SZ 
TL°S3 
69°SZ 
IL°S3 
9°SS 
09°SG 
P8°Ss 
69°93 
6L°S% 
LL°GS 
TL°a3 
EL°SS 
61°93 
66°93 
TL°93 
9°93 
TL°S3 
PL°SS 
GL'GS 


“Lojyomoreg 


“WN *d 6 : : 8 ae) 18°93 ‘NW'd g 
‘Nad g 18 G8 G98°SZ ‘WV 6 
‘N'V 6 aS G9 ° 66°93 “esTiung 
‘ostaung |: 69 LL 8L°SS ‘Hd 6 
‘N'd 6G 9L GL L8°SZ ‘Nd § 
‘Nd @ 69 PL '  88°G3 "W'V 6 
‘N‘V 6 19 GG 98°SG "osttang 
‘ostuNng OL ” 0g OL F8°SG ‘Hd 6 
‘HM 'd FL 6L/ 98°CS “H'd.§ 
‘Hd § gg 69 L8°SZ "N'Y 6 
‘N'V 6 6P #9 66°SG “OstIUNg 
‘astuung =| 6 ” bP L9 98°SS Ose 6 
‘Wd G 8 ” 69 GL BL°SS ‘Wd § 
‘Wd 6 0g OL 18°93 ‘NV 6 
‘NV 6 L ” 6P Gg IL°G3 “OSTIUNG 
‘Wd 6 99 GL FESS ‘W'd 6 
‘Wd § PL 8h FOSS Wd ¢ 
muvee |9 7» 19 I 91°93 "HV G 
‘NV 6G g ” Z9 TZ IL'9S | “OSTIUNg 
‘WM 'd 6 4 FL G9'GS “HW 'd 6 
Na @ 9¢ Pl 79°93 "WV 6 
‘esTIUNg fb . $g OL G9'°G3 "Wd 6 
‘Wd 6 69 BL 69°GS ‘Wd § 
“A 'V 6 g ” PL GL 19°SZ ‘WV 6 
‘Wd G 1g 09 19°93 “OsTIUNng 
‘Wd 18 61 TL°GS@ ‘Wd § 
‘N‘V 6 SL GL G8°G3 W'V 6 
‘ostuung |g Ane G9 $9 ¥8°SS ‘ostiung 
“INO; “£Uqq "001 “‘poyps}}V 


*1ojoULOIBET 


“OSST "UALTWONYLH I, 


“Penu2}U0)O—SNOILVAUASAQ TVOINOTOWOALA TL 


METEOROLOGICAL OBSERVATIONS. ATT 


APPENDIX G. 


G8 
68 
v6 
GL 
F6 


Oot 


68 


v8 
{6 
vL 
98 
¥8 
8L 
96 
L6 
TZ 
86 
08 
88 
68 
6L 
¥8 
68 
GL 
8L 
68 
94 
08 
G6 
68 
GL 
08 
G6 
LL 
IZ 


G6°SG 
68°SG 
66°96 
96°S3 
b8 GG 
L9°GG 
69°SG 
68°93 
68°SG 
86°SG 
68°S3 
L8°SG 
L6°GG 
68°SG 
16°96 
L6°GSG 
16°S6 
18°GG 
€8°GG 
68°96 
F8°9G 
68°96 
8°G6 
IL°S6 
19°SG 
L9°GG 
GL’GG 
GL°SG 
¥8°S6 
68°96 
16°96 
¥8°SG 
68°96 
L8°GG 
G6 GG 
98°96 


‘asTaung 
‘Wd 6 
‘Nd g 
‘NV 6 
‘Wd 6 
‘Nd g 
‘W'V 6 

‘asTAUNG 
‘Wd g 
‘N'V 6 

“OSTIUNG 
‘Nd g 
‘MN 'V 6G 

“OSTIUNG 
‘Wd ¢ 
‘N'V 6 

“asTIuNg 
‘Nd g 
N'd G 
‘N'd @ 
‘WV 6 

‘ostIung 
‘Nd g 
‘N‘V 6 

‘asTrung 
‘Nd @ 
‘N'V 6 

“OSTAUNS 
‘W'd 6G 
‘Nd g 
‘W‘V 6 

‘OSTIUM 
‘Wd 6G 
‘Nd @ 
‘MV G 

“estIuNg 


i 
1G 


0G 


61 


ST 
LI 


9T 


GT 


rT 


$I 


ol 


99 


9 


qsnsny 


9 


2? 


99 


99 
9? 


9) 


99 


9) 


99 


9 


‘Wd 
‘Wd @ 
‘WV 6G 
‘astauNg 
‘Wd 
‘Wd @ 
"Ii “VG 
‘Wd 
‘WV G 
‘esidung 
‘Wd @ 
‘HW 'V 6 
‘ostaung 
‘Wd 6 
‘Wd @ 
"W'V 6 
estiung 
"Wd 6 
‘Wd @ 
‘ll “VG 
‘ld 6 
‘WV 6 
“OSTAUNG 
"Wd 6 
‘Wd @ 
‘HV 6 
“asTUNg 
‘Wd 6 
‘Wd 
‘WV 6 
"Wd G 
"I'd @ 
‘WV 6 
‘astaung 
‘Wd 6 
WV 6G 


I 


06 


6G 


8% 


Ké 


96 


VG 


&% 


GG 
1G 


9? 


9? 


99 


9? 


99 


99 


29 


99 


99 


9) 


METEOROLOGICAL OBSERVATIONS. 


? 


2? 


?? 


APPENDIX G. 


478 


68 8 G6'9Z “WV 6 | 86 68 16°93 ‘WV 6 6 
GL GL 98°SZ ‘W ‘d 6 8 98 66°SG *W 'd 6 

68 68 08°SS ‘Nd g L6 6 69'GS “Wd g 

98 98 98 GZ “HV § PL sas 68 98 18°SZ ‘WV 6 8 
61 PL 66°GSZ ‘WV 6 G8 88 TL°SG ‘Wd 6 

9 G9 8°93 ‘OSLIUNG GL lM 86 G6 we ‘W ‘d § 

98 88 69°93 ‘W ‘d 6 88 L8 66°SG ‘W'V 6 4 
86 96 68°SG ‘Wd § 68 81 81'S ‘W'd 6 

68 88 G9°GS ‘WV 6 86 86 66°96 ‘Wid § 

L9 99 $6°SR ‘OSTAUNG LL ee 68 88 88°SG ‘W'V 6 9 
G8 L8 69°GZ ‘W'd 6 68 82 66°93 ‘WV 6 

L6 96 86'SZ ‘W ‘dg OL OL 6°92 ‘ostuung g 
ct 18 96°96 WH 'V 6 76 6g 66°96 ‘Wd g 

99 $9 66'S ‘OstaUNG OL ” 88 26 08°LZ “WV 6 p 
98 98 68°SG ‘W ‘d 6 8 G8 96°96 ‘Wd § 

66 G6 G9'GS ‘W ‘d 2 6 qsnsny 66 66 26'GZ ‘NV G e 
‘oa ‘poypuyyy ‘aNO}T “hog “oon “porpryyy ‘aMOTT 

$$] ‘sojom01eg ‘LOPOULOIBE, 
"-‘WaUWONaH, “OCR ‘WALTWONWAH TL, *OS8T 


"Panu? UIO—SNOILVAUESAQ TVOIDOTOUOALASY 


ALPHABETICAL INDEX. 


The asterisks indicate the page of the appendices where the locality or object is 
especially referred to. 


A 


Abronia mellifera, 395*. 

- Accident, 266. 

Accipiter fuscus, 314*. 
Acephala, 412*. 

Acerates decumbens, 394*. 
Achillea millefolium, 3891*. 
Actiturus bartramius, 3826*. 
Adobe Hall, 300*. 

Agelaius xanthocephalus, 326%. 
Agonum erythropum, 374*. 
Agropyrum repens, 397*. 
Allium stellatum, 397*. 

‘© reticulatum, 397*. 
Allorisma terminalis, 413*. 
Aloe, 82. 

- Amblirion pudicum, 396%. 
Ambrosia coronopifolia, 389%. 
Ameiva tesselata, 338%. 
American Creek, 294*. 
American Fork, 215. 
Ammophila aberti, 368%. 
Amorpha, 25, 26, 27, 34. 
Amsinckia lycopsoides, 393*. 
Anabrus simplex, 372%. 
Anas boschas, 322*. 


Ancient beach line of Salt Lake, 158, 160. 


Ancient Lake, 230. 
Anemone pennsylvanica, 383*. 
Anser canadensis, 321.* 

«¢ erythropus, 321*. 


Antelope, 28, 50, 82, 105, 205, 237, 238. 


Antelope Island, 157. 

Aph Creek, 294*. 

Aphodius strigatus, 375*. 

Arctomys ludoviciana, 37. 

Argemone hispida, 383%. 

Areoda lanigera, 375*. 

Artemisia, 25, 26, 27, 30, 92. 
a canadensis, 391*. 


Artemisia ludoviciana, 391*. 

frigida, 391*. 

ae tridentata, 391*. 
Arthrocnemum fruticosum, 394*. 
Ash Creek, 40. 

Ash Hollow, 272*, 289*, 301*. 
Aspen Creek, 285*. 

Aster oblongifolius, 389%. 
Astragalus, 102. 

i adsurgens, 385*. 
Athene hypugea, 314*. 
Atmosphere, dryness of, 72. 
Avicula, 58. 

Avicula custa, 412*. 
Avoset, 320%. 
Avoset, white-headed, 326%. 


B 


Badeau’s Post, 288*. 

Badger, 311*. 

Barometer, breaking of, 59. 

Basalt, 107. 

Base line of survey, 121. 

Bauchmin’s Creek, 217—276*, 278%. 

Beaches, ancient, 105. 

Bear, 237. 

Bear River, 77, 78, 94, 276%, 280*, 293. 

Bear River Bay, 165. 

Beaver, 82, 241. 

Belostoma, 370*. 

Berberis aquifolium, 383%. 

Big Blue River, 20, 271*, 291*. 

Big Horn, 312%. 

Big Kanyon Creek, 217, 277* 

Big Sandy River, 27. 

Big Vermilion River, 20. 

Birds collected, 314. 

Birds collected by Lt. Abert, 825%. 

Birds of N. America, not described in 
Audubon, List of, 327. 


479 


480 


Bishop Crosly’s, 294*. 
Bitter Creek, 55, 234, 273*. 
Bittern, 320*. 

Bitumen, 176. 

Black Flies, 205. 


Black’s Fork, 74, 228, 276*, 281,* 300*. 


Black Rock, 118, 169, 297*. 
Blacksmith’s Fork, 94. 
Bluebird, long-winged, 314*. 
Blackbird, yellow-headed, 326*. 
Blue Heron, 160. 

Boat, construction of, 150. 


Boats, their novelty to the Indians, 174. 


Bonneville, Col. 30. 

Botany, 383. 

Botaurus lentiginosus, 320*. 
Boudin, 35. 

Boulders, 101. 

Boulders of granite, 19, 91, 96, 164. 
Bowery, 130. 

Box-elder Creek, 96, 214, 293*. 
Brachiopoda, 409*, 

Bridger’s Fork, 238. 

Bright Creek, 284; 

Brown’s settlement, 83. 

Buffalo, 33, 34, 36, 287, 238, 247. 
Buffalo berry, 233. 

Bull boats, 214. 

Burrowing owl, 37. 

Buteo borealis, 314*. 

Butte Creek, 285*. 


C 


Cache Cave, 225, 280*. 
Cache Valley, 80, 94. 
Cactus, 34. 
Calamites, 403. 
Callipepla gambeli, 326*. 

- squamata, 326*, 

es venusta, 326%. 
Callirrhoe involucrata, 384*. 
Calochortus luteus, 160, 208, 397*. 
Calosoma scrutator, 373.* 
Camass prairie, 219, 225. 
Cantonment Loring, 93, 293%. 
Capote des acarte, 175. 
Carabus finitimus, 373*. 
Carboniferous limestone, 15. 
Carcajou, 311*. 
Carbonate of Soda, 62, 67. 
Cardinia, 58. 
Carrington, Albert, Esq., 122. 
Carrington Island, 297*. 
Castilleja hispida, 392*. 

a miniata, 392*. 

a sessiliflora, 892*. 
Castle Island, 159. 
Catfish, 22, 23. 


ALPHABETICAL INDEX. 


Cave, 202. 
Cedars, 239. 
Cedar City, 142. 
Cedar Grove, 294*, 
Cedar Trees, 210. 
Cerasphorus garganicus, 378*. 
Charadrius vociferus, 319*, 
Chemical analyses, 417*. 
Chenactis, 210. 
Chenactis stevioides, 390*. 
ee achillezfolia, 390*. 
Chenopodina linearis, 394*. 
Cherry Creek, 274*. 
Cheyennes,. 259. 
Cheyenne Pass, 300*. 
Chicken Cock Bluff, 224, 279*, 300*. 
Chimney Rock, 48, 51. 
Chironomus, 379*. 
Chleenius sericeus, 374*. 
vd vicinus, 374*, 
Cholera, 14, 15. 
Cholera among the Sioux, 48, 45. 
Cholera in the party, 55. 
Chonetes variolata, 410*. 
Chronometers, rate of, 801*. 
se most suitable kinds, 301. 
Chugwater Creek, 266. 
Chugwater River, 287*, 300*. 
Churchillia bellona, 350*. 
Cicada rej, 369*. 
‘¢ rimosa, 870*. 
‘<  striatipes, 369*. 
Cicer arietinum, 385*. 
Cicindela auduboni, 373*. 
Circulating medium of Utah, 133. 
Cirsium undulatum, 391*. 
Civil government of the Mormons, 182. 
Clangula albeola, 324*, 
Clematis ligusticefolia, 383*. 
Cleome lutea, 208, 384. 
Clymus striatus, 397*. ; 
Clytus irroratus, 378%. 
‘¢ flexuosus, 378%. 
Cnemidophorus tigris 338*. 
Coal, 16, 116, 225, 236, 241, 246. 
Coal Basin, 300*. 
Coal-bed, Bitter Creek, 282*. 
Cock of the Plains, 319*. 
Cold Springs, 218. 
Coleoptera, 873*. 
Collomia linearis, 393*. 
Coluber eximius, 351*. 
‘¢ mormon, 361%. 
Columba leucoptera, 326*. 
Columbia River, 92. 
Colymbus glacialis, 824*. 
Comandra umbellata, 209, 394*. 
Commelina, 26, 27. 
Communism not existing among the 
Mormons, 134. 


ALPHABETICAL INDEX. 


Conflicts with the Utahs, 149. 
Constitution Hill, 297*. 
Convention to form a constitution, 127. 
Cophosaurus, 341*. 
Copperas Spring, 227, 281*, 300%. 
Corals, 407*. 
Cormorants, 161, 207, 324*. 
Corralling, 16. 
Corydalis aurea, 383%. 
Cotalpa granicollis, 374%. 
Cotton-wood, 107. 
Cotton-wood Creek, 294*. 
Cotton-wood Trees, 244, 245. 
‘Court House,” 48. 
Cowania, 210. 
Cowania stansburiana, 386%. 
Crane, Brown, 319*. 
Crepis acuminata, 392*. 
Crotaphytus, 339*. 
Crotaphytus wislizenii, 340*. 
“ collaris, 340. 
Crow Creek, 259, 286, 301*. 
Curlew, long-billed, 320*. 
Currant bushes, 236. 
Cyathophyllidex, 407*. 
Cyathophyllum, 1738, 196, 211. 
Cygnus americanus, 321*. 
Cynthia cardui, 366*. 
Cypricardia occidentalis, 412*. 
Cystiphyllum, 211. 


D 


Dafila acuta, 323%. 

Diczlus splendidus, 878*. 

Dieteria pulverulenta, 389*. 

Difficulties of surveying the Lake, 187. 

Digitalis, 23, 50. 

Disappointment Island, 159. 

Dividing ridge between Forks of the 
Platte, 40. 

Dividing ridge of Atlantic and Pacific, 
241. 


Dividing ridge of Platte, 28. 
Deer Creek, 61, 274%. 
Deilephila lineata, 366%. 
Delphinium azureum, 383*. 
Density of Salt Lake water, 212. 
Deseret, 127. 
Deserters, capture of, 69. 
Devil’s Gate, 65, 274*. 
Dogs, as beasts of burden among the 
Indians, 47. 

Dolphin Island, 191, 297%. 
Dove, white-winged, 326%. 
Drift-wood, 101, 218. 
Dry Cotton-wood Creek, 215. 
Dry Creek, 48, 294*. 
Dry Sandy Creek, 71, 291*. 

31 


481 


Duchesne’s Pass, 219. 
Duck, Black-head, 324*, 
Duck, Bald-pate, 322. 
Duck, Butter-ball, 324*. 
Duck, Sprig-tail, 823%. 


AD) 


Early history of Great Salt Lake, 151. 
Earthquake, 150. 
Echinospermum floribundum, 393* 
Echo Creek, 224. 
Egg Island, 206. 
Kighteen-mile Point, 290*. 
Elaphidion marilandicum, 378%. 
Elder Creek, 274%, 289*. 
Eleodes cognata, 386*. 
cc obscimma.ie/ Ge 
Elgaria scincicauda, 348*. 
Elk, 50. 
Elk mountains, 238. 
Elymus, 209. 
Emigrants, 14, 18, 21,.24, 41, 44, 223, 
230. 
Emigrant’s Creek, 26. 
Emigrants, destruction and abandon: 
ment of their property by, 55, 57, 
60, 68, 67, 112, 118. 
Emigrant graves, 227. 
Emigration Creek, 277*. 
Encrinites, 16, 39, 42. 
Ephedra americana, 395%. 
Ephippigera trivavensis, 371*. 
Equisetum, 111. 
Erigeron coccineum, 389%. 
Erigonum umbellatum, 394*, 
a fremontii, 394%. 
Eriocoma cuspidata, 397%. 
Eritrichium glomeratum, 392*. 
Ermine, 311*. 
Erodium cicutarium, 388%. 
Erysimum asperum, 383%, 
Erythronium grandiflorum, 397%, 
Euomphalus subplanus, 414*. 
Euphoria cernii, 374*. 
Eutoca heterophylla, 393%. 
Evans’s Trace, 284*. 


i) 


Falco sparverius, 325*. 
Faphrentis multilamellata, 408*. 
&é stansburii, 408*. 

Faviphyllum rugosum, 407*. 
Favosites, 211. 

Fenestella, 211. 

Ferry on North Fork of Platte, 60, 
Fiber zibethicus, 311*. 


482 ALPHABETICAL INDEX. 


Finch, gray-crowned, 317*. 

Finch, Lincoln’s, 317*. 

Fish in Warm Springs, 117. 
Flat-head Indians, 214. 

Flat-rock Point, 172. 

Flowers on the prairie, 28. 

Fort Bridger, 74, 228, 276*, 281*, 800%. 
fort Hall, 93. 

Fort John, 53. 

Fort Kearny, 30, 272, 290*, 301*. 
Fort Laramie, 52, 273*, 288*, 301*. 
Fort Leavenworth, 270, 292, 301*. 


| Fossils,’ 32, 34, 35, 56, 58, 60, 61, 68, 


74, 89, 91, 116, 196, 211. 
Fossil bones, 33, 34, 39, 402. 
Fossils described, 407. 

Fossil plants, 62, 72, 73. 

Fossil trees, 72. 

Fourche Boise River, 274*. 

Fourth of July, celebration of, 42. 

Fox, great-tailed, 309*. 

Frappe’s Creek, 250, 285*, 300*. 
Frémont Island, 160, 297. 

Fresh water on Salt Lake, 1738, 178, 184. 
Froximon cuspidatum, 392*. 

Fuligula affinis, 824*. 


G 


Gasteropoda, 413%. 

Gayophytum ramosissimum, 387*. 

Gentiana affinis, 394*. 

Geographical positions, table of, 300*. 

Geology, 38, 69, 71, 75, 78, 79, 83, 89, 
90, 96, 103, 111, 1138, 159, 168, 228, 
230, 232, 234, 244. 

Geology of the route described, 401*. 

Gilia pulchella, 393*. 

Gnats, annoyance from, 174, 176, 190. 

Golden Pass, 218, 278%. 

Goose Creek, 89. 

Goose, Canada, 321*. 

Goose, white-fronted, 321*. 

Government train, 28. 

Grand Island, 29. 

Grayia polygonoides, 394*. 

Grayia spinosa, 209. 

Grease-wood, 172. 

Grease-wood Creek, 274*. 

Great Basin, 227, 264. 

Great Salt Lake, 101. 

Great Salt Lake City, 84, 277*, 278*. 

Green River, 72, 276*, 282%. 

Grindelia squarrosa, 389*. 

Gros Bois Creek, 94. 

Grus canadensis, 319*. 

Gulls, 102, 179, 188, 191,205, 206. 

Gulls’ eggs, 197. 

‘Gulo luscus, 311*. 


Gunnison’s Island, 197, 297*. 
Gypsum, 2438, 246. 
H 

Ham’s Fork, 74, 231, 276%. ' 
Hastings’s Cut-off, 112. 
Hat Island, 204. 
Hawk, sharp shinned, 314*. 
Hawk, red-tailed, 314*. 
Hedspeth’s Cut-off, 293*. 
Hedysarum mackenzii, 385*. 
Hemiptera, 369%. 
Henous techanus, 377*. 
Heterodon nasicus, 852*. 

a platyrhinos, 353*. 

6 simus, 853%, 
Heuchera, 210. 
Heuchera rubescens, 388*. 
High Fork, 94. 
Hill Creek, 281*. 
Holbrookia, 840*. 
Holbrookia maculata 342*, 308. 
Hordeum jubatum, 397%. 
Horia stansburii, 377*. 
Horned Frogs, 178. 
Horned Frogs, monograph of, 354, 
Horse Creek, 273*, 287%. 
Horse-shoe Creek, 56, 273%. 
Horses stolen by Indians, 22. . 
Hot Springs, 213, 215, 293*, 294%. 
Hot Spring water, analysis of, 420*. 
Hydrochus foveatus, 375*. 
Hydrophyllum capitatum, 393*. 
Hymenoptera, 867%. 
Hypoxis erecta, 397*. — 


I 


Independence Creek, 292*. 
Independence Rock, 65. 
Indians, 100, 103, 107, 251. 
Indians, alarm of, 26. 
Indian burial lodge, 40, 42. 
Indian camp, 46. 

Indian Creek, 270*. 

Indian lodges, 111, 181. 
Indian pottery, 182. 

Indian Springs, 185, 192. 
Indian traces, 244, 245, 248. 
Inoceramus, 402. 

Insects described, 366*. 


J 


Jealousy of the Mormons, 85. 
Jordan River, 141, 156. 


ALPHABETICAL INDEX. 483 


Jordan, valley of, 128. 
Juncus balticus, 397*. 


K 


Kansas Branch, 270. 
Kanyon Creek, 294*, 
Ketcham’s Creek, 23, 271*. 
Killdeer, 319%. 

Keeleria cristata, 397. 


L 


Labidus harrisii, 367*. 
‘¢  melshemeri, 368*. 
Labidus saji, 367*. 
La Bonté Creek, 58. 
La Bonté River, 273*. 
La Hontan’s knowledge of Great Salt 
Lake, 151. 
Lake Fork, 286%. 
Language of signs, 254. 
La Préle River, 59, 2738*. 
Laramie Fork, 286*. 
Laramie Peak, 250. 
Laramie Plains, 300*. 
Laramie River, 252. 
Lark, Western, 316%. 
Larve of insects, 164, 177, 379*. 
Latitudes and longitudes in Salt Lake 
Valley, 297%. 
Lava, 163. 
Lawrence’s Fork, 48. 
Layia glandulosa, 391*. 
Legerette Creek, 19. 
Lepidoptera, 366*. 
Leucosticte tephrocotis, 307, 317%. 
Lewis’s Fork, 92. 
Lias, 78. 
Lima, 211. 
Lignite, 49, 234. 
Limestone, 16, 17, 18, 19, 20, 176. 
Limestone, carboniferous, 15. 
Linosyris serrulata, 389%. 
Lithospermum circumscissum, 393*. 
Lithostrontion, 408*. 
Little Blue River, 271*, 291*, 301*. 
Little Sandy Creek, 72. 
Little Snake River, 300*. 
Loasa nitida, 23. 
Locust, 3871. 
Lodge-pole Branch, 287*. 
Lodge-pole Creek, 251, 260, 301*. 
Logan’s Fork, 94. 
Loon, 324*. 
Loyalty of the Mormons, 144. 
Lupinus albicaulis, 386%. 
Lygodesmia juncea, 391* 


M 


Malacothrix, 209. 
Malacothrix sonchoides, 892*.! 
Malade River, 88, 98. 
Mallard, 322%, 
Malvastrum coccineum, 208, 384*. 
Misc Us collected, 309%, 
“ by Lt. Abert, 313*. 

Maple Creek, 293*, 
Marble, 172, 174. 
Mareca americana, 322%, 
Marsh Creek, 294*, 
Mauvaise Terre, 234. 
Medicine-bow Mountain, 300*. 
Medicine-bow River, 285*, 
Medicine Butte, 77. 
Megaderus corallipes, 378*. 
Meles labradoria, 311*. 
Meloe parvus, 377%. 
Melolontha decemlineata, 375*. 
Melops constrictus, 374*. 
Mentzelia ornata, 587*. 

$6 albicaulis, 388%. 
Mertensia drummondii, 398*. | 
Metamorphic rocks, 79. 
Meteorological observations, 423. 
Military post, eligible site for, 95. 
Miller’s settlement, 215. 
Mink, 130, 311*. 
Mirage, 101, 119, 184. 
Mocking birds, 205. 
Monothrix stansburiana, 389*, 
Monotis, 403. 
Morin’s Fork, 279*. 
Mormons, 139. 
Mormon battalion, 125. 
Mormons, obligations to, 217. 
Morin’s Creek, 222. 
Mouna Creek, 294%. 
Mountain sheep, 312*. 
Mud cakes, 164. 
Muddy Creek, 238. 
Muddy Fork of Green River, 77. 
Mud Island, 168, 297%. 
Mule killed by a snake, 70. 
Muskrat, 28, 311*. 


N 


Nautilus, 404. 

Nebraska, 40. 

Necrophorus obscurus, 376%. 
Needle Creek, 226, 

Needles, the, 280*. 
Neemaha, 270*. 

Nemaha, 292*. 

Nemahaw River, 23, 


a 


484 ALPHABETICAL INDEX. 


Niphoea oregona, 316*. 
Nitrate of soda, 62. 

Noche, 175. 

North Fork of Platte, 243. 
North Platte River, 301*. 
Nucula arata, 413*. 
Numenius longirostris, 320*. 
Nyctobatus intermedia, 376*. 


O 


Oak grove, 291*. 
Obione canescens, 395*. 

‘“< confertiflora, 395*. 
(Edipoda corallipes, 371*. 
CEnothera, 23, 25, 27, 32, 48. 
(Enothera albicaulis, 387*. 


ee cxspitosa, 287*, 
- hispida, 68. 
- scapoidea, 387*. 


Ogallalah Sioux, 253. 
Ogden City, 126, 214. 
Ogden’s Creek, 81. 
Ogden’s Hole, 78. 
Ogden’s Pass, 293*, 
Ogden River, 214. 
Ona-pah Creek, 294*. 
Oolitic limestone, 113. 
Oquirrh Mountain, 141. 
Orthis, 196. 

Orthis umbraculum, 412*. 
Orthoptera, 371*. 
Otocoris occidentalis, 318. 
Ovis montana, 312*. 

Owl, burrowing, 314*. 
Ox-bow Creek, 291*. 
Oxen poisoned by water, 63. 
Oxytropis lamberti, 385*. 


Er 


Pacific Creek, 275*. 
Pacific Springs, 70, 275*. 
Pah Utes, 148. 


Palzontology of the expedition, 407. 


Panageus distinctus, 373*. 
< caliginosus, 373*. 
Pannack, 293*. 
Pannack River, 87. 
Papilio asterias, 566%. 
Park Mountains, 239, 342. 
Parley’s Park, 300*. 
Partridge, Blue, 326%. 
5 Scaly, 326*. 
Parus septentrionalis, 316*. 
Pass Creek, 285*. 
| Paysan City, 142. 
' Pelecanus trachyrrhynchus, 324*. 


Pelicans, 117, 179, 183, 188, 191, 193, 
195, 206, 324*., 

Pelidnota tripunctata, 375*. 

Pentstemon grandiforum, 392*. 

Pequinnetty Creek, 294*, 

Peretyle, 210, 

Perthostoma, 370. 

Petenete Creek, 294*, 

Petroleum, 226. 

Peucea lincolnlii, 317*. 

Peucedanum citernatum, 889*. 

Phaca mollissima, var. utahensis, 385*. 

Phalacrocorax dilophus, 324*, 

Phalangopsis, 378*. 

Philonthus comptus, 375*. 

Phlox, 23. 

Phlox hoodii, 394*. 

Phlox longifolia, 3894*, 

Pholadomya, 402. 

Phrynosoma, monograph of, 354. 

“ cornutum, 360*. 


e coronatum, 360*. 

2% douglassii, 8361*, 362*. 
ee modestum, 361*, 362%. 
- orbiculare, 359*. 

= platyrhinos, 361*, 362%. 


Physalis lanceolata, 394*. 
Physaria didymocarpa, 384*, . 
Picus torquatus, 319%. 

‘¢ varius, 326%, 
Pieris protodice, 366*. 
Pilot Butte, 231. 
Pilot Peak, 111. 
Pimquan Creek, 294*, 
Pifion, 111. 
Pipilo aberti, 325%. 
Plan of Salt Lake City, 128. 
Plants, 68. 
Plants described, 383*. 
Platte Ferry, 274*. 
Platte Ford, 284*, 
Platte, Forks of, 33. 
Platte River, 272*, 289*, 
Pleasant Creek, 294*. 
Plestiodon skiltonianum, 349*, 
Pleurotomaria coronula, 413%. 
Plochionus timidus, 373%. 
Plover, 17, 27. 
Plover, Field, 326%. 
Pecilus scitulus, 374%. 
Polygamy among the Mormons, 136. 
Polygonatum canaliculatum, 396*. 
Pomont Creek, 215. 
Port Neuf, 87. 
Port Neuf River, 273%. 
Prairie-dog village, 37. 
Prairie hens, 17. 
Prairie squirrel, 312%. 
Préle, 27. 


G 


SB 


ALPHABETICAL INDEX. 485 


Preservative qualitiés of the water of 

Salt Lake, 171. S 
Proctrotretus, 341*. 
Productus, 16, 19, 401. Sabbath, importance in travyelli 
Productus costatus, 411*. day of ie 99. BA 
Productus semi-reticulatus, 54, 411*. Sage, 160. 


Promontory Point, 169, 172, 297*. Sage Cock, 319*. 
Provaux City, 142. Sage Creek, 242, 300%. 
Provaux River, 219, 225. Sage Hens, 70. 
Pseudostoma castanops, 313*. Saleeratus, 417%. 
Pterocyanea rafflesii, 307, 322*. Saleratus Lake, 274*, 
Pumbar’s Creek, 78. Salt bed, 110. 

Pungun Spring, 294*. Salt deposites, 158. 


Putorius erminea, 311*. Salt Lake, antiseptic quality of its water, 
ee vison, 311*. Tits 
Saltness of the water of Great Salt Lake, 
102. 
Q Salt pools, 62. 
Salt, solid crust of, 190. 
Quartz, 102 Salt Springs, 101, 117, 118. 
; . ‘ Sandstone, 16, 27. 
¥ ? ? 
Querquedula carolinensis, 3822*. Sandy, 275*. 
Sandy Creek, 26, 70, 271*. 
San Pete Creek, 294%, 
R Sarcobatus, 160. 
Sarcobatus vermicularis, 394*. 


Rabbit Hollow, 232. Sauk Indians levying tribute, 17. 
Railroad route, 220. Scarcity of timber about Salt Lake, 
Railroad to Pacific, 263. 122, 
Rattlesnakes, 174. Sceleporus graciosus, 346*. 
Rattlesnakes in holes of prairie dog, 37. ee grammicus, 344*. 
Rattlesnake, mule killed by, 70. ae microlepidotus, 344*. 
Rattlesnake Creek, 246, 285*, 300*. < scalaris, 847*. 
Recurvirostra americana, 320*. Scirpus torreyi, 397*. 
- occidentalis, 326*. Scott’s Bluff, 52, 272, 288%. 
Red Buttes, 62. Sealed wives, 136. 
Red Chimney Fork, 79. Sections of strata, 20, 21, 23, 25, 34, 35. 
Red Fork, 222, 224, 276, 279. 39, 42, 44, 48, 54, 56, 57, 58, 67, 73, 
Red Kanyon, 300*. wie 
Red Springs, 214, 274*. Sego, 160, 169, 208. 
Reptiles collected, 336. Senecio filifolius, 391*. 
Revenues of Utah, 132. Ge hookeri, 891*. 
. Richard’s Post, 288*. «¢  hydrophilus. 391*. 
Rim Base, 280%*. Sessions settlement, 215*. 
Rock, Independence, 274*. Settlements, mode of establishing by the 
Rock-gate Camp, 166. Mormons, 142. 
Roofing slate, 161. Shale, 16. 
Root-digger Indians, 82, 148. Sheep Rock, 88. 
Roseaux River, 88. Sheep, Rocky Mountain, 812*. 
Route, new, through the Rocky Moun- | Shepherdia argentea, 395*, 
tains, 261. Shoshonee Indians, 77, 97, 174, 232. 
Routes from Fort Bridger to Salt Lake | Sialia macroptera, 307, 314. 
City, 75. Sidalcia malveeflora, 384*. 


Route from Fort Hall to Salt Lake, 98. | Sidalcia neomexicana, 208. 
Route from Salt Lake City to California, | Sigillaria, 403. 


87. Silk-plant, 175. 
Routes to Salt Lake, 139. Silver Creek, 222, 279%. 
Routes to Salt Lake Valley, 227. Sioux, 44. 
Rumex venosus, 394*. Sioux, dead bodies of, 43. 


Rush Creek, 94. Siredon lichenoides, 336%. 


486 ALPHABETICAL INDEX. 


Siredon maculatus, 337*. 

«©  mexicanus, 337*. 
Sisymbrium canescens, 384*, 
Sisyrinchium bermudiana, 397*. 
Small-pox, 253. 

Smith’s Fork, 230, 276*, 281*. 
Smith, Joseph, 124. 
Smithsonian Institution, 53. 
Snapping-turtle, 28. 
Snowbud, Oregon, 316*. 
Soft-shelled turtle, 23. 
Selidago missouriensis, 387*. 
South Pass, 70, 274*. 
Spanish Fork, 294*. 
Spermophilus 13-lineatus, 312*. 
Spirea dumosa, 387*. 

‘¢ opulifolia, var. 387*. 
Spiritual wives, 136. 
Spirifer, 16, 18, 19. 

‘¢ _ hemiplicata, 409*, 

obs octoplicata, 409*. 

‘¢ triplicata, 410*. 
Spring Camp, 270*. 

Spring Creek, 215, 290*. 
Spring Valley, 118. 
Slate, 208. 
Stampede, 16, 222. 
Stansbury Island, 297*. 
Staphylinus villosus, 875*. 
Stenopelmatus fuscus, 372*. 
Stenotus cespitosus, 389%. 
Stephanomeria minor, 208. 

es runcinata, 391*. 
Stethoxus'triangularis, 374*. 
Stipa striatus, 209. 
Strawberries, 23. 
Streptanthus crassicaulis, 383*. 

xe sagittatus, 384*. 

Strepletasma, 173, 211. 
Strix cunicularia, 37. 
Strong’s Knob, 195, 199, 202, 297*. 
Sturnella neglecta, 316*. 
Sufferings for want of water, 108. 
Sulphur Creek, 226, 280*, 300*. 


Sulphur Springs, 101, 213, 236, 282*. 


Summary of Mormon faith, 135. 
Summary of work done, 216. 
Survey of lake completed, 211. 
Swan, 100, 821*. 

Sweetwater, 238. 

Sweetwater River, 63, 274*. 
Symphemia semipalmata, 320*. 
Syringopora, 211. 


Ai 


Tanning, Use of lea in, 421. 
Tar-spring Creek, 77. 
Teal, Red-breasted, 322*. 


Teal, Green-wing, 322*. 
Tellinomya protensa, 412*. 
Temperature of spring water, 49. 
Terebratula, 401. 

Terebratula subtilita, 409*. 
Teton Peak, 286*. 

Tetradymia nuttallii, 391*. 
Tetrao urophasianus, 70, 319*. 
Thaspium montanum, 389*. 
Thirty-two Mile Creek, 271*, 272%. 
Three Buttes, 93. 

Thunder storm, 221. 

Timber, 70, 94, 218, 250. 
Timber, character of, 79. 

Timpa Creek, 294*. 
Timpanogas, 215. 

Timpanogas River, 294*. 
Titmouse, black-head, 316*. 
Tophet, 189. 

Tradescantia, 32, 34. 

Trap dikes, 222. 

Trees, their destruction by fire, 32. 
Triangles, tabulation of, 298*. 
Triglochin maritimum, 396*. 
Triteleia grandiflora, 397*. 
Trochalus explanatus, 374*. 
Trona, 420*. 

Tropidolepis scincicaudus, 348*, 
Trout, 77, 98, 94, 141, 241. 
Tufa, 202. 

Tuilla Valley, 18. 

Turkey Creek, 23, 271*, 291*. 
Turn Creek, 291*. 

Turret Rock, 181. 


U 


Uintah Mountains, 73. 
Uintahs, 120. 
University, 143. 
Uta, 344%. 

“ ornata, 344*. 

‘¢ gtansburiana, 345*, 
Utah Creek, 294*. 
Utah Indians, 202. 


Vy 


Vache 4 lait, 175. 


Vegetation, change of, 50. 
Vermilion Creek, 292*. 
Vicia americana, 385%. 
Viola pedunculata, 383*. 
Volcanic rocks, 87. 
Vulpes macrourus, 309*, 


ALPHABETICAL INDEX. 487 


Wolverene, 311*. 


W Woodpecxer, Lewis, 819*. 
Woodpecker, yellow-bellied, 326%. 
Wahsatch Mountains, 78. Woollen Factory, 142. 
Wa-ka-tiky Creek, 294*. Wyeth’s Creek, 25. 


Wa-ke-te-ke Creek, 215. 

Walnut Creek, 15, 25, 271*.. 

Warm Spring, 55, 117, 129, 273%. Dg 
Warm Spring Creek, 55. 

Warm Spring water, analysis of, 419*. | Yampah, 237. 


Watage Creek, 294*. Yampas, 120. 

Water of Salt Lake, its chemical campo- | Yellow Creek, 225, 280*, 300%. 
sition, 212. Young, Brigham, 84, 125, 1388, 146. 

Water vessels of Indians, 104. Young’s Peak, 297. 


Weber River, 78, 214, 276*, 293%. 
Wild ducks, 173. 


Wild fowl, 157, 161. ZL 

Wild geese, 100. 

Willet, 820*. Zaitha bifoveata, 870*. 

Willow Creek, 294*. ‘¢ reticulata, 370*. 

Windmill Rock, 92, 293*. Zoology, 307. 

Wind River Mountain, 67. Zopherus variolosus, 376*. 
THE END. 


STEREOTYPED BY L. JOHNSON & CQ. 
PHILADELPHIA. 


\ 


mntsenthrmn Pevieto 
114 

etenncrt Gos lidar § 
CRS Rrijee , Corre 
LO ps MAE hy fete 

het SS) OSES 
Cok<ren Orn O10 a - Freon rien ph Sr 

Op Vite Maps- PHSEEo ry"! AbSF feo -/G60 


v 49 
oT 
ei 

* 
f 
s 
1 

: 

‘ 

A 


x G, re 


M4 


We 
en: 


ty = “Wars S Ga = MW GC |e = 
ty 4, y G NS y 
4 <5 Y = Vity = Ss = ae y 
Ss wes > = > “in gnise 

7) > ” + 


RARIES SMITHSONIAN INSTITUTION NOILNLILSNI NVINOSHLINS S31 


SS 


- 


+. 


NOILNLILSNI! 
LIBRARIES 


SMITHSONIAN 


Wy, 
Ot 


LNLILSNI 


we 
’ fe 


OM 


NVINOSHLINS S3IYVYdIT LIBRARIES 


any 


: 


. 


RARIES SMITHSONIAN 


INSTITUTION NOILMLILSNI 
SJINVUGIT LIBRARIES SMITH! 


S3I1YVUEIT LIBRARIES 


INSTITUTION 
Saiuvyd! 


INSTITUTION NOILNLILSNI NVINOSHLINS S31} 


of. i ee 
oY, LT, 
tg 
NVINOSHLIWS 


NWINOSHLIWS 


“A . 
: 
> 


LNLILSNI_ NVINOSHLINS S3!1YVYEIT LIBRARIES SMITHSONIAN NSTI 


SMITHSONIAN 
NVINOSHLIWS 
SMITHSONIAN 


NOILNLILSNI 


LIBRARIES 
NOILNLILSNI 


NOILNLILSNI 


RARIES SMITHSONIAN INSTITUTION NOILNLILSNI NVINOSHLINS S3I¥ 


li 


LINLILSNI NVINOSHLINS S3IYVYEIT LIBRARIES 


SAIUVYGIT LIBRARIES 


INSTITUTION 
INSTITUTION 
INSTITUTION 


SMITHSONIAN INSTI 


ir, 
% 
* 


3RARIES SMITHSONIAN INSTITUTION NOILNLILSNI 


NVINOSHLINS S3I1YVYdIt 


SMITHSONIAN 


NVINOSHLIWS 
SMITHSONIAN 
SMITHSONIAN 


w ir) 

79) ~ ty ; ut 

a : 3g 
oc . = o 2 = (a 
<3, = Ga" 
cc a ac c = Re, 
a =a [ae] _ foe) rrr, 
- we Oo ca o Not 
= 2 — = i a 

LNLILSNI_NVINOSHLIWS _S3 1}uYVvug ioe LIBRARI ES_ SMITHSONIAN INSTIT 
a = wo - , ee Lox 
aA a A > xa fs 
0 ie > = > 
; o aa ry 2 

o p p- 
wn = a 2 m 


}RARIES SMITHSONIAN _ INSTITUTION NOILNLILSNI_ NVINOSHLINS, S31YW] 
. ie - Ww D 
= <Gill> <x = <x san = if 
. 7 KM, 2 XN. 5 Gy, zkMm 3s FY 
Qo = KM® 2 AQ Tt HY 2 LM 


iy. & a xi "4 baa 
* Ps 


ARIES SMITHSONIAN INSTITUTION NOILNLILSNI NVINOSHLINS S3ZI1YVYa 


SMI 
NVI 
SMI 
NVI 

SMI 


pal oe Ay a 
w 2 u 2 w 
wc. = oc te = 
< 3 pa <x a < 
oc = ac a 

y co -- oa) ae oO 

.D ews ; O ~ O ins 

uk all Zz 4 = pipe 

JIILSNI NVINOSHLINS S3!IY¥VUdIT LIBRARIES SMITHSONIAN 

ee oe Ss if re - 

2 : : 5 $42 
> rs > = Ll > 

Ya s, 

: i= - “i Ye is 
3 va Si die - 

ys Zz ~ z Hi 

INSTITUTION NOILONLILSNI NVINOSHLINS S31ly¥vuag 

” a ” = Ww 
= i. ae ee = 
eed z = pe 4 
oH 3 e Me - 
Oo » oO keen O 
= 2, - =. 
e eae S = . 


NVINOSHLINS SSIYVYEIT LIBRARIES SMITHSONIAN INSTITUTI 
% 


ay - an fs a 
o uw a uw 2 
= oc = ow os 
= o on oc a 
Oo sie O mag ro) 5 
= ae z me, = 
_, SMITHSONIAN _INSTITUTION NOILNLILSNI NVINOSHLINS SS3I1YVUYE 
= z= 
es 
= * 5 
. 2 Pe) es es 
= a i I~ 
‘ = me ne hy), vee 
NALILSNI NVINOSHLINS S3IYVYE!IT LIBRARIES SMITHSONIAN 


INSTITUT! 


NVINOSHLINS S3ty¥Vvudl 


SMITHSONIAN 
NVINOSHLIWS 
SMITHSONIAN 


SMITHSONIAN 


IARIES SMITHSONIAN. INSTITUTION NOILNLILSNI 


is an EA ian 
r a u 2 u 
< % < < 
oe 5 0 S o 
(aa) = joa) nee oO 
ate “4 O — Oo as 

x ail Ba cul 2 me | 

NALILSNI NVINOSHLINS S3SIYWVYSIT LIBRARIES 
©) > 0 = pe 8) 
2 : 2 EY 2 
m 2 al m 
w < Hi = 7) 


NVINOSHLINS S3JIYVYE 


rf a ql A yy g 
KS = WLP, 4 4; 
x ty Ut, 
io t ~ MB * 


Fo, OUND « 


SSMITHSONIAN | INSTITUTION 


% 
4'e 


HLINWS 


SONIAN 
SHLINS 


SMITHSONIAN INSTITUTION LIBRARIES 


, he: Caw A. aka ' a . 
Pi oP 442 . 4 n - 

’ r 94 48 ge as ae : : 

wt) Paths As y; +5 ee ¢ acs . 24 4 


3 9088 00758 1663 


- pos 
Z 7 t 
z ? a Z 
Pe 
f