Skip to main content

Full text of "A treatise on the geology of Armenia .."

See other formats 


Google 


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

to  make  the  world's  books  discoverable  online. 

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

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

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

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

publisher  to  a  library  and  finally  to  you. 

Usage  guidelines 

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

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

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

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

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

About  Google  Book  Search 

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

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


'm'  SL'^^-^'-«?-?f-^)  OS  iS 


l^ilfpflpiLibrary 


!S!3a:gcSri|i3s:cat5avnisnY 
.^.  .g.  .^.  .g.  .^.  .,j^.  .,j^.  .5. 


i 


\i 


WW-l 


-  .^.  .^.  .^.  .^.  .^.  .^.  - 


* 


]ii  fV 


A    TREATISE 


ON     TRB 


GEOLOGY  OF  ARMENIA 


FELIX    OSWALD,  B.A.,  D.Sc. 


Thesis  accepted  by  the  University  of  Umdon 
for  the  Degree  of  Doctor  of  Science 


IN    TWO    PARTS 

I.    :  'G9OI^0QICi^'  ;I^PLTS    OP  ':JK:-  |pU|iNS)r  *  BT   THE 
AXJTBOJ(a^iU>Uti6ihmtiSH:J«Mkl£«     '' 


•    •••    •«<• 


U.     TBB   GB01O6ICAL   RBCORO   OF  ARMENIA 


PUBLISHED  BY   THE  AUTHOR  AT 

loNA,  Beeston,  Notts. 

J906. 
One  Guinea  nett. 


6-I- 


a:2"35»'S"-a.*w, ;  '¥". 


\i 


to 


MY    WIFE 


PREFACE 

Jn  1898  my  friend  Mr.  H.F.B.  Lynch  suggested  that 
I  ^ould  accompany  him  on  his  second  journey 
through  Turkish  Armenia,  chiefly  with  the  obje<^ 
of  investigating  the  physical  geography  and  geology 
of  the  country.  M.^  participation  in  this  journey  was 
only  renderea  possible  by  the  kindness  of  the  late 
Lord  St.  Helier,  who  granted  m^  a  special  extension 
of  leave  for  the  purpose. 

A  few  words  are  necessary  to  explain  the  delay 
in  publishing  my  observations  and  conclusions.  In 
the  first  plaod,  my  official  duties  occupy  the  greater 
part  of  the  day;  much  of  my  time  was  taken 
up  in  preparing  maps,  plans  and  ^me  of  the 
illustrations  for  Mr.  Lynch's  book  (Armenia^  2  vols. 
Longmans,  (^l)»  and  especially  in  executing  the 
hill-work  of  the  large  map  of  the  country  between 
the  Black  and  the  Caspian  Seas  (Scale  i  :  1,000,000), 
in  which  I  have  endeavoured  to  bring  out  the  natural 
features  of  the  Armenian  pl^iteau  and  its  high  border- 
ranges.  The  examination  of  my  numerous  rock- 
sedhons^  the  marshalling  of  my  observations  and  an 
exhaustive  research  into  ^e  literature  of  the  subjeA 
necessitated  still  further  clelay.  Finally,  the  require* 
ments  of  the  University  and  the  impossibility  of 
having  the  book  published  at  my  own  expense 
impelled  me — ^at  the  suggestion  c^f  my  friend  Mr. 
Russell  Norrie — to  invest  in  a  handpress  and  the 


iv  Geology  of  Armenia 

necessary  type  and  to  print  it  myself  pas^e  by  page^io^  copied 
I  am  indebted  also  to  Mr.  Norrie  for  kindly  initiating 
me  into  the  mysteries  of  setting  up  type.  Any  mis- 
prints or  irregularities  in  the  impressions  will,  I 
hope,  be  regarded  with  some  indulgence  as  the  work 
of  a  novice  in  this  mechanical  and  laborious  process. 
I  have  reproduced  my  sketch-maps,  sections  and 
figures  by  means  of  the  autocopyist. 

1  owe  much  to  the  advice  of  Frofessor  T.G.  Bonney, 
F.R.S,  who  has  been  so  good  as  to  allow  me  to 
incorporate  his  descriptions  of  some  rocks  collected  by 
Mr.  Lynch  during  his  previous  journey  in  Armenia. 
I  gratefully  acknowledge  my  obligations  to  Mr.  R. 
Bullen  Newton,  F.G.S.,  and  to  Mr.  G.C.  Crick,  F.G;S., 
for  determining  the  fossils  I  coUedled;  to  Colonel 
F.R.  Maunsell  for  furnishing  me  with  many  valuable 
geological  data,  and  to  Colonel  G.S.  Elliot  for  placing 
at  my  disposal  his  colledlion  of  rocks  from  the  Nimrud 
Dagh  and  the  coast  of  Lake  Van. 

The  first  part  (257  pp.)  of  this  thesis  contains  the 
purely  original  worK — the  result  of  my  own  observa- 
tions in  the  field  and  the  examination  of  my  rock- 
s^edlicns  vith  the  microscope.  The  second  part  is 
based  on  the  observations  of  other  travellers,  in- 
cluding my  own  researches,  and  is  the  first  attempt 
which  has  been  made  to  write  a  systematic  geological 
history  of  Armenia. 

Owing  to  want  of  space  I  have  been  obliged  to 
withhold  until  some  future  opportunity  my  dmpters 
on  the  volcanoes,  rocks,  minerals  and  mineral  springs 
of  Armenia,  together  with  appendices  on  the  plants 
and  insedls  which  I  coUedled  during  the  journey. 

FELIX  OSWALD. 

OAK   OOrPTAODi  DOIiWIOP,  LOMDOW. 

December,  1905. 


CONTENTS 

CBAPTBR  I 
• I 

CHAPTER  II 

ComzAHTDroFiB  TO  TaamnoBD       ..•.••     13 

CHAPTBR  III 

TftmsoRD  10  tHB  Vavuk  Pass     ....••      19 

CHAPTER  IV 

TBI  Vavue  Pass  10  Erzmxuu      ••«•••     $4 

CHAPTER  V 

^ggt^^yif  f0  HChTWIS  •••••••••       AO 

CHAPTBR  VI 
Khdds  to  TOTAIII •        •        •         *      68 

CHAPTER  VII 

Tdtaxh  to  Asblat      •...••••     85 

CHAPTER  VIII 

Tme  Taukic  HnoBTS r       .    100 


vi  Geology  of  Armenia 

CHAPTBR  DC 

Tbb  Nmini  Volcaxo    ••••#»••    tij 

CHAPTER  X 
CBAPTBR  Xt 

SiPAN  id  Ksucim  •  •       •       •       •       •       •    tyt 

CHAPTBR  Ztl 
Khamuii  to  TB9  B»Gte.  Cum,    •       •       •   '    •       «       •    iM 

CHAPTBR  XItt 
Tbb  B1K06L  VoLCura   •••«••••    §04 

CHAPTBR  XIV 
Emirvm  to  Trbbizond  ••••••.••    tt9 

CHAPTBR  ZV 

PBB-DBTOKiAir  Roon  or  AmmMA       ••«..•     t6t 


• 


CHAPTBR  XVI 

DsToiiXAir  AND  CAKBommovs      ••••••    tSj 

CHAPTBR  XVn 
Puxo^TkiAttic  (JuLVA  Bbm).       •       •       •       •       •       c    joo 

CHAPTBR  XVZII 

JUftASKC JO6 

CHAPTBR  XIX 

LOWSB  CUTACSOUl  ^  •     JJf 


contents  vii 

CHAPTSR  zx 
UvnOt  ClBTACBOOl  348^ 


CHAFTXR  ZZI 


•        •••••,•• 


.    38a 


CHAPTER  ZZn 


CBAPTBR  ZZIII 


•        • 


435 


CBAPTXR  XZIV 

PtiooiMi  jjm  Ptiiwocaui    ,....*.  469 

AFFENDIX 485 

BIBUOGItAPHT  •  487 

INDEX*        <»««••«••••  501 


V 


^ 


/ 


PART    I 


GEOLOGICAL    RESULTS 


OP  M7  JOURNEY  THROUGH 

TDEEISH  ARMENIA 


I.I 


•5!: 


^ 


» 

m 

-      •      • 


CHAPTER  I 


INTRODUCTORY 


Historical  events  have  repeatedly  shown  that  the 
Armenian  plateau  forms  the  natural  bridge  between 
Central  Asia  and  Southern  Europe.  The  periodical 
surging  westwards  of  tribal  hordes  through  this  narrow 
belt  of  high  land  is  due  to  the  fadl  that  Armenia  is 
merely  a  connedling  link  in  the  long  chain  of  plateaux 
which  extend  from  the  confines  of  China  right  across 
Asia  to  the  /Egesm  Sea.  This  plateau  belt  is  to  be 
regarded,  broadly  speaking,  as  a  colossal  wrinkle  of 
the  earth's  crust  extending  on  the  one  hand  through 
Europe  to  the  Atlantic  Ocean,  and  on  the  other  hand 
to  the  Pacific,  where  it  is  bordered  by  the  volcanic  fes- 
toons of  the  Malayan,  Philippine,  Japanese,  Kurile  and 
Aleutian  islands. 

The  general  character  of  this  Asiatic  earth-wrinkle 
is  typically  expressed  by  a  secftion  across  the  Tibetan 
region.  In  its  central  part  we  find  a  high  undulating 
plateau,  with  an  average  height  of  15,000  feet,  cradled 
between  rugged  border  ranges,  the  colossal  Himalayas 
on  the  south,  the  lower  but  more  ancient  Kuenlun 
Mountains  on  the  north.  To  the  north  of  the 
Kuenlun — "  the  nucleus  of  Asia  " — a  mountain  mass 
ever  since  the  Silurian  period,  there  is  an  abrupt 
fall  to   the  vast  Tarim  and  Gobi  depression  which 

B 


• 


2     .,.'..    '  Geology  of  Armenia 

at';tob  Nor  stands  only  20CX)  feet  above  the  sea- level. 
It&Vplains  and  sandy  deserts  were  formerly  flooded, 
.even  in  historic  times,  by  the  great  Mediterranean 
Sea  of  Central  Asia,  the  Hanhai  of  the  Chinese,  now 
represented  only  by  minor  evaporating  salt  lakes.  This 
great  Tarim  depression  is  framed  in  on  the  north  by 
the  Altai  Mountains,  and  on  the  N.W.  by  the  ranges 
in  echelon  of  the  Tienshan,  while  its  high  western 
wall  is  formed  by  the  break-off  of  the  Pamirs  above 
Kashgar. 

This  sedtion  reveals  the  presence  of  three  compound 
earth-waves  lying  between  ancient  tablelands,  that  of 
Siberia  on  the  north,  that  of  India  on  the  south. 
Each  wave  consists  of  a  geo-syncline  and  a  ge-anti- 
cline,  the  latter  lying  to  the  north  of  the  former.  The 
median  geo-syncline  is  represented  by  the  gently 
folded  plateau-belt  of  the  Asiatic  continent ;  indeed, 
this  scheme  of  three  successive  latitudinal  earth-waves 
can  be  traced  from  Tibet  to  Armenia  and  Asia  Minor. 
Hence  it  can  hardly  be  treated  as  a  fortuitous  coinci- 
dence, but  should  rather  be  regarded  as  the  expression 
of  a  fundamental  structural  law  affedling  the  whole 
region.  The  following  scheme  emphasizes  the  various 
homologies : — 

ANTERIOR      WAVE 


ANTERIOR 
TABLELAND 

Arabia 

Arabia  and  bed  of 
Indian  Ocean 

Hindustan 

Hindustan 


ANTERIOR 
GEO-SYNCLINE 

Mesopotamian  plain 
filled  with  alluvial  deposits 

Persian  Gulf  to  Indus  valley 


Punjab  plains 

filled  with  alluvial  deposits 

Ganges  plains 

filled  with  alluvial  deposits 


POSTERIOR 
GE-ANTICLINE 

Taurus 

Zagros  to  Makran 
and  Suliman 
(Outer  Iranian  Arc) 

Salt  Range 
Himalayas 


Introdudiory 


MEDIAN        WAVE 

ANTERIOR    GEO-SYNCLINE      POSTERIOR    GE-ANTICLINE 


Armenia  and  Azerbaijan 

Iran(Persia  and  Afghanistan) 

Kashmir 
Tibet 


Pontic,     Thrialetic,      Gokcha    and 
Karabagh     Ranges 

Alburs    and    Khorassan    Mountains 
(Inner     Iranian     Arc) 

Hindu  Kush  and  part  of  Pamirs 

Kuenlun 


POSTERIOR     WAVE 


ANTERIOR 
GEO-SYNCLINE 

Rion  and  Kur  plains 
fiiL'd  Hith  alluvial  beds 

South  Caspian 


POSTERIOR 
GE-ANTICLINE 

Caucasus 

Mid-Caspian  ridge 
to  Great  Balkhan 
and  Parapomisus 

Tienshan 


POSTERIOR 
TABLELAND 

Russia 
Siberia 


Siberia 


Part  of  the  Pamirs 
(here  much  compressed) 

Tarim  and  Gobi  depression  Tienshan  and  Altai         Siberia 

The    position    of    this    mountain -belt    of    Asia    is 

accounted  for  by  the  fadl  that  in  the  Cretaceous,  and 

particularly  in  the  Eocene  period  its  site  was  occupied 

by  a  vast  sea,   extending  from   the  Atlantic  to  the 

Pacific^    in    which    great    masses   of   limestone   were 

deposited    by    organic    agency    {e.g.,    Hippuritic  and 

Nummulitic  Limestones).     Subsidence  must  have  been 

gradual    to    enable    sedimentation    to    continue    for 

such   long  periods  of   time  without  resulting  in   an 

early  silting  up  of  the  sea.     It  is  only  this  zone  of 

thick  sediments  which  has  been  plastic  enough  (owing 

to  its  retained  water  of  sedimentation)  to  be  thrown 

into  mountain -folds.     On  the  other  hand,  where  the 

^rust  was  thick,  old  and  unresisting,  as  in  Siberia  and 

Jndia,  no  folding  was  possible. 


4  Geology  of  Armenia 

On  referring  to  the  map  of  Asia,  it  will  be  seen  that 
the  mountain  folds  in  the  central  zone  have  become 
constricted  at  certain  points,  e.g.,  in  the  Pamirs  and  in 
Armenia,  while  in  the  intervening  regions  they  spread 
out  and  form  arcs  of  large  curvature,  which  are  more 
pronounced  in  the  high  border-ranges  than  in  the 
enclosed  plateau-country.  The  great  constrictions 
occur  only  at  those  places  where  there  is  a  considerable 
mass  of  old  tableland,  e.g.,  India  and  Arabia,  to 
oppose  the  full  development  of  the  resultant  tangential 
pressure  which  in  Asia  has  proceeded  from  the  north, 
owing  to  the  vaster  extent  of  the  Siberian  tableland. 
In  Europe,  on  the  other  hand,  (excepting  in  tlie 
Pyrenees),  the  main  force  has  proceeded  from  tlie 
greater  mass  of  Africa  in  the  south.  This  constriction 
of  the  Asiatic  folds  has  resulted  in  a  considerable 
elevation  of  the  Armenian  area  above  the  neighbour- 
ing wider  plateaux  of  Anatolia  and  Iran. 

My  diagrammatic  map  of  the  structure  of  Armenia 
is  intended  to  graphically  express  the  conclusions  at 
whic:h  I  have  arrived,  with  regard  both  to  the  position 
and  the  direction  of  the  main  folds  and  fractures.  It 
shows  also  how  the  continuity  of  these  folds  has  in 
some  cases  been  interrupted  by  fractures  and  con- 
sequent depression  of  the  ground,  or  else  conc^ealed  by 
lavas  from  the  volcanoes  which  arose  along  the 
fractures. 

The  central  and  least  interrupted  fold  of  the 
Armenian  plateau  may  be  regarded  as  the  link  between 
the  Alburs  of  Northern  Persia  and  the  Antitaurus  of 
Asia  Minor.  The  Alburs  folds,  after  bending  round  to 
the  N.W.,  are  broken  off  near  Harau,  and  only  rise  up 
again  N.W.  of  the  great  volcano  Savalan  to  form  the 
little  known  mountains  of  the  Karadagh.  The  folds 
are  cut  through  by  the  transverse  valley  of  the  Araxes 
below  Ordubad,   and  are  continued   with   the  same 


f\ 


Introductory  5 

N.W.  diredtion  through  the  Karabagh  region.    The 
Western  Karabagh  range  is  the  highest  of  the  folds, 
reaching  12,832  feet  in  the  granitic  Kapujik  Dagh ;   it 
gradually  bends  round  to  W.N.W.,  through  Dammry 
Ua.gh  (11,093  feet)  and  Salvarty  Dagh  (10,422  feet),  in- 
to the  Daralagoz  district,  finally  breaking  off  in  the 
Dsynlerly  Dagh  towards  the  broad  Araxes  plain  north 
of  Ararat.    Here  the  fold  has  undergone  fradlure  and 
depression,  and  has  been  concealed,  not  only  by  the 
lavas  of  Ararat  and  Alagoz,  but  also  by  the  loss  and 
alluvial  deposits  of  the  Araxes.    A  fragment,  however, 
comes  to  light  in  the  Oligocene  outlier  of  the  Kizilyeri 
Dagh  near  Igdir,  N.W.  of  Ararat,  and  the  fold  resumes 
its  continuity  a  little  further  west  as  the  Takvaltu 
Dagh,  near  Kulpi.    Thence  it  forms  the  Shatin  Dagh, 
*'the  spine  of  .A^enia,'*  with  a  dioritic  axis,  and  it 
bends  round  to  a  W.  by  S.  direction,  dividing  Pasin 
from  Alashkert  and  Tekman.    After  passing  through 
Palanddken  (10,694  feet),  where  it  forms  the  southern 
boundary  of  the  Erzerum  plain,  this  line  of  heights 
appears  to  die  out  in  the  Karakaya  Dagh,  and  its 
place  is  taken  on  the  south  by  the  Khadi  Dagh, 
Girdim  Dagh,  Ba^ir  Dagh  and  Merjan  Dagh; — on 
the  north  by  the  ^rdige  Dagh  and  Miriam  Dagh ; 
thence  this  line  crosses  the  Frat,  to  attain  greater 
heights  fwith  W.S.W.  direction)  in  the  Melpert  Dagh 
and  Keshish  Dagh  (c.  12,000  feet).     Here  it  is  inter- 
nipted  by  the  Erzingan  depression,  but  it  soon  rises 
higher  in  the  Muzur  Dagh.    It  is  again  traversed  by 
the  Frat  above  Egin  for  the  third  time,  and  is  con- 
tinued to  the  S.W.  into  the  Antitaurus. 

The  Central  Karabagh  line  forms  part  of  the  same 
Alburs — ^Antitaurus  series.  It  is  indicated,  ^irst  of  all, 
by  the  Ardebil  depression,  traversed  by  the  S.E. — ^N.W. 
course  of  the  Kara  Su,  and  flanked  on  the  west  by  the 
great  volcano  Savalan  (15,792  feet).    The  line  is  con- 


6  Geology  of  Armenia 

tinued  to  the  N.W.  through  the  Cretaceous  Karadagh, 
surmounted  by  the  trachytic  Gaishtasar  Dagh  (9630 
feet)  and  Saigrama  Dagh  (9060  feet).  Crossing  the 
Araxes,  its  S.E. — N.W.  axis  can  be  traced  through  the 
relative  depression  of  the  Central  Karabagh  Plateau 
— Si  depression,  which  has  been  more  than  filled  up  by 
the  lavas  and  tuffs  of  numerous  volcanoes,  arranged 
in  series  along  its  axis.  The  most  important  of  these 
peaks  are  the  Kechal  Dagh,  Ishikhly,  Kizil  Bogaz, 
Mukhortolian,  Galingaya  and  Syr-er-syrchaly,  all  of 
which  rise  to  heights  between  10,000  and  12,000  feet. 
The  Central  Karabagh  depression  is  then  continued 
through  Lake  Gokcha  into  the  Darachichak  heights, 
and  this  line  finally  disappears  beneath  the  northern 
lava-flows  of  .Alagoz. 

Starting  once  more  from  the  Caspian,  we  find  a 
series  of  S.E. — N.W.  folds  adjacent  to  the  Alburs  folds, 
and  rising  up  to  form  the  Talish  ranges.  They  are  on 
the  same  line  of  strike  as  the  folds  which  have  been 
broken  off  on  the  east  coast  of  the  Caspian,  between 
Ashref  and  Asterabad.  It  is  evident  that  this  inter-r. 
ruption  in  their  continuity  is  due  to  the  depression 
of  the  area  now  occupied  by  the  South  Caspian.  The 
Talish  ranges  disappear  beneath  the  transverse  valley 
of  the  Araxes,  and  rise  up  again  to  form  the  East 
Karabagh  ranges,  among  which  Shusha  lies  centrally 
situated.  They  are  continued  with  the  same  N.W. 
direcftion  as  the  Gokcha  ranges,  or  so-called  Little 
Caucasus,  occupying  the  space  between  Lake  Gokcha 
and  the  valley  of  the  Lower  Kur.  The  richest  ores 
of  Russian  Transcaucasia  are  found  in  these  rugged 
Eastern  Border- ranges  of  Armenia.  At  the  forty-fifth 
meridian  the  folds  begin  to  bend  round  to  the  west, 
and  consist  of  three  main  lines,  w>.,  Pambak,  Be- 
zobdal  and  Lialvar,  which  disappear  completely 
beneath  the  volcanic  plateau  of  Russian  Armenia, 


IntroduAary  7 

with  the  exception  of  a  small  outlier  of  Cretaceous 
rocks  near  Modotappa  lake.  The  Aglagan — Bezobdal 
fold  already  shows  indications  of  bending  round  to 
W.  by  S.  Although  now  completely  hidden  by  vast 
accumulations  of  lava  and  tuff,  the  continuation  of 
these  folds  may  perhaps  be  indicated  by  the  W.S.W. 
dire<ftion  of  a  part  of  the  Kars-chai  valley  below  Kais, 
and  by  the  same  direAion  of  the  volcanic  watershed 
between  the  lakes  of  Chaldir  and  Khozapin. 

On  the  west  side  of  the  volcanic  plateau  this  series 
of  folds  reappears  to  view,  forming  a  N.E. — S.W.  izone 
through  the  Olti  depression.  They  are  still  more 
clearly  indicated  further  to  the  south-west  by  the 
Cretaceous  heights  lying  between  the  longitudinal 
valleys  of  the  Chorokh  (which  doubles  upon  itself 
so  remarkably) — and  further  west  by  the  synclinal 
area  of  the  Varzahan  plain  and  the  Kelkid  valley. 
The  chief  line  of  these  folds  is  marked  by  the  heights 
of  Akhbaba,  Khoshabpunar,  Jejen,  Kop,  Otluk,  Sipi- 
kor  and  Chardaklu — a  line  which  clearly  defines  tlie 
northern  edge  of  the  Frat  basin,  and  which,  together 
with  the  opposite  Muzur  Dagh  line  of  heights,  passes 
over  into  the  Antitauric  system. 

A  still  more  northerly  series  of  folds,  the  Thrialetic- 
Imeretian  Mountains,  or  Northern  Border-ranges  of 
Armenia,  is  separated  from  the  Lialvar  line  of  folds 
by  a  hard,  resisting  block,,  which  I  have  called  the 
Khram  *  horst^'  a  mass  of  gneiss,  ancient  schists  and 
Palaeozoic  rocks  (see  Chapter  XV).  Among  these 
folds  four  main  lines  may  be  distinguished : — 

1.  Bielyikliuch  to  Yagluja  Dagh. 

2.  Manglis  to  Tiflis  (the  waterparting  between 
the  Alget  and  Vera  rivers). 

3.  Arjevan  through  Digom  to  Lilo.  This  third 
fold  bends  roimd  to  the  W.S.W. ,  west  of  the  Arjevan 
Dagh,  through  Karakaya  Sanislo  (9351  feet)  and  the 


8  Geology  of  Armenia 

Oshora  Dagh  (8551  feet),  to  cross  the  Kur  valley 
just  above  Akhaltsykh;  it  then  disappears  beneath 
the  volcanic  plateau  of  Kargapunar,  Dokhuspunar  and 
the  Arzian  Dagh.  At  the  base  of  the  deep  canyons 
of  the  Poskhov  river  the  Cretaceous  rocks  of  this 
fold  are  disclosed.  West  of  the  Arzian  Dagh  it  is 
seen  to  reappear,  bounding  the  longitudinal  valley 
of  the  Imerkhevi  river,  and  it  is  continued  to  the  S.W,, 
between  Arlvin  and  Ardanuch,  into  the  Diduwe  Dagh 
(10,558  feet).  Thence  it  forms  the  inner  line  of  the 
Pontic  Range,  along  the  left  bank  of  the  Chorokh, 
and  passes  through  the  Vavuk  Dagh  to  bound  the 
Kelkid  river  on  the  right. 

4.  The  fourth  line  is  noiarked  by  the  Thrialetic- 
Imeretian  Range,  through  the  Tortizi  Dagh,  across 
the  Kur  below  Borjom ;  thence  it  is  surmounted  by 
the  peaks  of  Sagalatlo  (8288  feet),  Kageba,  Nepiskaro, 
Gotimeria  and  Taginauri.  At  this  point  it  curves 
over  to  the  S.W.,  and,  after  crossing  the  Chorokh  just 
below  the  Ajaris  confluence,  this  line  proceeds  along 
the  coast  behind  Khopa  and  Arkhava.  It  appears  to 
me  likely  that  the  Karchkal  Dagh  (11,248  feet)  and 
parts  of  the  Pontic  Ranges  have  acfted  as  a  ^horst^' 
round  which  the  folds  have  been  curved  and  bent. 

On  the  other  hand,  the  extension  of  this  Thrialetic 
fold  from  Mtskhet  e9stwards  appears  to  lie  through 
Mamkodi,  bending  round  E.  by  S.  through  Jatistavi 
Dagh  to  vanish  beneatli  the  great  Kur  depression.  A 
remnaiit  of  this  fold  may  be  seen  in  the  Cretaceous 
outlier,  with  S.E.  strike,  which  lies  south  of  Signakh ; 
but,  with  this  exception,  it  is  concealed  beneath  Sau:- 
matian  deposits  up  to  the  forty-eighth  meridian, 
where  it  is  indicated  by  the  Eocene  foot-hills  of  the 
Caucasus,  and  it  finally  vanishes  beneath  the  Caspian^ 

Between  the  Antitauric  and  Tauric  series  of  folds 
the  Armenian  plateau  is  only  traversed   by  minor 


Introduilory  g 

folds,  chiefly  affedting  Lower  Miocene  limestone,  e.g., 
the  Sharian  Dagh  (south  of  Kuseh  Dagh),  the  Suteh 
Dagh  between  Sipan  and  Bilejan,  and  the  gentle  folds 
of  similar  strata  oetween  Lakes  Van  and  Urmi.  All 
these  folds  show  a  general  agreement  with  the  Anti- 
tauric  series,  but  their  continuity  is  frequently  broken 
by  fault-scarps,  and  concealed  either  by  the  ejecfta- 
menta  of  volcanoes,  or  by  recent  lake-deposits. 

The  Tauric  folds  form  a  S.W. — N.E.  series  in  the 
zone  between  Malatia  and  Samsat,  and  extend  to  the 
N.E.,  embracing  the  longitudinal  lake  Goljik,  and 
proceeding  to  Palu.  Here,  however,  they  appear  to 
meet  a  ^horst,'  composed  of  mica-schists,  slates  and 
marbles  (see  Chapter  XV).  A  southerly  line  of  folding 
bends  roimd  this  ^horst^*  through  Haini  and  Hazo, 
towards  Sert.  A  still  more  southerly  member  of  the 
Tauric  series  lies  between  Samsat  and  Urfa,  with 
a  W.S.W. — E.N.E.  strike ;  it  disappears  beneath  the 
basalt  of  the  Karaja  Dagh,  and  rises  again  in  the 
Mardin  hills,  running  nearly  due  east.  This  line  of 
folding  crosses  the  Tigris  above  Jezireh,  and  gradually 
bends  round  to  the  E.S.E.,  through  Judi  Dagh  and 
Shemakh  Dagh,  to  pass  into  the  Zagros  Mountains 
with  a  constant  S.E.  diredtion. 

In  Armenia  the  chief  eras  of  moimtain-folding 
seem  to  have  occurred  in  the  Lower  Permian,  ante- 
Tithonian  and  post-Oligocene  periods ;  and  in  all  of 
these  the  resultant  pressure  came  from  the  south,  i.e., 
from  the  Arabian  tableland.  The  northern  limit  to 
all  this  mountain-folding  in  Armenia  was  formed 
by  the  great  granitic  *  horst '  of  the  Meschic  Mountains. 
The  Caucasus  in  those  times  was  represented  by  an 
dliptical  island,  lying  to  the  north  of  this  resisting 
mass,  and  was  only  slightly  affeAed  by  the  force  from 
the  south.  But  in  the  post-Miocene  era  of  mountain- 
makingy  to  which  the  Caucasus  owes  its  present  great 


10  -  Geology  of  Armenia 

development,  the  force  proceeded  from  the  N.E.;  and 
by  this  time  the  sediments  of  Armenia  had  lost  their 
plasticity,  and  could  not  undergo  any  fiurther  folding. 
This  hard,  unyielding  area  now  became  fradlured  into 
blocks,  most  of  which  show  a  parallelism  to  the 
Caucasus,  and  are  as  much  an  expression  of  the 
tangential  stresses  in  the  earth's  crust  as  the  folds 
of  the  Caucasus  itself. 

The  lines  of  fracfture  in  Armenia  which  are  due 
to  this  post-Miocene  mountain- making  can  be  divided 
into  three  classes : — 

1.  Fradures  parallel  to  the  Caucasus,  z.e.,  N.W. — 
S.E,  They  have  given  rise  to  the  depressions  of  the 
Lower  Kur  from  Gori  to  the  Caspian,  the  Gokcha — 
Karabagh — Savalan  zone,  the  Kars — Ararat — Nakhi- 
chevan zone,  the  Alashkert  zone,  the  Erzerum — ^Ala 
Dagh — Sahend  zone,  the  Khinis — Melaskert  zone,  the 
Giimgiim — Liz  zone,  the  Erzingan — Mush — Lake  Van 
zone,  and  the  Diarbekr  zone. 

2.  Two  fradlures,  almost  meridional  (N.  by  W.  to 
S.  by  E.),  lying  on  either  side  of  the  Meschic  *  horstf 
and,  in  all  probability,  directly  due  to  its  resistance. 
The  western  line  runs  from  Elburz,  through  the  Arzian 
and  Soganly  series  of  volcanoes,  to  Kuseh  Dagh,  and 
thence  to  Sipan ;  the  eastern  line  extends  from  Kasbek 
through  Alagoz  to  Ararat,  and  along  the  Turco- 
Persian  frontier  to  Rowanduz. 

3.  A  fradure  crossing  the  Armenian  mountain- 
folds  at  their  concave  bend,  nmning  from  Karajadagh 
through  the  huge  volcanoes  of  Nimrud,  Sipan,  Tan- 
durek  and  Ararat  to  the  Murow  Dagh,  and  across 
the  Kur  depression  to  Shemakha.  In  fad:,  the  largest 
volcanoes  of  Armenia  have  arisen  along  this  line, 
especially  where  it  intersedte  other  lines  of  fradlure. 
This  line  may  be  traced  southwards  into  the  great  rift 
of  the  Jordan  valley,  the  Red  Sea  and  through  East 


Introduihry  ii 

Africa  as  far  as  the  rift-lakes  Tanganyika  and  Nyassa. 
Perhaps  another  line  may  also  be  traced  radiating 
from  the  Karaja  Dagh,  through  the  volcanic  Bingol 
Dagh  and  Palandoken  to  Erzerum  and  the  Deveh- 
boynn.  It  is  noteworthy  that  the  most  violent  of  the 
Armenian  earthquakes  occur  alon^  these  two  lines, 
that  the  volcano  Tandurek  is  still  m  a  solfatara  con- 
dition, and  that  Nimrud  was  in  eruption  so  late  as 
the  year  1441,  whilst  the  recent  disastrous  outbreak 
at  Shemakha  shows  that  the  old  forces  are  far  from 
being  extindt. 

A  study  of  these  fradhires  indicates  that  the  position 
of  the  remarkably  numerous  volcanoes  of  Armenia 
is  by  no  means  fortuitous,  that  on  the  contrary  they 
have  invariably  arisen  along  these  lines,  and  that  the 
volcanoes  which  occur  at  points  of  intersedHon  are 
proportionately  larger.  Moreover,  just  as  the  volcanic 
plateau  of  the  Central  Karabagh  is  an  area  of  relative 
depression,  lying  between  the  lofty,  serrate  border- 
ranges  of  the  East  and  West  Karabagh — ^so,  in  like 
manner,  the  Armenian  plateau,  as  a  whole,  may  be 
regarded  as  an  area  of  relative  depression  between  its 
northern  and  southern  border-ranges.  Gently  folded 
Miocene  limestones  are  characteristic  of  the  Ajmenian 
plateau,  but  never  occur  in  the  border-ranges;  on 
the  other  hand,  while  metamorphic  schists  are  very 
charadleristic  of  the  Taurus  (the  southern  border- 
range),  they  occur  only  quite  exceptionally  in  the 
plateau-region,  e.  g.,  in  the  Akdagh,  north  of  Khinis. 

The  depressions  caused  by  the  subsidence  of  blocks 
of  land  along  the  lines  of  fracture  became  filled  by 
lakes  during  the  Pliocene  and  Pleistocene  (see  Chapter 
XXVI),  The  lavas  and  tuffe  of  the  numerous  con- 
temporary volcanoes  became  interbedded  with  the 
lake-deposits,  and  assisted  in  levelling  the  pre-existing 
inequalities  of  the  ground.    In  this  way  plains  were 


I  a  Geology  of  Armenia 

formed,  varying  in  size  from  the  great  mid-Araxes 
depression  at  the  foot  of  Ararat  to  the  plains  of 
Alashkert,  Pasin.  Erzerum,  Khinis,  Mush,  etc.,  down 
to  little  fertile  *ovas^'  sudb  as  those  of  Bashkent  and 
Gundemir.  The  draining  of  the  lakes,  caused  by  the 
outflowing  rivers  gradu^y  deq>enin^  their  courses, 
has  perhaps  contriouted  to  the  cessation  of  volcanic 
adtion.  Even  within  historic  times  (as  stated  above, 
and  see  Chapter  VIII),  an  eruption  has  been  recorded 
of  the  Nimrud  Daeh  oa  the  snores  of  Lake  Van — the 
largest  survival  of  the  sheets  of  water,  which  once 
extended  not  only  over  Armenia,  but  over  the  adjacent 
plateaux  of  Anatolia  and  Persia. 


(?)  ioruu   ojT  toiUAnacaj 


CHAPTER  II 

CONSTANTINOPLE  TO  TREBIZOND 

A  preliminary  word  is  necessary  to  explain  the  paucity 
of  my  geological  observations  during  the  first  part  of 
our  journey  as  far  as  Erzerum.  Owing  to  the  dis- 
trust with  which  the  Porte  regarded  the  motives  of 
Europeans  travelling  into  the  interior  of  Asia  Minor 
and  especially  into  the  disturbed  Armenian  provinces, 
it  was  not  possible  for  us  to  obtain  even  an  ordinary 
passport  or  tezkereh.  But  since,  every  British  subject 
has  a  treaty  right  to  travel  in  Turkey,  our  Embassy 
decided  to  make  a  test-case  of  our  intended  journey. 
Accordingly  we  embarked  under  its  auspices,  without 
any  official  permission  from  the  Turkish  Government. 
It  was  not  until  we  reached  Erzerum  that  we  heard  of 
the  successful  issue  of  the  negotiations.  Therefore, 
until  the  matter  had  been  decided  in  our  favour,  it 
was  considered  advisable  to  travel  through  the  coast 
province  as  rapidly  as  possible,  and  to  avoid  making 
frequent  notes  or  observations,  which  might  have  ex- 
cited the  suspicions  of  officials. 

We  left  Constantinople  in  the  afternoon  of  June  3, 
1898,  and  soon  our  steamer  was  threading  its  way 
through  the  winding  channel  of  the  Bosphorus.  It 
was  easy  to  recognize  that  this  narrow  strait  presents 
the  characteristics  of  a  drowned  valley,  with  several 


14  Geology  of  Armenia 

arms  running  into  the  land,  of  which  the  Golden 
Horn  and  Buyukdereh  are  the  most  notable.  Some- 
times the  steamboat  approaches  a  cliflF  of  Devonian 
limestone  or  clay-slate,  e.g.^  at  Anadoli  Hissar,  where 
the  Bosphorus  is  narrowest ;  at  other  times  the  strait 
widens  so  greatly  as  to  appear  like  a  land-locked  lake. 
The  rounded  hill  behind  the  Asiatic  fort  of  Anadoli 
Kavak  is  a  mass  of  intrusive  rock  rising  out  of  clay- 
slates  of  Devonian  age. 

At  the  exit  to  the  Black  Sea  the  low,  flat-topped 
Cyanean  Islands  appear  to  consist  of  a  brown  con- 
;lomerate,  dipping  about  17®  to  N.E. ;  doubtless  the 
:ragmental  nature  of  the  rock  has  greatly  aided  wave- 
adlion  in  reducing  these  islands,  since  the  days  of  the 
Argonauts,  to  their  present  insignificant  condition. 

The  north-easterly  course  of  our  vessel  soon  caused 
the  disappearance  of  the  low,  white  cliffs  of  Bithynia. 
By  eight  o'clock  on  the  following  morning  we  were 
off  Amasra,  and  the  coast  was  gradually  coming  into 
sight  again  ;  it  remained  well  under  observation  from 
Cape  Karembeh  onward.  Although  numerous  small 
transverse  valleys  break  the  continuity  of  the  lofty 
mountain -wall,  yet  the  absence  of  natural  harbours 
and  inlets  along  this  rock-bound  coast  forms  a  most 
striking  contrast  to  the  deep  indentations  of  the  west 
coast  of  Asia  Minor,  from  the  Bosphorus  to  Lycia. 
The  conclusion  is  irresistible  that  the  western  part  of 
Anatolia  has  sunk  down  since  the  formation  of  its 
valleys,  while  the  eastern  part  has  risen.  This  in- 
ference receives  support  from  the  numerous  traces  of 
recent  beach-conglomerates  near  Kerasun,  Platana* 
and  Trebizond,  often  at  a  considerable  height  above 

*  Hamilton,  /Researches  in  Asia  Minor ^  i.  246.     ''  The  low  hills  just 

before  reaching  Platana  [f.^.,  east  of  the  town]  consist  of  a  soft  shelly 
limestone,  wit^  many  fragments  and  casts  of  recent  or  post-Tertiary 
shells." 


Constantinople  to  Trebizond  15 

the  sea,  as  well  as  by  the  rapid  growth  of  the  deltas 
of  the  Halys  and  Iris. 

So  far  as  I  could  judge  from  the  deck  of  our  vessel, 
the  strata  along  the  coast  appear  to  be  composed  of 
massive  limestones,  dipping  about  30°  N.N.W. ;  on 
the  whole,  the  coast -line  nearly  follows  the  axis  of  a 
syncline,  which  is  occasionally  revealed.  High  cliffs 
rarely  occur,  doubtless  owing  to  the  seaward  dip,  and 
the  mountains  rise  up  from  the  shore  with  a  con- 
tinuous slope. 

Ineboli  was  reached  soon  after  one  o'clock.  The 
broad  valley  of  the  Iki-chai  gradually  recedes  into 
the  high  mountains  towards  Kastamuni ;  the  river 
at  this  time  of  year  is  a  mere  rivulet,  but  the  wide, 
stony  bed  bears  eloquent  witness  to  the  great  volume 
of  water  which  must  rush  down  from  the  mountains 
in  time  of  storm.  The  fields  on  the  hillside  west 
of  the  Iki-chai  show  a  very  black,  fertile  soil,  forming 
a  marked  contrast  to  the  brown  or  purple-brown 
colour  of  the  fields  east  of  the  river.  This  circum- 
stance might  lead  one  to  conclude  that  Tchihatcheff's 
map  is  corredt  in  marking  an  outcrop  of  coal-measures 
at  Ineboli.  However,  we  were  informed  by  an  Italian 
resident — the  contractor  for  repairing  the  road  to 
Kastamuni — that  no  coal  is  found  here,  but  that  an 
excellent  mine  occurs  at  Kastamuni,  yielding  "fine 
black  coal,  as  good  as  any  from  Cardiff."  Lead  and 
copper  are  also  obtained  in  its  proximity. 

The  Acropolis  hill  lies  just  on  the  east  side  of 
the  Iki-chai  and  displays  two  cliff -sedlions,  revealing 
a  series  of  pale  yellow  limestones  alternating  with 
blue-grey  shales,  dipping  about  45°  S.S.E.  The  lime- 
stone flags  vary  in  thickness,  but  do  not  exceed  one 
foot;  they  appeared  to  be  barren  of  fossils,  but  the 
time  at  my  disposal  only  permitted  a  very  cursory 
examination.    The  other  cliff-sedlion  was  nearer  the 


i6  Geology  of  Armenia 

sea,  and  consisted  of  pale  yellow-brown,  calcareous 
flagstones,  dipping  60°  N.N.W.,  and  apparently  over- 
lying the  limestones  and  shales.  Darlmess  soon  set  in 
aiter  leaving  Ineboli,  but  the  flagstones  clearly  con- 
tinued for  some  distance  eastwards  along  the  coast, 
with  nearly  vertical  dip. 

Samsun  was  reached  early  the  next  morning,  but 
it  was  not  considered  advisable  to  land.  The  rolling 
downs  surrounding  the  town  would  appear  to  be 
composed  of  limestone,  dipping  about  10°  N.W. ;  small 
caves  occur  in  a  low  cliff  at  the  west  end  of  the 
wide  bay.  The  east  end  presents  different  features. 
Here  three  remarkably  abrupt  bosses,  clothed  with 
thick  scrub,  rise  out  of  the  treeless  downs,  one  behind 
the  other,  from  S.W.  to  N.E.,  the  last  one  facing 
the  sea  in  a  low  black  cliff.  It  is  highly  probable 
that  they  represent  intrusive  masses  of  a  doleritic 
nature. 

The  extensive  delta  of  the  Iris  or  Yeshil  Irmak 
necessitated  a  considerable  detour;  the  sea  is  dis- 
coloured for  miles  by  the  yellow  sediment  of  the  river.  • 
When  the  low,  featureless  delta  is  at  last  left  behind, 
the  mountains  of  the  seaboard  assume  the  appearance 
of  mammillated  cones ;  snow  still  covered  the  summit 
heights  far  inland.  The  land,  however,  soon  receded 
again  from  sight  as  we  stood  out  in  order  to  double 
the  bold  headland  Yasun  Burun  (Jason's  cape).  This 
promontory,  according  to  Tchihatchefifs  map,  is  com- 
posed of  tradiyte,  and  it  certainly  presents  a  contrast 
to  the  rounded  contours  of  the  limestones,  which  thus 
far  prevail  along  the  coast.  Its  hills  are  deeply 
furrowed  by  steep-sided  valleys,  and  the  soil  is  a 
rich  red-brown. 

Kerasun  was  reached  shortly  before  sunset.  The 
citadel  hill,  which  bounds  the  wide  bay  on  the  east,  is 
a  rugged,  precipitous  height  of  iron-grey  lava,  "  pyro- 


Constantinople  to  Trehtzond  17 

xenic  porphyry*'  according  to  Tchihatcheff,^  and 
perhaps  similar  to  the  grey  "pyroxenic  porphyry" 
(a  leucite-tephritej  of  Trebizond,  but  I  did  not  have 
the  opportunity  ot  landing  here. 

Finally  we  reached  Trebizond  at  sunrise  on  the 
6th  of  June.  The  low  Kalmek  promontory,  on  the 
west  side  of  the  present  anchorage,  is  crowned  by 
the  lighthouse  and  the  ruined  Genoese  castle;  the 
cliff  here  clearly  shows  a  dark  lava- flow,  interbedded 
in  a  paler  rock,  which  may  be  sedimentary,  for  Koch| 
states  that  the  road  behind  this  point,  from  shore 
to  shore,  cuts  through  a  "porphyry-conglomerate" 
with  a  very  compadl,  argillaceous  cement. 

Whenever  the  bare  rock  is  exposed  to  view  at 
Trebizond,  it  consists  of  a  grey  lava,  a  leucite-tephrite 
(vide  infra)y  with  conspicuous  crystals  of  augite.  On 
the  east  side  of  the  town,  Boztepeh,  the  flat-topped, 
*  grey  hill,*  rises  to  about  800  feet  above  the  sea, 
with  precipitous  sides,  excepting  where  it  joins  on 
to  an  outspur  of  the  Pontic  Range.  It  is  mainly 
composed  of  massive  flows  of  grey  lava,  but  at  a 
spot  about  halfway  up  the  hill  on  the  west  side, 
near  a  cave  with  a  spring  of  clear  water,  I  noticed 
a  horizontal  bed,  10 — 15  leet  thick,  of  well  stratified 
sand,  altematitig  with  a  fine  brown  gravel  of  pebbles 
up  to  an  inch  in  diameter.  This  bed  is  intercalated 
in  the  lava-flows  and  may  represent  a  former  littoral 
deposit,  but  no  fossils  of  any  kind  could  be  detected. 

A  similar  occurrence  is  visible  in  the  cliff  beneath 
the  eastern  wall  of  the  citadel  of  Trebizond ;  here  the 
wall  rests  on  a  coarse  conglomerate,  full  of  rounded 

*  Tchihatdieff,  Asie  Mttuure^  GidogU^  L  260,  states  that  the  rock 
has  a  reddish-grey  matrix,  with  crystab  of  labradorite  and  dark  green, 
lustrous  pyroxene  (2-5  mm.) ;  and  that  it  passes  into  a  breoda  forming 
th  *  picturesque  heights  on  which  Kerasun  lies. 

t   Koch,  Rme  im pontischen  Gtbirge^  ii.  35. 

c 


1 8  Geology  of  Armenia 

boulders  of  shale,  limestone  and  grey  lava,  with  a 
calcareous  cement  enclosing  augite  crystals.  Perhaps 
the  nearly  level  summits  of  Boztepeh  and  the  citadel- 
height  may  indicate  the  ancient  marine  terrace  of 
denudation. 

The  specimen  of  rock  (No.  i),**  which  I  obtained 
near  the  Kizlar  monastery,  halfway  up  Boztepeh, 
agrees  very  closely  with  the  leucite-tephrite  from 
Trebizond,  described  by  Lacroix.f 

•  The  rock  is  light-grey  and  compact,  with  rough  irregular  fracture ; 
it  contains  abundant  augite  crystals  (up  to  3  mm.),  and  numerous  tiny 
white  specks  of  leucite.  A  few  irregular  c  a\aties  are  lined  by  a  zeolite. 
Under  the  microscope  the  augite  idiomoiphs  are  quite  fresh,  a  pale 
yellow-green,  often  with  darker  green  zonal  layers ;  frequently  twinned, 
sometimes  repeatedly ;  beautifully  zoned,  one  crystal  showing  herring- 
bone twinning ;  frequently  in  groups,  often  radiate,  macles  also  occur ; 
very  feebly  pleochroic,  tncloses  apatite  and  magnetite.  Leucite  in 
small,  fairly  abundant  idiomorphs,  \^ith  faint  intersecting  twinning  only 
in  the  larger  crystals ;  it  encloses  a  few  tiny  augite  and  magnetite 
granules  and  dust,  usually  with  no  definite  arrangement,  but  sometimes 
in  a  peripheral  zone ;  either  quite  fresh,  or  altered  to  analcime,  and 
often  bordered  by  a  narrow  wreath  of  augite  granules.  Apatite 
occurs  in  the  stout  idiomorphs  characteristic  of  felspathoid  rocks, 
rather  larger  than  the  leudtes,  occasionally  in  groups.  Magnetite 
grains  fairly  frequent,  chiefly  associated  with  augite.  The  groundbiass 
contains  abundant  small  leucites,  differing  from  the  larger  leucites  in 
being  quite  isotropic  or  replaced  by  analcime ;  sometimes  with  crystal- 
outlines,  but  often  merely  rounded.  The  interstitial  substance  consists 
of  abundant  tiny  plagioclase-laths  (difhcult  to  observe  owing  to  zeo- 
litization^,  numerous  small  augites,  and  magnetite  granules.  The  rock 
is  practically  saturated  with  finely  granular,  isotropic  analcime.  A 
second  zeolite  (christianite)  is  also  present,  either  lining  or  filling 
cavities  and  cracks,  in  clear,  stumpy,  radiating  crystals,  polarizing 
in  low  greySj  with  straight  extinction.  A  cavity,  to  which  it  forms 
a  lining,  has  been  filled  with  calcite. 

t  Bull.  Soc,  Geol,  France,^  idx.  732 — 740. 


CHAPTER  III 

TREBIZOND  TO  THE  VAVUK  PASS 

We  started  early  the  next  morning  (June  7)  for 
Erzeruniy  driving  in  an  open  carriage  owing  to  the 
impossibility  of  obtaining  satisfactory  horses  at  short 
notice.  The  road  passes  in  front  of  Boztepeh  to  the 
stony  delta  of  the  Deirmendereh  (Pyxitis),  and  then 
proceeds  up  the  left  bank  of  the  river.  A  great 
quantity  of  gravel  and  sand  is  brought  down  by 
the  swift  stream,  discolouring  the  sea  for  a  con- 
siderable distance.  The  wide  embouchure  rapidly 
narrows,  and  the  grey  rock  of  Boztepeh  gives  place 
to  precipitous  cliffs  of  black,  columnar,  basaltic  lava. 
Near  the  first  bridge  (4^  miles)  some  buff  strata 
(occasionally  conglomeratic)  occur  interbedded  in  dark 
purple  lavas,  and  are  puckered  in  small  wrinkles, 
which  have  probably  been  caused  by  the  lava  whilst 
flowing  over  these  sediments.  A  little  further  on 
a  band  of  dark  green  lava,  about  20  feet  thick,  is 
intercalated  in  the  series.  This  alternation  of  voir 
canic  and  sedimentary  rocks  (with  a  general  dip  of 
about  35^  N.  by  W.)  prevails  for  several  miles  up 
the  river ;  the  latter  soon  predominate  and  the  valley 
immediately  widens  out,  the  hills  becoming  more 
rounded,  with  gentle  contours.  However,  a  conical 
ridge   of  igneous  rock,   with  E. — W.   axis,   forms  a 


20  Geology  of  Armenia 

conspicuous  feature  on  the  right  bank.  We  made 
a  short  halt  at  the  tenth  milestone,  near  a  small 
quarry  for  road-metal.  The  rock  is  a  greenish -white 
tuflF  (No.  2),  too  much  decomposed  to  permit  of  an 
exadl  determination  of  the  original  constituents.  It  is 
highly  chloritic  and  calcareous,  and  yields  an  ar- 
gillaceous odour  when  moistened.  It  weathers  to  a 
pale  buff  colour  and  is  not  improbably  identical  with 
the  interbedded  sedimentary  strata  already  noticed. 
Volcanic  dikes  soon  became  evident,  cutting  through 
reddened  shales,  which  dip  30°  N.  by  W,  Lavas  now 
predominate,  and  the  valley  narrows  considerably, 
forming  a  deep,  rocky  gorge.  Near  a  wayside  burial 
ground  with  a  solitary  cypress  tree,  the  road  crosses 
over  to  the  right  bank,  and  rises  above  the  river-bed. 
Near  this  place  a  cliff-sedlion  shows  that  the  lava 
has  caught  up  some  shaly  strata,  baking  the  large 
flakes  of  rock  to  a  pale  pink  colour. 

The  left  bank  at  Haseveran  rises  in  an  imposing 
precipitous  bluff  to  a  height  of  about  2000  feet  above 
the  river ;  it  is  composed  of  successive  flows  of  black 
lava,  in  long  narrow  columns.  Near  the  summit 
the  lava-flows  are  interbedded  with  nearly  horizontal 
strata  of  a  buff  colour,  and  the  \^  hole  mass  is  crowned 
by  a  cap  of  similar  beds.  A  deep  gorge  of  a  left 
tributary  bounds  it  on  the  south,  and  gives  full  ex- 
pression to  its  height. 

Half  an  hour  after  passing  this  locality  we  reached 
the  pidluresque  village  of  Jevizlik  (990  feet,  20  miles 
from  Trebizond).  Just  above  the  bridge  there  occurs 
a  thin  bed  of  grey  brecciated  limestone,  overlaid  by 
a  dark  green  decomposed  lava,  and  this  again  by 
shaly  beds  dipping  30°  N.  by  W.  So  far  as  I  could 
judge,  lavas  prevailed  in  the  cliffs  above  these  strata. 

Our  road  now  turned  up  the  Machkadereh,  and  rose 
up,  higher  and  higher,  on  the  right  bank,  which  is 


4 

r 


:®:'S(?'      ,-  "M"      to 


Trehizahd  to  the  Vavuk  Pass  21 

composed  of  dark  green  or  black,  columnar  lavas 
and  tuffs.  Thin,  grey  intercalary  layers  were  some- 
times visible ;  they  are  probably  seams  of  foraminiferal 
limestone  (vide  infra).  We  entered  the  rhododendron 
zone,  with  its  luxuriant  forests,  at  a  height  of  about 
2000  feet  above  the  sea.  On  the  opposite  side  of 
the  valley  the  little  white  monastery  Aiama  clings 
halfway  up  the  face  of  a  precipice;  at  this  spot 
the  rock  by  the  roadside  proved  to  be  a  fine-grained 
andesitic  tuff  (No.  3).^  An  hour  later,  at  a  much 
higher  level,  large  boulders  of  a  pale  grey  limestone 
fNo.  4)  encumbered  the  road  and  had  evidently  fallen 
from  the  heights  above.  It  is  very  hard  and  splinter\% 
with  numerous  thin  veins  of  calcite.  No  fossils  could 
be  seen  except  an  obscure  indication  of  a  Pe£ten. 
Under  the  microscope  numerous  foraminifera  become 
revealed,    consisting   chiefly  of  a  Miliolina^  with   a 

*  The  rock  is  compact,  greyish  green  and  fragmental,  consisting 
of  white  or  pinkish  partides  of  felspar,  up  to  0.8  mm.,  in  a  green, 
chloritic  cement.  A  rude  stratification  is  visible ;  the  rock  is  highly 
calcareous,  and  is  slighUy  magnetic  (2"^;  in  this,  and  the  following 
rock-descriptions  this  figiure  represents  the  number  of  degrees  which 
the  hand-specimen  permanently  deflected  the  needle  of  my  field 
compass,  when  held  dose  to  its  extremity;  in  the  most  magnetic 
spedmen  of  my  collection — a  basalt  from  Nimnid — this  figure  reached 
32'').  Under  die  microscope,  plagioclase  (andesine  aboutAb5An3), 
in  small  idiomorphs  and  fragments,  forms  the  bulk  of  the  rock;  a 
little  subordinate  orthoclase  is  present  in  smaller  fragments;  the 
felspars  are  turbid  with  brown,  dusty  decomposition-products,  and 
the  plagiodases  are  often  filled  centrally  with  chlorite  and  caJcite ; 
zoning  not  seldom  occurs.  A  good  many  grains  of  magnetite  and 
small  partides  of  limonite  are  frequent.  Apatite  is  accessory.  There 
are  no  pseudomorphs  after  ferro-magnesian  minerals.  The  larger 
plagiodases  sometimes  occur  still  embedded  in  an  andesitic  felt 
of  oligodase  microlites  with  interstitial  chloritic  matter  and  dis- 
seminated caldte.  Epidote  occurs  in  aggregates  of  small  grains, 
filling  cracks  and  lenticular  areas  or  in  irregular  patches  along  bedding 
planes;  it  is  everywhere  assodated  with  chlorite.  Calcite  in  fre- 
qnent  irregular  grains  and  equally  abundant  chlorite  form  the 
interstitial  material. 


22  Geology  of  Armenia 

less  number  of  a  Textularta.  The  rock  in  situ  by 
the  roadside  was,  however,  still  a  dark  green,  andesitic 
lava.  Now  Tchihatcheff,*  whilst  travelling  along  a 
more  westerly  route,  found  narrow  bands  of  limestone, 
fwith  Nummulites  Ramondi^  Defr.,  and  other  Eocene 
lossils),  similarly  interbedded  in  augite  andesites,  all 
the  way  from  the  Hassanly  plateau  (south  of  Tireboli) 
up  the  Kharshut  valley  as  far  as  Giimiishkhaneh. 
It  would  seem,  therefore,  very  probable  that  the  lavas 
which  form  the  northern  slopes  of  the  Pontic  Range 
were  submarine  and  of  Eocene  date. 

After  another  hour's  drive  the  sombre  lavas  were 
interrupted  for  30 — 40  yards  by  a  friable  yellowish 
brown  travertine  (No.  5),  saturated  with  water;  it 
is  evidently  the  same  deposit  which  Wagnerj  noticed 
in  1852.  In  a  quarter  of  an  hour  the  houses  of  Lower 
Hamsikoi  came  into  sight,  just  after  passing  a  high 
precipitous  bluff  on  our  left.  It  was  already  dark 
before  we  climbed  up  the  opposite  bank  by  a  zigzag 
road  to  Upper  Hamsikoi  (4060  feet,  15^  miles  from 
Jevizlik).  The  rock  here  is  still  a  dark  green,  compac5l 
andesite,  with  glancing  felspar  and  augite  crystals, 
but  the  bare  rock  is  rarely  exposed  to  view  in  this 
moist  zone  owing  to  the  luxuriance  of  the  vegetation 
and  the  thickness  of  the  soil.  The  road  now  turns  up 
the  valley  of  the  Maden  Su,  which  derives  its  name 
from  some  disused  copper  mines.  Whilst  driving 
through  the  dense  forest  I  noticed  a  bold  crag  of 
mica-porphyrite  (No.   6)f   projedling   into   the   road 

*  Asie  Mineitrt:  Geohgie^  i.  266,  ii.  428,  and  Palkontologie^  195,  422  ; 
and  see  my  Chapter  XXII  00  the  Eocene  of  Armenia. 

t  ReUe  nach  Persien^  i.  166.  His  place-names  along  this  part  of 
the  route  need  some  modification :  his  '  Tchebislik '  should  be 
Jevizlik,  'Hapsikoi/  Hamsikoi,  'Sehana,'  Zigana,  'Ardaso,'  Ardasa, 
•Dekoi,'  Tekkeh. 

%  The  rock  is  very  compact,    a  pale  green-grey,  with  abundant 


TYehitond  tty  the  Vavuk  Pass  23 

near  the  31st  milestone.  It  is  evidently  a  dike  which 
has  traversed  the  more  easily  decomposed  andesit^ 
and  porphyrites.  Finally  we  reached  the  tree-limit 
near  the  35th  milestone  at  about  5500  feet  above 
the  sea,  and  an  entire  change  of  scenery  takes  place. 
The  Zigana  pass  is  approached  by  a  laborious  and 
winding  ascent  up  the  sides  of  steep,  stony,  rounded 
slopes,  quite  bare  of  shrubs  or  trees,  and  the  snow 
in  the  hollows  had  not  yet  melted  (June  8th).  The 
rock  is  decomposed  to  so  great  a  depth  that  no 
time  could  be  spared  for  an  attempt  to  obtain  a  fresh 
specimen.  The  occasional  sections  along  the  road 
merely  exposed  a  soft,  pale  greenish-white  rock, 
probably  a  much  kaolinized  and  chloritized,  granitic 
or  dioritic  rock.  Since,  during  our  second  traverse 
of  the  Pontic  Range  (Chapter  XIV),  I  found  the  axis 
(the  Kazikli  Dagh)  to  consist  of  a  mica-diorite,  easily 
decomposing,  it  is  not  improbable  that  the  rock  of  the 

white  striated  felspars  (up  to  6  mm.),  and  occasional  specks  of  biotite 
(up  to  2  mm.) ;  fracture  splintery ;  brisk  effervescence  with  acid, 
under  the  microscope,  the  plagioclass  (an  acid  labradorite  about 
(AbiAni)  occurs  in  large  tabular  idiomorphs,  sometimes  in  groups; 
frequently  zoned,  and  much  traversed  by  wavy,  roughly  parallel 
cracks,  along  which  much  calcite  occurs ;  the  crystals  have  sometimes 
been  nearly  replaced  by  calcite.  Biotite  is  fairly  frequent,  down  to 
qiute  small  dimensions ;  alteration  is  here  very  unequsd  :  the  crystals 
are  sometimes  quite  fresh,  with  a  peripheral  zone  of  magnetite  granules, 
but  they  have  frequently  been  altered  into  yellow-green  chlorite  (some- 
times spherulitic)  and  magnetite^  although  some  of  the  smallest  flakes 
may  remain  fresh.  A  few  pseudomorphs  after  hornblende  occur, 
now  quite  altered  into  an  earthy  mixture  of  calcite,  chlorite  and 
magnetite  granules,  but  with  still  characteristic  outlines.  Apatite 
is  accessory,  chiefly  enclosed  by  magnetite  and  biotite.  The  ground- 
mass  is  microcrystalline  and  somewhat  obscured  by  dusty  decom- 
position-products ;  it  is  an  orthophyric  felspar-mosaic,  mainly  of  short, 
stumpy  prisms  of  orthoclase,  and  also  some  oligodase  microlites, 
with  intersdtial  quartz,  a  few  tiny  biotite  flakes,  and  granules  of 
magnetite;  some  quartz  crystals  project  into  a  cavity  now  filled  by 
calcite. 


wwm-w—rm 


24  Geology  of  Armenia 

Zigana  Dagh  (which  lies  on  the  same  axis)  is  of 
the  same  nature.  Unaltered  rock  could  certainly  not 
be  reached  at  a  less  depth  than  20  feet  below  the 
surface.  From  the  guardhouse  at  the  summit  of  the 
pass  (6680  feet ;  10^  miles  from  Hamsikoi),  a  fine 
panorama  of  broken  heights  lay  at  our  feet,  bounded 
on  the  distant  horizon  by  the  dark,  serrated  peaks 
of  the  Giaur  Dagh,  still  swathed  in  snow.  Here 
a  zaptieh  showed  me  a  specimen  of  galena  in  quartz, 
which  (according  to  him)  had  been  obtained  close 
at  hand,  probably  from  the  old  workings  of  Zigana 
Maden,  below  the  pass. 

The  descent  into  the  basin  of  the  Kharshut  river 
is  very  steep ;  on  this  side  the  rocks  are  likewise 
greatly  decomposed,  and  appear  to  be  almost  entirely 
kaolinized  (No.  7).  Huge  vertical  dikes,  30 — 40  feet 
thick,  similarly  altered,  cut  through  the  softer  rocks 
and  seam  the  precipitous  slopes  from  top  to  bottom. 
Although  the  descent  is  rapid,  yet  there  is  no  return 
of  the  luxuriant  forests  of  the  northern  slopes,  for 
the  saturated  winds  from  the  Black  Sea,  in  striking 
against  the  lofty  barrier  of  the  Pontic  Range,  have 
lost  their  superabundant  moisture. 

About  a  mile  and  a  half  below  the  pass  there 
lies  the  little  village  of  Maden,  probably  the  site 
*  of  ancient  copper  or  lead  mines.  At  the  Lower 
Zigana  village  (4330  feet),  a  few  miles  further  on, 
dark  green  stratified  rocks  appeared  with  thin  seams 
of  ironstone,  dipping  44°  N.  These  chloritic  sand- 
stones (Nos.  8,  9,  and  10)^  are  composed  entirely  of 

•  The  rock  is  dark  green,  compact  and  fine-grained,  the  layers 
varying  somewhat  in  fineness  of  grain,  with  lenticular  angular  fragments 
of  dark  green  chlorite  (up  to  1.5  cm.  wide,  and  3  mm.  thick), 
white  felspars  (up  to  0.8  mm.),  and  small  particles  of  limonite. 
The  groundmass  contains  the  same  constituents  on  a  finer  scale,  and  is 
very  calcareous.  No.  9  shows  an  even  layer  (4  mm.  thick)  deeply 
stained  with  iron  oxide,  and  No.  10  is  traversed  by  a  vein  of  calcite 


KftCVKMN 
I 


900' 


I   r 


V^ 


Xf  ^  AN  A 


MfiMDi 


Oftf. 


SUMfiXAt 


I 


TO  son  EL  AS 


I 


1    I   '-  •         •    • 


%    -x 


^ 


i  - 


\f 


^h 


•    .  A 


•»•«>-. 


.a 


Trebtzond  to  the  Vavuk  Pass  25 

detritus  from  igneous  rocks.  It  was  doubtless  near 
the  Upper  Zigana  village  (*  Sehana  *)  that  Wagner  * 
observed  "a  compact,  crystalline,  white  limestone, 
much  altered  by  contacft  with  intrusive,  plutonic 
rocks,"  but  I  was  unable  to  confirm  this,  for  a 
blinding  storm  of  rain  effecftually  prevented  any 
further  observations  during  our  drive  along  the  road 
leading  down  to  the  bridge  over  the  Kharshut  river 
(3100  feet),  but,  so  far  as  I  could  judge,  dark  green 
porphyritic  rocks,  probably  decomposed  andesites, 
seemed  to  prevail. 

Ardasa  (3148  feet)  lies  about  two  miles  further 
up  the  river;  its  hovels  nestle  at  the  foot  of  a  high 
brown  crag,  crowned  by  the  ruins  of  an  ancient  castle, 
a  thousand  feet  or  more  above  the  river.  Our  road 
now  followed  the  right  bank  of  the  Kharshut,  and 
ascended  nearly  to  its  sources  in  the  Vavuk  Dagh. 
The  deep  valley  is  entirely  hemmed  in  by  high, 
pinnacled  cliffs  of  dark  greenish- grey,  pyroxenic  rocks, 
weathering  a  rusty  brown.  The  Chit-chai  enters 
on  the  left  bank,  its  clear  waters  forming  a  marked 

In  No.  8,  under  the  microscope,  the  most  abundant  crystals  consist 
of  felspar,  chiefly  plagioclase  (andesine  about  Ab4An3),  but  some 
ORTHOCLASE  is  also  present;  both  are  turbid  with  brown,  dusty 
decomposition-products;  the  plagioclase  frequently  contains  chlorite 
inclusions  centrally  and  also  much  calcite ;  occasionally  the  crystals  are 
a  little  rounded  at  comers.  Several  pseudomorphs  after  hornblende 
occur,  outlined  by  limonite,  the  interior  ff  the  crystals  consisting  of  an 
aggregate  of  calcite,  chlorite  and  epidote.  A  moderate  number  of 
MAGNETITE  grains  are  present,  with  much  calcite  and  limonite. 
Interstitial  chlorite  is  even  more  abimdant  than  the  calcite ;  some 
of  the  larger  patches  enclose  plagioclase  crystals  and  microlites,  with 
magnetite  granules,  and  may  perhaps  represent  original  glass.  There 
are  also  some  much  decomposed,  apparently  diabasic  fragments,  not 
very  clearly  defined,  consisting  of  chloritic  and  earthy,  limonitic  ag- 
gregates, often  with  lenticular  outlines,  enclosing  small,  prismatic 
felspars. 

*  Jieise  nach  Persien^  L  172. 


26  Geology  of  Armenia 

contrast  to  the  muddy  brown  Khaishut.  On  looking 
up  this  side- valley,  nearly  horizontal  grey  strata, 
probably  limestones,  were  visible. 

We  soon  passed  into  a  zone  of  pink,  fine-grained 
hornblende-granite  (No.  ii),*  formmg  bold,  rugged 
heights,  barren  and  bleak.  The  dark  grey,  pyroxenic 
rock  re-appeared,  and  was  traversed  by  a  vein  of  the 
pink  granite.  It  gave  way  in  turn  to  a  pale  grey 
limestone  (No.  12J,  dipping  25°  S.;  its  fracfture  is 
splintery,  and  it  is  traversed  by  thin  veins  of  calcite. 
In  fadl  it  is  very  similar  to  the  limestone  I  noticed 
on  the  previous  day  near  Hamsikoi  (p.  21^.  The 
microscope  reveals  a  few  foraminifera,  chiefly  Tex- 

*  The  rock  is  holocrystalline,  composed  chiefly  of  felspars,  both 
pink  [orthodase],  and  greenish-white  and  striated  [plagiodase],  the 
former  rather  more  than  twice  the  amount  of  the  latter ;  but  the 
plagiodase  is  idiomorphic  and  reaches  slightly  larger  dimensions  (up 
to  5  mm.).  Lustrous  hornblende  crystals  (up  to  6  mm.)  are  fairly 
abundant,  as  well  as  smaller  flakes  and  hexagonal  plates  of  chlorite. 
Quartz  granules  (up  to  2  mm.)  are  subordinate,  and  interstitial.  A  thin 
vein  of  epidote  traverses  the  specimen.  The  rock  is  slightly  magnetic 
(5^).  Under  the  microscope,  the  orthoclase  is  very  turbid,  with 
brownish  decomposition-products  and  tiny  flakes  of  muscovite,  with 
some  epidote ;  it  is  allotriomorphic,  and  clearly  later  than  the  plagio- 
CLASS  (ANDESiNE  about  Ab5An3),  which  tends  to  be  idiomorphic,  and 
is  less  decomposed;  however,  it  contains  abundant  epidote,  with 
chlorite  and  calcite.  Green  hornblende  is  fairly  abundant,  but  never 
truly  idiomorphic;  it  is  rather  reedy,  sometimes  twinned,  and  has 
been  occasionally  altered,  in  part,  to  chlorite,  and  often  contains 
epidote  granules.  Chlorite  pseudomorphs  after  biotite  are  slightly 
less  abundant  than  the  hornblende.  Quartz  is  present  in  allotrio- 
moiphic,  interstitial  grains,  sometimes  enclosing  orthoclase  in  an 
irregularly  pegmatitic  manner;  fluid  inclusions,  with  a  bubble,  are 
frequent,  sometimes  in  strings.  A  few  scattered  grains  of  MAGNETrrs 
are  present,  and  leucoxene  occurs  in  irregular  grains,  sometimes  still 
with  a  core  of  ilmenite.  Zircon  and  apatite  are  accessory.  Epi- 
dote is  present  in  the  form  of  a  vein,  as  well  as  enclosed  in  felspar, 
hornblende  and  chlorite.  The  order  of  crystallization  seems  to  hiave 
been  :  i.  Apatite,  zircon  and  magnetite ;  a.  Biotite ;  3.  Hornblende ; 
4.  Plagiodase. ;  5.  Orthoclase  and  quartz. 


Trebizond  to  the  Vavuk  Pass  27 

tularia^  a  Miliolina  indistinguishable  in  secftion  from 
M.  trigonula^  Lam.  of  the  Paris  basin,  and  Globigerina. 
It  weathers  into  rusty-brown,  picturesque  crags.  This 
zone  of  limestone  (presumably  Eocene)  becomes  in- 
truded by  a  greenish  grey  homblende-pyroxene-por- 
phyrite  (No.  13).*  Indeed,  the  limestone  has  been 
repeatedly  intruded  by  this  porphyrite  as  far  as  the 
Kharava,  a  clear  stream  coming  from  the  north. 
Here  a  breccia  (No.  14)  of  dark  red  limestone  frag- 
ments, in  a  cement  01  calcite,  underlies  the  grey 
limestones,  which  still  dip  to  S.S.E.  After  passing 
over  some  marshy  ground,  we  crossed  to  the  left  bank 
of  the  Kharshut,  still  within  the  limestone  region. 
However,  at  a  little  distance  below  this  bridge,  I 
noticed  a  clear  instance  of  an  intrusive  sill  of  reddish 
rock,  about  15  feet  thick,  causing  alteration  of  the 
sedimentary  strata,  both  above  and  below.    A  series 

*  The  rock  is  iron-grey  and  compact,  with  numerous  white,  tabular 
felspars  (up  to  $  mm.),  and  dull  greenish  black  crystals  (up  to  6  mm.), 
with  the  characteristic  outlines  of  hornblende.  The  rock  is  calcareous 
in  places.  Under  the  microscope,  plagioclasb  (andesine  about 
AbjAna)  occurs  in  large,  tabular  idiomorphs,  often  zoned  j  albite 
twinning  associated  with  Carlsbad,  rarely  with  pericline;  rather  free 
from  indusions,  except  for  an  occasional  granule  of  magnetite,  but  the 
cxystals  are  much  cracked,  with  dusty  calcite  and  chlorite  along  the 
cracks.  Pseudomorphs  after  hornblende  are  numerous,  down  to 
quite  small  dimensions ;  it  has  been  entirely  altered  into  a  mixture  of 
pale  green  chlorite  and  abundant  magnetite  granules — the  latter  being 
more  closely  congregated  along  the  margins — together  with  subsidiary 
epidote  and  calcite.  A  few  idiomorphs  of  very  pale  green  augitb 
are  present,  partly  replaced  by  calcite  and  chlorite.  An  occasional 
pseudomorph  of  clear  chlorite  free  from  magnetite  seems  to  represent 
a  wmmmm  pyroxene.  A  few  large  grains  of  magnetite  are  scattered 
through  the  slide,  and  apatite  is  also  accessory  The  groundmass  is 
a  felt  of  orthoclase  in  stumpy  prisms,  and  of  oligoclase  in  laths 
and  microlites,  with  subordinate  magnetite  granules  and  chlorite  (after 
augite),  and  also  abundant  interstitial  quartz,  which  also  occurs  in 
larger  secondary  grains,  associated  with  spherulicic  chlorite,  in  original 
cavities.     Calcite  is  generally  distributed  in  the  groundmass. 


28  Geology  of  Armenia 

of  shales  now  came  to  light,  overlying  the  limestones 
and  sometimes  reddened  by  intrusive  rocks.  The 
softer  nature  of  the  strata  causes  a  widening  of  the 
valley  at  Besh  Kilissa,  with  a  remarkably  sudden 
change  from  rugged,  barren  rocks  to  orchards  and 
smiling  meadows.  The  valley  soon  narrowed  again 
owing  to  the  re- appearance  of  hard,  dark  green 
porphyritic  rocks,  still  intrusive  into  limestones,  which 
became  vertical,  and  then  dipped  steeply  to  S.E.  But 
the  porphy rites  soon  prevailed  on  both  sides  of  the 
valley.  I  obtained  two  specimens :  the  first  (No.  15)  * 
is  a  brecciated  homblende-porphyrite,  collected  at  a 
spot  distant  about  three  quarters  of  an  hour's  drive 
from   Besh  Kilissa;  and  the  second  specimen   is  a 

*  It  is  a  compact,  greenish  grey  rock,  blotched  with  purple  and 
containing  numerous  white  felspars  (up  to  is  mm.),  lustrous  horn- 
blende crystals  (up  to  a  mm.),  and  still  smaller  specks  of  magnetite. 
Under  the  microscope,  plagioclasb  (andesinb  about  AbsAnj)  occurs 
in  idiomorphs  mostly  of  prismatic  habit,  sometimes  in  groups ;  often 
zoned,  and  much  cracked,  with  dusty  decomposition-products  along 
the  cracks.  A  little  subordinate  orthoclasx  is  also  present.  Horn- 
blende is  present  in  frequent  idiomorphs,  green  in  colour,  sometimes 
with  small  patches  stained  an  orange-brown;  pleochroism  strong: 
a — ^pale  greenish  yellow,  A— dull  sap-green  with  a  brownish  tinge, 
r— dark  green  ;  sometimes  twinned ;  it  has  suffered  a  variable  amoimt 
of  alteration,  often  showing  a  thick  resorbtion-border  [hematitic]  which 
occasionally  encroaches  on  the  entire  crystal,  or  else  it  has  been 
replaced  by  a  granular  aggregate  of  chalcedony  with  a  core  of  chlorite. 
AuGiTE  in  pale  green  idiomorphs  is  less  frequent  than  the  hornblende, 
and  usually  in  smaller  individuals;  it  is  quite  fresh;  occasionally 
twinned,  sometimes  repeatedly;  and  it  encloses  magnetite  grains. 
Magnetite'  and  apatite  [sometimes  pinkish]  are  accessory.  The 
GROUNDMASS  is  fine-grained,  and  shows  a  brecciated  structure,  com- 
posed of  brownish,  angular  areas  in  a  clear  matrix,  which  only  differs 
in  the  fact  that  the  former  contain  nematitic  dust  and  more  magnetite 
and  enclose  the  more  strongly  resorbed  hornblendes.  The  ground- 
mass  in  both  cases  is  a  felt  of  tiny  plagioclase-felspar  laths,  showing 
fluidal  arrangement,  with  abundant  magnetite  granules,  and  more 
sparing  augite  granules  ;  sometimes  in  a  quartzose  (obscurely  micro- 
poecilitic)  matrix,  at  other  times  with  interstitial  colourless  glass. 


TrMxmuL  to  the  Vmfmk  Pass  ag 

pyroxene-porphyrite  (No.  16),*  taken  a  quarter  of  an 
oour  further  on. 

Gumiishkhaneh,  famous  for  its  silver  mines,  soon 
came  into  view.  It  is  sit^iated  halfway  up  barren 
limestone  mountains,  which  recede  gradually  from  the 
Khaishut  river  and  rise  up  into  pinnacled  ridges, 
CTowned  by  tiie  ruined  towers  of  the  mediaeval  castle 
khutda  Oglu,  about  three  or  four  miles  east  of 
Gumiishkhaneh  (16^  miles  from  Ardasa).  As  we 
proceeded,  it  became  apparent  that  these  limestones 
have  been  intruded  on  a  grand  scale  by  a  pink  biotite 
granite,  which  decomposes  into  red  and  orange  sands. 
The  limestone,  which  at  first  dipped  to  S.E.,  soon 
formed  a  syncline,  coming  down  to  the  level  of  the 
river,  and  then  changing  its  dip  to  20°  N,N.W.  The 
granite  again  rose  to  view  from  beneath  the  limestone 
in  which  a  contaA-zone  is  clearly  visible,  a  purple 
band  next  to  the  granite  passing  into  a  red-brown. 

^  The  rock  is  dark  grey,  in  places  with  a  purplish  tinge,  compact, 
and  with  irregular  fracture.     It  contams  abundant  striated  felspars  (up 
to  2  mm.) ;  with  the  lens,  still  smaUer  green,  chloride  crystals  (less 
than  a  mm.)  can  be  seen  on  the  cut  surface.     Und;:r  the  microscope, 
the  PLAGiocLASE  (lab'radorite  4bout  Ab5An6)  occurs  in  large,  mostly 
prismatic  idioDioiphs,  occasionally  intergrown ;  fairly  cleir,  but  often 
with  a  peripheral  zone  of  dust ;  boning  infrequent ;  it  encloses  augite, 
and  also  ddorite  and  ealdte..    A  fair  amount  of  augite  in  very 
pal;;  green  idiomorphs  is  presient,   quite  fresh,   sometimes  showing 
lamelkr  twiiming;  encloses  apatite  and  magnetite..   Pseudomorphs 
after  a  mmmmm  pyroxene  are  earlier  and  more  numerous,  down 
to  quite  small  dimensions;  the  original  idiomorphs  have  been  entirely 
altered   into  bhl6rite,   fibrous  and  yellow-green  along  the  margins 
and  ctoss'pardi^s,  blue-green  and  micrbspherulitic  internally;  in  some 
cases  further  alteration  has  resulted  in  the  formition  of  dusty  ealdte 
and  hematite  along  the'  outlines  ahd  cracks.     Magnetite  is  present  in 
rather  large  grains.     Apatttb  js  acces^ry.*    The  grpundmass  is  a 
pale  brown,  diist^  and  obscurely  micropoedlitic  felt  of  felspar  laths 
and  microlites  (oligodase),  augite  granules  (sometimes  stained  by  iron 
oxide),   and  magnetite  m  granules  and  dust    Calcite  occurs  in 
patches  and  filling  cracks. 


30  Geology  of  Armenia 

Denudation  has  proceeded  more  rapidly  with  the 
limestone  than  with  the  gr^iite,  which  has  formed 
rounded,  pyramidal  hills,  cloaked  in  orange  sands. 

On  leaving  this  granitic  zone  we  emerged  into  open 
undulating  country  near  Tekke,  finely  situated  on 
a  hill  to  the  north  of  the  Kharsbut.  The  rather  tame 
scenery  is  due  to  the  outcrop  of  calcareous  sandstones 
(showing  current  bedding),  overlaid  by  dark  grey 
shales,  dipping  to  the  south  and  soon  forming  a 
gentle  syncline, 

Just  before  a  road  branches  off  to  Erzingan  a  thick 
vertical  dike  of  dark  grey  rock  cuts  through  the 
shales.  A  little  further  on,  a  black  rock  exhibits  a 
spheroidal  structure  by  weathering  in  concentric  coats 
like  an  onion ;  it  is  probably  of  a  doleritic  nature. 
The  valley  is  soon  constricted  again  by  limestones, 
rising  up  in  lofty  heights.  It  was  evidently  close 
to  this  spot  that  Hamilton*  found  "a  fragment  of 
a  large  Ammonite.**  This  renders  it  probable  that 
these  limestones  are  at  any  rate  of  Cretaceous  age, 
if  not  older.  We  soon  emerged  from  the  rocky 
gorge  into  the  small  alluvial  plain  of  Murad  Khan 
(4430  feet),  our  resting-place  for  the  night.  Here 
we  were  surrounded  by  high  cliffs  of  unfossiliferous 
limestone,  weathering  a  rusty  brown,  and  honey- 
combed by  numerous  caves. 

After  leaving  Murad  Khan,  we  had  no  sooner 
passed  through  the  limestone  barrier  than  shales 
re-appeared  for  a  short  space,  forming  open  tmdulating 
country,  which,  however,  is  bounded  on  the  east  by 
a  precipitous  mass  of  limestone.  The  jagged  edges 
of  the  vertical  strata  stand  silhouetted  against  the 
sky,  like  the  teeth  of  a  saw,  and  are  crowned  by 
the  ruins  of  a  mediaeval  castle. 

*  jReseanches  in  Asia  Minor ^  L  170. 


TrebtMond  tb  the  VOimk  Pass  31 

The  eastern  side  of  Kalajik  ot  Geniskaleh  (Genoese 
castle)  is  not  so  precipitous  &&  the  western,  but  the 
rugged  slope  has  oeen  adiply  fortified  to  make  good 
t!he  deficienaes  ol  nature  on  this  side.  After  passing 
througli  nearly  vertical  shales  reddened  by  intrusive 
rocks,  the  Kharshut  valley  is  nartowed  a^ain  owing 
to  a  zone  of  hard  igneous  rocks ;  a  specimen  taken 
near  the  junction  of  the  shales  is  an  altered  brc^izite- 
porphyritfe  (No.  17^*  It  weathers  deeply  along  its 
fiiassive  joints  into  fantastic,  wcolpack  forms^ 

Small  outcrops  of  shale  were  still  met  with  here  Atid 
there  for  a  short  time,  but  the  igneous  rock  soon 

Srevailed  entirely,   frequently   traversed   by   vertical 
ikes.     Here  the  rock  became  very  dark  green  and 

^  The  rock  is  a  dull  brownish  crimson,  compact)  with  abutldant 
white  felspars  (up  to  3  mm.},  a  few  small  gteen  chlorites  (up  to  a  mM.) 
and  some  glassy,  pale  bluish  opal  in  small  irregular  masses.  Utidet 
the  microscope,  the  plagioclasi  (labradorite  about  Abs An6)  occurs 
in  large  tabular  idiomorphs^  often  in  groups ;  usually  fairly  ftesh, 
but  cracked  and  sometimes  with  opal  or  a  zeolite  &long  the  cis^cksj 
toning  is  frequent;  it  sometimes  contains  chlorite  granules  or  little 
oblong  inclusions. of  dusty  brown  glass.  Pseudomorphs  after  bronzite 
are  fairly  numerous;  sharp  ciystal-outlines  are  present,  but  mostly 
a  little  rounded  at  the  angles ;  usually  in  groups ;  the  outlines  and 
cross-partings  are  marked  out  by  hematite,  the  interior  beng  occupied 
by  opal,  often  enclosing  remnants  of  a  green,  fibrous,  basdte-likfc 
mineral ;  a  core  of  still  unaltered  bronzite  occurs  in  one  case  in 
the  slide:  it  is  4  tHrih,  a  Very  pale  green,  slightly  pleochroic  [to 
pale  yellow] ;  in  another  case,  some  plagioclase  is  enclosed  in  a  group 
6f  these  pseudottiotphs.  Magnetite  grains  are  fairly  numerous, 
and  are  especially  associated  with  the  pseudotnorphs.  The  ground- 
ICASS  is  a  somewhat  orthophyric  felt  of  plagioclase  laths  [andesine 
about  AbjAna],  with  stumpy  prisms  of  orthoclase,  somewhat  less 
in  number,  abundant  magnetite  granules,  ahd  rather  sparing  pale 
green  augite  granules,  frequeiitly  stained  with  iron  oxide, — indeed,  the 
Whole  rock  has  been,  so  to  speak,  saturated  with  hematite,  much 
obscuring  the  structure.  Small  cavities  have  been  filled  with 
secondary  products  :-^a  narrow  lining  of  a  zeolite  [waterclear,  and 
polarizing  in  interlocking,  crystalline  granules],  an  inner  layer  of  opal, 
ind  finally  a  core  of  chlorite. 


32  Geology  of  Armenia 

more   compadl,    with    conchoidal   fracfture;   it    is   a 
pyroxene-porphyritc  (No.  i8),*  somewhat  altered. 

At  last  we  left  the  Kharshut  river,  and  made  a  long 
and  gradual  ascent,  up  the  valley  of  a  left  tributary, 
to  the  Little  Vavuk  pass.  The  contours  of  the  valley 
are  gently  rounded,  and  consist  of  an  andesitic  ag- 
glomerate (No.    ig),!  which  has  experienced  not  a 

*  The  rock  has  a  dull,  pitchy  lustre,  and  contains  numerous  striated 
felspars  ^up  to  6  mm.,  often  with  a  dark  green  core),  dark  green,  silky 
crystals  (up  to  a  mm.,  showing  prismatic  and  octagonal  outlines),  and 
small  specks  of  magnetite.  The  rock  is  slightly  magnetic  (a°),  and 
rather  calcareous.  Under  the  microscope,  the  plagioclase  (an  add 
LABRADORITE  about  AbiAui)  is  in  numerous  large,  tabular  idiomorphs 
much  traversed  by  cracks,  along  which  chlorite  and  much  caldte  have 
formed;  the  core  is  frequently  full  of  caldte  and  chlorite  while  the 
more  acid  periphery  is  clear  and  fresh.  There  are  numerous  pseudo- 
morphs  after  a  wmmmm  pyroxene,  in  well -formed  crystals,  composed 
entirely  of  bluish  green  chlorite ;  fibrous  along  the  margins  and  cross- 
partings;  sometimes  further  altered  into  caldte  along  the  cracks. 
A  few  pseudomorphs,  composed  of  dusty  caldte  with  a  little  chlorite, 
have  the  characteristic  forms  of  augite.  Scattered  crystals  of  mag- 
netite are  present.  Apatite,  pinkish  brown  and  slightly  pleochroic, 
is  accessory.  The  groundmass  is  an  obscurely  micropcedlitic  felt  of 
crowded  felspar  laths  and  microlites  (the  larger^  are  andesine,  often 
with  forked  ends,  the  smaller  are  oligoclase),  with*  extremely  abundant 
magnetite  in  dust  and  granules.  Calcite  occurs  in  frequent  patches 
throughout  the  slide,  and  there  has  been  much  staining  with  iron 
oxide. 

t  The  rock  contains  irregular  dark  green  fragments  (up  to  9  mm.), 
and  endosing  pink  felspars  (up  to  2  mm.),  and  smaller,  pale  brownish, 
quartzose  areas  in  a  paler  green  matrix;  specks  of  iron  pyrites  and 
magnetite,  and  some  dark  green  chloritic  crystals  (up  to  a  mm.)  can 
just  be  seen  with  the  lens.  Under  the  microscope,  the  numerous 
angular  fragments  of  lava  are  either  cryptocrystalline,  with  perlitic 
markings,  or  else  are  andesitic  microfelts,  varying  in  degree  of  texture, 
consisting  of  oligoclase  microlites  and  magnetite  granules,  with  inter- 
stitial chloritic  matter,  phenocrysts  of  andesine  idiomorphs  (much 
obscured  by  caldte  and  dilorite),  pseudomorphs  of  opal  after  bronzite 
with  occasional  remnants  of  the  original  mineral,  and  finally  pseudo- 
morphs after  augite,  consisting  either  wholly  of  dilorite,  or  of  caldte, 
or  of  both  minerals.     The  groundmass  is  a  confused  aggregate  of 


Trcbizond  to  the  Vavuk  Pass  33 

little  alteration;  it  varies  a  good  deal  in  texture, 
sometimes  becoming  a  rather  coarse  breccia.  A  small 
outcrop  of  columnar,  basaltic  rock  was  noticeable 
on  the  left  bank  as  we  proceeded  up  the  valley,  but 
this  soon  gave  place  to  a  dark  grey,  impure  limestone 
(No.  20),  forming  high,  roimded  downs.  Exposures 
seldom  occurred,  and  I  could  find  no  traces  of  fossils ; 
even  the  microscope  revealed  no  organisms.  Wagner* 
in  1852  has  recorded  ao  equal  want  of  success  after 
a  search  of  some  hours.  However,  he  did  not  hesitate 
to  consider  this  limestone  to  be  of  Cretaceous  age. 
This  is  indeed  very  probable,  because  the  Vavuk  Dagh 
forms  the  broad,  western  marginal  rim  surrounding 
and  underlying  the  Eocene  plateau-basin  of  Varzahan ; 
and  at  Baiburt,  on  the  eastern  rim  of  this  plain, 
similar  limestones  have  been  established  by  Abich  t  to 
be  Turonian  in  age. 

cUorite,  caldte,  numerous  patches  of  a  dusky,  pale  brownish  zeoUtie 
(with  a  dark  border,  and  nulially  fibrous  structure),  opal,  groups  of 
secondary  quarts  grains,  and  gnuiules  of  iron  pyrites  and  magnetite. 

*  i?m^  nort  Armrn^  i.  192. 

t  Giologtsdu  Forschtmgm  in  dm  hmkamchai  Landem^  ii.  1x3— »Z20. 


I 


CHAPTER  IV 

THB  VAVUK  PASS  TO  ERZERUM 

The  view  from  the  summit  of  the  Little  Vavuk  pass 
(6450  feet;  loi  miles  from  Murad  Khan)  recalls  the 
gentle  charadleristics  of  our  Chalk  downs,  but  on 
a  far  grander  scale.  High  rolling  downs,  alternating 
with  wide  plains,  form  the  dominant  features  of 
the  Armenian  plateau-country,  cradled  between  the 
jagged  peaks  of  the  Pontic  Range  on  the  north,  and 
those  of  the  Tauric  Alps  on  the  south.  Perhaps 
the  chief  criterion  of  this  plateau-region  (possessing 
an  average  elevation  of  5000  feet)  is  to  be  found  in 
the  character  of  the  rivers.  The  larger  streams,  such 
as  the  Frat  and  Murad,  flow  with  a  full,  sluggish 
current,  often  meandering  slowly  over  extensive, 
alluvial  flats, — a  strong  contrast  to  the  violent  and 
impetuous  torrents  of  the  border-ranges. 

Grey  shales  underlie  the  limestones  of  the  Vavuk 
Range,  and  wherever  they  crop  out  they  give  rise 
to  lower,  undulating  ground,  marshy  and  golden  with 
buttercups,  while  the  turf  of  the  high  limestone  downs 
was  parched  and  brown  (June  10).  A  marked  feature 
of  this  limestone  zone  is  the  number  of  narrow  escarp- 
ments, occurring  at  intervals  one  above  the  other, 
running  continuously  for  miles  over  the  hills,  and 
looking   deceptively   like  human   masonry.    Several 


The  Vavuh  Pass  to  BrMerum  55 

folds  were  noticeable  as  we  drove  along,  the  dip 
not  exceading  25""  to  N.N.W,  and  S,S.E.  respeAively, 
At  Khadrak  (88th  milestone,  4^  miles  from  the  Vavuk 
pass)  the  folds  are  somewhat  more  numerous,  but  still 
rather  shallow  and  symmetrical.  The  lowe^  slopes 
here  consist  of  rounded  hills,  wilh  a  reddish  brown 
soil ;  it  soon  became  evident  that  this  colour  is  due  to 
the  decomposition  of  a  pink  microgranite,  intrusive  in 
the  limestones,  and  apparently  very  similar  to  the 
granite  I  observed  near  Giimushkhaneh  (p.  29). 

The  river,  which  we  were  following,  flows  gently 
and  placidly  through  its  alluvial  flats ;  the  limestone 
downs^  with  the  underlying  granite,  gradually  retreats 
away  to  the  N.E.,  but  sends  down  an  occasional  spur 
towards  the  road,  with  strata  dipping  10®  S.S.E.  The 
alluvial  plain  increases  in  width  to  over  a  mile,  owing 
to  the  river  joining  the  main  stream  at  Balakhor. 
Below  this  village  a  considerable  ridge  of  the  barren 
limestone  comes  down  from  the  rugged,  snowy  heights 
of  the  Kitowa  Dagh,  and  the  folds  of  the  strata  are 
clearly  marked  on  the  bare,  rounded  hillside.  The 
Osluk  river  cuts  through  this  barrier  by  a  narrow 
gorge,,  by  which  we  entered  the  extensive  Eocene 
plain  of  Varzahan  (5400  feet).  At  Osluk  Khan  (8 
miles  from  Khadrak),  where  we  made  our  midday 
halt,  the  walls  are  mostly  built  of  a  white,  crystalline 
limestone  derived  from  the  above-mentioned  barrier, 
but  I  noticed  a  couple  of  blocks  of  a  dark  brown 
conglomerate,  containing  pebbles  of  serpentine  (up  to 
2  cm.),  and  abundant  small  Nummulites  (6 — 10  cm.  in 
diameter).  It  was  unfortunately  impossible  to  take  a 
specimen,  and  I  ocnild  not  see  this  rock  anywhere  in 
utu.  My  subsequent  discovery  (see  Chapter  XIV) 
ct  Nummulitic  Limestone  at  Kerzi  (6  miles  N.E.  of 
Osluk  Khan),  on  the  northern  border  of  the  plain, 
tenders  it  probable  that  these  blocks  <A  Nummulitic 


36  ,  Geqlogy  of  Armenia 

conglomerate  had  not  been  brought  from  any!  great 
distance.  The  road  soon  leaves  the  riverj  and  passes 
eastward  over  low  hillocks,  composed  of  gteenish  And 
grey  shales,  dipping  at  a  low  angle  to  S.E. 

At  Vgirzahan  (5620  feet ;  6  miles  from  Osluk  Khan) 
we  stopped  to  examine  the  three  rui^-*d  Armenian 
churches,  surrounded^  by  tombstones,  many  of  which 
show  the  ram  type.  Although  these  buildings  date 
back  to  the  twelfth. century,  if  not  earlier,  yet,  o^^'ing 
to  the  extremely  diy  climate,  the  calcareous  sandstone 
of  which  they  are  built  is  still  fresh  and  yellow, 
w;ithout  the  slightest  covering  of  moss  or  lichen  to 
emphasize  their  antiquity.  The  keystones  ever  the  • 
doorway  of  Surp  Khach  consist  of  a  pale  green  sand- 
stone, which  derives  its  colour  frcm  serpentincus  ^ 
elements.  ^ 

After  leaving  Varzahan,  the  low,  undulating  downs 
of  the  great  plain  form  slightly  higher  ground,  and 
consist  of  yellow,  calcareous  sandstones.  Shortly 
before  arriving  at  Baiburt  (5075  feet;  6  miles  from 
Varzahan)  the  road-cuttings  reveal  a  small  anticline 
with  a  dip  of  20°  both  to  N.W.  and  S.E.  The  narrow 
esc?Lrpments  also  clearly  show  the  N.E. — S.W.  strike.  1 

The  last  stage  (Baiburt  to  Maden  Khan)  of  our  day's 
journey  of  45  miles  w^  traversed  in  total  darkness, 
and  it  was  only  on  our  return  to  the  coast  (Chapter 
XlV)  that  I  was  able  to  take  observations  of  this  part 
of  the  route. 

Maden  Khan  (5^34  feet;  lof  miles  from  Baiburt)  is 
situated  on  the  right  bank  of  the  Chorokh  valley, 
which  is  still  fairly  wide  but  shut  in.  by  high,  rounded 
downs  of  a  barren,  grey,  unfossiliferous  limestone, 
similar  to  that  at  Baiburt,  and  probaWv  Crietacebus 
in  age.  It  is  clear  that  the  valleyirtiQS  along  .the  axis 
of  an  anticline,  the  strata  having  a;  southerly  dip  of 
20^  on  the  south  bank  and  a  sinaildir  j3^i::^ination  to  the 


M'^    TO 


si. 


L^) 


f 


1^ 


"iaStXhor- 


ICfH 


^*/ 


Toy 


KIT 


fcxncK^ 


Sg 


-'H 


r^ 


^^ 


'(f^ 


k 


Mi<-^ 


TOWA        ^ACrH 


'C^CHT^L^     c£e^    Jttvyu^^ 


^^^Al,  y^ 


{{  L*landl  Siauford,  Jr.'; 


The  Vavuk  Piiss  to  Erzerum  37 

north  on  the  opposite  bank.  After  crossing  the 
wooden  bridjg^e  on  the  way  to  the  Kop  Dagh,  the  road 
passes  over  horizontal  strata,  which  reach  a  level  oiF 
about  100  feet  above  the  river.  The  beds  are  greyish 
brown;  a  harder,  conglomeratic  layer  is  less  easily 
weathered,  and  forms  an  overhanging  cornice ;  these 
horizontal  beds  are  probably  travertine-deposits  (see 
Chapter  XIV).  Two  terraces  could  be  distinguished, 
one  about  30  feet  above  the  other. 

We  soon  left  the  Chorokh,  and  drove  up  the  valley 
of  the  Kop  Su,  leaving  on  our  right  an  intrusive 
boss  as  we  entered  the  passage  through  the  barren 
limestones  of  the  Chorokh.  In  this  gorge  I  noted 
a  steep  and  symmetrical  syncline,  dipping  65^  first 
to  S.S.E.,  and  then  to  N.N.W.  After  crossing  over 
to  the  left  bank,  we  emerged  into  open,  undulating 
country;  the  limestone  assumed  a  cavernous,  some- 
what travertinous  appearance,  but  no  fossils  were 
visible.  This  upland  plain,  with  low  hills,  owes  its 
gentle  contours  to  the  outcrop  of  soft  shales,  with 
numerous,  thin  seams  of  limestone,  just  as  in  the 
Vavuk  region.  My  observations  on  our  return  journey 
(Chapter  XIV)  on  this  series  of  beds  at  Akhviren, 
16  miles  further  east,  show  that  they  are  Neocomian 
(Hauterivian)  in  age.  In  one  place  a  greenish  sand- 
stone (No.  21)  is  intercalated  in  the  series  of  shaly 
strata ;  it  contains  carbonized,  reed- like  plant  remains 
which,  unfortunately  are  not  well  preserved ;  they  are 
probably  fucoidal  in  nature.  This  sandstone  indeed 
is  very  similar  to  the  Flysch  facies,  which  is  presented 
by  the  Cretaceous  in  several  parts  of  Armenia,  e.g.  on 
tifie  northern  slopes  of  the  Aghri  Dagh,  near  Eailpi, 
where  Abich  found  an  Inoceramus,  sp.  indet.,  with 
numerous  f ucoids.* 

Thei  shales  soon  re-appeared,  forming  a  sharp  anti- 

*  Anthula,  Krddefossmm  i,  JStmk.^  Bdtr.  Pal.  Oest.-Ung.  xii.  430. 


38  Geology  of  Armenia 

cline,  then  becoming  vertical,  and  finally  dipping 
S.S.E.  When  the  inclination  had  been  reduced  to 
20°  S.S.E.  they  became  overlaid  unconformably  by 
horizontal,  mud-brown  strata  of  either  a  travertine,  or 
an  old  river  deposit,  but  the  beds  are  how  about  30 
feet  above  the  present  level  of  the  Kop  Su.  The 
valley  here  narrows  considerably,  becoming  a  rocky 
defile,  shut  in  by  high  bare  cliffs  of  a  limestone,  which 
forms  a  syncline,  dipping  first  to  S.S.E.,  and  then 
20°  to  N.N.W. ;  it  overlies  the  shaly  strata.  This 
limestone  is  not  massive  and  marmoraceous  like  that 
of  the  Chorokh  zone,  but  is  concretionary,  and 
weathers  to  a  warm,  rusty  brown,  colour.  It  is, 
however,  equally  unfossiliferous.  When  fradlured,  it 
is  bluish  grey,  with  white  veins  of  calcite.  Further 
on,  it  becomes  much  contorted  and  puckered  to  a 
most  extraordinary  extent.  The  ,  shales  with  lime- 
stone seams  rose  up  again  to  view  from  beneath 
the  concretionary  limestone,  and  displayed  even  more 
acute  contortions. 

Before  we.  reached  the  Northern  Kop  Khan  (6358 
feet;  6^  miles  from  Maden  Khan),  the  horizontal, 
muddy-brown  strata  (perhaps  travertinous),  which  I 
had  already  noticed  twice  on  this  stage,  are  strikingly 
exposed  in  a  large  block  isolated  by  the  river,  with 
cliffs  about  100  feet  high.  Its  base  is  encumbered 
by  the  fallen  debris  of  the  harder  conglomeratic 
beds,  which  form  overhanging  eaves,  the  nesting- 
places  of  innumerable  jackdaws.  About  two  miles 
further  on,  the  much  contorted  and  crushed  shaly 
strata  are  again  overlaid  by  a  limestone  f weathering. a 
yellow-brown),  which  shows  less  sign  ot  disturbance, 
although  exhibiting  several  folds  and  even  becoming 
vertical  in  places.  The  increasing  gradient  of  the 
road  enabled  me  to  observe  that  this  limestone 
becomes  more  massive  and  whiter  in  its  upper  part. 


:# 


The  Vavuk  Pass  to  Erzerum  39 

To  the  west  of  our  road,  however,  the  country  fell 
away  in  gentle  contours,  with  green,  rolling  slopes 
often  marshy  and  golden  with  buttercups, — marking 
the  continuance  (rf-4he  shales  (with  E.N.E. — W.S.W. 
strike)  in  this  direction.  There  is  in  fadl  a  con- 
siderable resemblance  to  the  landscape  of  the  Vavuk 
pass.  During  the  steep  and  laborious  zigzag  of  the 
final  ascent  to  the  ICop  pass,  I  examined  a  long 
stretch  of  the  dazzling,  white,  barren  limestone  (No. 
22),  but  the  only  distindt  fossil  which  I  could  find  was 
a  fragment  of  a  Belemnite;  Mr.  G.  C.  Crick,  of  the 
British  Museum,  has  kindly  examined  it  and  Considers 
it  to  be  Duvalia  cf.  polygonalis^  Blainville.  Some 
obscure  silicified  sponges,  and  an  indeterminable 
sedlion  in  iron  oxide  of  a  Gastropod  shell  were  as- 
sociated with  this  fossil,  which  indicates  the  Upper 
Neocomian  (Hauterivian)  age  of  the  Kop  limestone. 
It  is  therefore  evident  that  Abich's  statements  that 
"indistincft  and  very  sparing  traces  of  fossils  point 
to  the  limestone  of  the  Kop  Dagh  and  the  Khoshab- 
punar  system  being  Hippuntic  Limestone,*  "  and  that 
"  the  limestone  of  Baiburt  and  the  Khoshabpunar  pkss 
belongs  to  the  Upper  Rudistes  zone,!  "  can  no  longer 
be  considered  accurate. 

A  specimen  of  the  Kop  limestone,  gathered  by 
Mr.  H.  F.  B.  Lynch  on  his  previous  journey,  revealed 
abundant  Radiolarian  remains  on  microscopic  ex- 
amination by  Professor  T.  G.  Bonney,  F.R.S.  During 
our  ascent  I  noticed  several  thick  beds  of  chert,  which 
have  probably  derived  their  silica  from  these  or- 
ganisms as  well  as  from  siliceous  sponges. 

The  Kop  Dagh  lies  to  the  east  of  the  pass;  its 
summit    is   broad    and   flat,    very   charadleristic   in 

*  Geol.  Farsch.  in  den  kauk.  Ldndem^  ii.  Z09. 

t  Ibid.  ii.  lao ;  and  see  Chapter  XK  on  the  Cretaceous  of  Armenia. 


43  Geology  of  Armenia 

meandering  streams.  To  the  west,  the  lower  ground 
Is  composed  of  crimson  maris  and  sandstoites.  Before 
reaching  Ashkala,  the  road  passes  over  an  old  river^ 
terrace,  about  30  feet  above  the  present  alluvial  plain 
of  the  Frat. 

The  south  bank  of  the  river,  opposite  to  Ashkala 
(5500  feet;  10  miles  from  Pimakapan),  is  formed 
by  a  high  cliff  of  massive,  buff  strata,  probably  sandy 
limestones,  dipping  about  20**  W.N.W.  On  looking 
towards  the  Jejen  peak  (N.5°W.  of  Ashkala),  I  noticed 
a  hill  of  similar  strata,  dipping  in  the  opposite 
direction ;  probably  the  axis  of  the  broad  syncline  lies 
a  little  to  the  north  of  Pimakapan,  for  ix  W.N.W. 
dip  is  clearly  evident  in  the  Miocene  beds  on  bdth 
banks  of  the  Frat.  The  houses  at  Ashkala.  are  built 
of  a  white,  chalky  limestone ;  in  one  of  the  blocks 
I  noticed  a  small  Pe6ten.  Doubtless  the  stone  had 
been  brought  from  lio  great  distance,  and  indeed 
the  limestone  hill  of  Kardarich  (Kagdarich),  so  rich  in 
Lx)wer  Miocene  fossils  (see  Chapter  aXIIIX  was  clearly 
visible  in  the  east. 

On  continuing  our  drive  to  Erzerum,  we  ascended 
to  the  old  alluvial  plain,  about  two  miles  in  breadth. 
Both  to  north  and  south  of  the  plain  light  brown 
clays  or  marls  rise  up,  dipping  about  30  W.N.W;, 
with  occasional,  thick  seams  of  a  buff,  conglomeratic 
limestone.  These  strata  are  very  conspicuous  up  to 
the  wooden  bridge,  where  the  Ashkala  plain  suddenly 
narrows.  Here  the  limestone  predominates,  dipping 
30°  W.N.W.  conformably  beneath  the  clays;  and 
on  the  north  side  of  the  bridge  (by  which  we  crossed, 
to  the  left  bank  of  the  Frat)  the  buff  limestone  occurs, 
exclusively,  in  thick  beds.  Little  Kardarich  lies  at 
the  southern  base  of  a  conspicuous  and  imposing  hill 
of  this  limestone.  Unfortunately  it  was  not  possible 
for  me  to  turn  aside  to  visit  this  excellent  locality  for 


v/yy.v/ 


«.*r 


MW. 


JCo'Tt^L^nich 


SunxJ^uJuLoyv  oftfuL   Ma/^voi^'>Y  at  ^(cudoKccK^, 


£.S.£. 


W.VH( 


?*0«Mfil*J^  Ar  f^'yfxp. 


Til 


MA9eN  KHATI 

4 


^^toVurHvt 


'         a-nttcli"nal  a^s 


ir 


rHAT  VAULCY 


lo'rA 


J 


40*W 


\ 


C-HA* 


■■^i 


Aorti 


MA9EN  KMM« 


amti'clt'nal  oju's 


^ 


1 


i^lMY 


rtd  fnarl$ 


\o'H 


I 


{ 


v.- 


V) 


:f 


^   ^fi 


r\- 


The  Vavtik  Pas$  to  Erzerum  43 

Lower  Miocene  fossils,  which  Abich*^  studied  and 
described.  The  hillside  on  our  right  (i.e.  on  the  left 
bank)  was  covered  with  large  blocks  of  limestone 
from  the  high  escarpments,  which  show  a  dip  of 
about  70°  E.S.E. ;  hence  the  river  cuts  through  an 
anticline  at  this  spot,  our  road  lying  a  little  to  the 
south  of  its  axis. 

The  Frat  valley  widens  out  again  above  its  con- 
striAion  at  Kardaricii.  Along  its  centre  a  low  line  of 
hills  extends  on  the  north  side  of  the  river,  which 
has  cut  a  deep  channel  for  itself  here.  Massive 
alabaster  cropped  out  on  our  right,  and  on  the  left 
side  of  the  Jinis  tributary  I  again  observed  the  red 
marls,  which  I  had  noticed  lower  down  the  Frat 
valley,  west  of  Ashkala.  Doubtless  gypsum  (and  not 
improbably  rocksalt)  is  associated  here,  as  in  so  many 

?arts  of  Armenia,  with  red  marls  and  sandstones  of 
Ipper  Miocene  (Sarmatian)  age.  They  are  overlaid 
by  a  later  series  of  gravel  and  sand,  evidently  the 
westerly  continuation  of  Abich's  '  Myttlus  [Dreyssensia 
polymorpha^  Pallas]  marls,'  which  occur  all  round 
the  outer  margin  or  the  plain  of  Erzerum. 

A  little  further  on,  at  Karabairkh,  these  beds  pass 
up  into  friable,  grey  marls,  with  thin  seams  of  lime- 
stone,— the  whole  series  forming  a  shallow  syncline, 
dipping  30^  to  W.N.W.  and  E.S.E.  In  a  riverside 
cliff  these  beds  showed  a  series  of  small  dip-faults 
hading  to  E.S.E.  They  give  rise  to  a  succession  of 
flat-topped  hills.  At  Evrani  the  snow -clad  plateau  of 
Kargabazar  (10,000  feet)  came  well  into  view,  bound- 
ing the  horizon  on  the  east.  The  walls  of  this  village 
are  built  of  black,  basaltic  lava,  either  highly  scoria- 
ceous,  or  else  compadl  and  streaky,  with  a  pitchy 
lustre.    The  blocks  have  apparently  been  derived  from 

:^  GmI.  Forsch,  in  dm  kauk.  Ldndem^  ii.  105;  and  see  Chapter 
XXIlI  on  the  Miocene  of  Armenia. 


44  Geology,  of  Armenia 

the  heights  lying  to  the  south  of  the  road.  It  is  only 
at  Alaja  that  the  broad  Frat  valley  definitely  widens 
out  still  further  to  form  the  extensive  plain  of 
Erzerum. 

The  upper  Euphrates  thus  furnishes  a  marked  ex- 
ception to  the  general  rule  that  a  river-valley  becomes 
narrower  when  followed  upwards.  Here,  on  the 
contrary,  we  find  a  succession  of  broad  plains,  divided 
from  each  other  by  constriAions  of  the  valley, — either 
narrow,  as  at  Kardarich,  or  else  extending  for  a 
considerable  distance,  as  in  the  long  gorge  between 
the  plains  of  Erzingan  and  Mamakhatun.  This 
unusual  state  of  things  finds  its  solution  in  the  faA 
that  Armcoia  during  the  Upper  Pliocene  and  Pleisto- 
cene was,  to  a  great  extent,  covered  by  lakes,  in  which 
Dreyssensia  polymorpha  flourished^  often  in  great  abun- 
dance. The  larger  rivers^  such  as  the  Frat,  Murad, 
Kur  and  Araxes,  were  evidently  formed  by  the 
overflow  at  these  lakes  across  the  border-ranges ;  and, 
as  the  rivers  gradually  worked  down  to  their  base- 
levels,  the  chains  of  lakes  became  successively  drained 
dry,  leaving  large  walled  plains  or  the  smaller  ova$^ 
which  are  so  charadleristic  of  the  Armenian  plateau. 

Low^  imdulating  foot-hills  of  dusty  white  lacustrine 
marls  occur  on  both  sides  of  the  plain  of  Erzerum, 
and  rise  up  to  a  height  of  about  200 — 300  feet  above 
the  plain,  which  near  the  road  is  composed  of  brown, 
alluvial  gravel,  containing  large,  subangular  blocks  of 
a  black,  basaltic  lava.  The  white  marls  and  shaly 
limestones  were  soon  seen  to  conformably  overlie  a 
blue  clay,  dipping  about  lo**  N.N.W.  The  plain  con- 
tinued to  widen  out,  and  these  lacustrine  beds  receded 
gradually  froYn  view. 

At  Ilija,  the  hot  springs,  which  have  been  used  for 
baths  since  Roman  times,  are  enclosed  in  two  small, 
octagonal  buildings  on  the  north  side  of  the  road. 


The  Vavuk  Pms  to  Erzerum  45 

The  village  is  built  at  the  foot  of  a  hill  of  brown, 
much  weathered  basalt,  in  broad,  stumpy  columns, 
the  ends  of  which  form  a  smooth  pavement  of  hexa- 
gonal blocks.  Unfortunately  I  had  no  opportunity 
of  visiting  the  springs  or  of  examining  the  bed  of 
diatom-earth,  which  Koch*  found  underiying  the 
basalt. 

Boulders  of  a  black,  highly  scoriaceous  lava  lie 
scattered  all  over  the  plain ;  the  steam-cavities  often 
reach  a  length  of  2 — 3  inches.  Now  and  again  the 
rich,  black  soil  of.  the  plain  is  exposed  to  view  before 
reaching  Erzerum  (6168  feet ;  33  miles,  from  Ashkala). 
It  is  situated  on  the  broad  detrital  fan,  which  has 
been  brought  down  from  the  great  cirque  of  Palan- 
doken  and  Eyerli  Dagh.  A  ravine,  which  the  torrent 
has  cut  through  this  fragmental  material,  divides  the 
citadel  and  the  old  town  from  the  larger  and  more 
modem  part.  The  old  site,  moreover,  owes  its  greater 
height,  and  comparative  immunity  from  the  violent 
earthquakes  of  this  region,  to  the  fadl  of  its  having 
been  built  on  solid  rock,  which  rises,  like  an  island, 
out  of  the  enveloping  gravel  and  sand.  The  citadel 
and  the  ancient  towers  of  the  Chifteh  Minareh  have 
resisted  the  severe  shocks,  not  only  of  the  earthquake 
of  June  1859)  which  destroyed  4500  houses  and 
levelled  nine  minarets, — but  also  of  the  more  recent 
disastrous  shocks  of  November  8th,  1901,  which 
destroyed  the  greater  part  of  the  town  situated  on  the 
yielding  fragmental  deposits. 

*  I^dse  impaPiHschen  GMrge,  u/  364-5 ;  and  see  Chapter  XXTV  on 
the  Pliocene  and  Pleistocene  of  Armenia. 


CHAPTER  V 


ERZERUM  TO   KHINIS 


On  June  20th  we  set  out  on  horseback  for  our  journey 
to  Lake  Van.  At  first  there  is  a  gentle  rise  up  the 
great  detrital  fan  issuing  from  the  cirque  of  Palan- 
doken.  A  military  road  has  been  made  from  Erzerum 
up  to  the  two  forts  on  the  southern  wall  of  the 
cirque;  but,  with  this  exception,  the  whole  of  our 
subsequent  journey  was  performed  in  a  country  totally 
devoid  of  any  kind  of  routes  excepting  mule- tracks. 

The  northern  wall  of  Palandoken  shows  rugged, 
brown  precipices,  presenting  all  the  charaAeristic 
features  of  the  serpentinous  olivine-pyroxene  rock 
which  I  met  with  in  the  interior  of  the  cirque.  The 
rocky  cliffs  on  the  north  side  of  the  Erzerum  plain  (to 
the  south  of  Sheikh jik)  also  consisted  of  this  ultra- 
basic  rock  (Chapter  XIV),  and  presented  precisely  the 
same  appearance. 

Subsequent  observations  on  the  plains  of  Mush, 
Khinis  and  Giimgiim  led  me  to  the  conclusion  that 
these  walled-in  plains  of  the  Armenian  plateau  have 
been  formed,  in  the  first  place,  by  a  faulted  depression 
of  the  ground,  at  a  time  subsec^uent  to  the  Miocene 
period.  The  facft  that  these  plains  are  more  affedled 
by  earthquakes  than  any  other  part  of  Armenia  seems 
to  show  that  the  slipping  down  of  the  blocks  is  strll 


Erxerum  to  Kkinis  47 

taking  place.  The  cliffs  bounding  these  plains  are,  in 
my  opinion,  to  be  regarded  as  fault-scarps;  their 
nature  is,  however,  occasionally  obscured,  partly  by 
the  deposits  of  the  lakes  which  usually  occupied  these 
depressions,  partly  by  the  lavas  of  the  volcanoes 
which  arose  on  their  borders,  along  the  lines  of 
fradhire. 

It  took  us  twenty  minutes  to  traverse  (at  a  walk) 
the  northern  mountain- wall  of  the  Palandoken  cirque' 
The  only  rocks  visible  in  the  gorge  were  dark, 
andesitic  lavas,  displaying  spheroidal  jointing  in 
places,  or  else  splitting  mto  slabs.  Some  layers, 
about  a  foot  in  thickness,  have  weathered  more  easily 
than  others,  and  have  become  a  rich  orange  or  a 
cinnabar-red  in  colour.  The  inclination  of  the  lava 
flows  is  as  much  as  45°  N.W.;  the  boulders  lying 
on  the  road  are  often  scoriaceous. 

After  entering  the  great  cirque,  the  road  keeps  to 
the  eastern  side,  rising  at  first  over  grass- grown 
hummocks.  Soon,  however,  we  were  at  a  sufficient 
height  to  look  down  into  the  interior,  which  shows 
a  very  irregular  and  uneven  surface,  with  brown, 
nigged  masses  of  serpentine  and  limestone  rising 
out  of  grassy  slopes.  A  little  stream  trickled  down 
from  the  eastern  wall  of  the  cirque.  Marmorized 
limestone  occurred  here,  dipping  about  60°  S.E. 

We  reached  the  crest  of  the  southern  wall  of  the 
Palandoken  cirque  (9780  feet ;  7J  miles  from  Erzerum) 
one  hour  after  our  entry  into  the  gorge.  During 
our  ascent  we  frequently  obtained  a  glimpse  of 
Erzerum  and  of  the  snow-capped  mountains  on  the 
north  side  of  the  plain — a.  view  framed  in  on  the  east 
by  the  cone  of  Palandoken  (10,694  feet),  ^^  ^^^  west 
by  the  double-crested  outspur  of  Eyerli  Dagh.  The 
southern  wall  of  the  cirque  consists  of  a  white  marble, 
intruded  and  altered  by  serpentine  (olivine-enstatite 


4S  Geology  of  Armenia 

rock,  No.  24)  *  associated  with  a  fine-grained  epi- 
diorite  (No.  25),!  and  capped  on  the  east  oy  a  brown, 
scoriaceous  lava,  which  is  inclined  15®  to  S.E.  The 
chief  point  of  emission  of  this  lava  must  have  been 

*  The  rock  b  a  very  dark  green  lerpentitie,  with  rather  numerous 
grains  of  paler  green,  lustrous  BASTm  (up  to  6  mm.) ;  it  is  frequently 
traversed  by  strings  of  magnetite  and  thm  veins  of  a  lighter  ^een 
serpentine.  It  is  only  very  slightly  magnetic  (x*).  Under  the  micro- 
scope, none  of  the  olivine  or  enstatitb  has  escaped  alteration  into 
serpentine  and  bastite  respectively.  A.  few  small  crystalline  grains 
of  CHROMrrs  are  present,  bordered  with  magnetite  and  sometimes 
sightly  translucent  in  the  centre.  Secondly  magnktitb  occtirs  in 
stnngs  along  the  veins  of  serpentine.  Becke  [Abich,  dot.  Forsck. 
in  den  kauk.  Landem^  ii]  has  examined  a  serpentine  (his  No.  14)1 
taken  from  the  floor  of  the  drque;  it  is  precise^  similar  excepting 
that  portions  of  the  olivine  and  enstatite  stul  remam  unaltered. 

t  The  rock  is  compact  and  ratha:  finely  aystalline,  conmsting 
of  black  -hornblende  and  white  felspar  crystals  witii  a  pale  bluish 
grey  tinge  (both  up  to  2  mm.),  the  former  sUghtly  exceeding  the  latter 
in  amount.  The  felspar  has  a  prismatic  tendenqr  and  on  the  whole 
presents  a  rather  jagged,  hieroglyphic  appearance.  Under  the  micro- 
scope^ the  green  hornblende  is  seen  to  be  actinolitic,  only  seldom 
compact  or  with  well-marked  crystal-outlines ;  most  of  the  individuals 
are  merely  sheaves  of  actinolitic  needles,  with  the  usual  pleochroism 
(a  pale  straw,  b  sap  green,  c  deep  blue-green,  €>h>a\  when  compact 
c  is  dull  sap  green  without  a  bluish  tinge) ;  it  surrounds  the  felspar  in  a 
subophitic  manner,  and  the  needles  often  project  into  the  felspar ;  it  is 
probably  derived  frcm  augite,  although  none  of  this  mineral  is  actually 
present.  The  felspar  is  both  primary  and  secondary;  the  former 
is  of  prismatic  habit,  while  the  latter  usually  forms  a  dear,  narrow 
margin,  or  even  completes  the  tennination  of  a  turbid  felspiar  lath ; 
the  earthy  decomposition-products  have  peady  obscured  the  twinning 
of  the  primary  felspar,  but  the  extinction-angles  point  to  an  acia 
LABRADORFiE  (about  AbiAur),  while  the  secondary  felspar  is  albitx. 
Apattte  18  accessory.  Ilmenitb  occurs  in  plates  and  skeleton- 
rhombohedra,  with  ragged  outlines ;  it  shows  incipient  alteration  into 
granular  leucoxene. 

Very  similar  rocks  have  been  described  by  Becke  [/.c],  viz.  hit 
Nos.  33  &  34  from  the  Shatin  Dagh  above  Kag^zman  (east  of  Palan- 
doken,  on  the  same  line  of  heights),  and  m&  No.  37  from  the 
west,  near  Sipikor,  in  the  mountain  chain  north  of  Erzingan.  In  both 
cases  they  occur  ''  in  close  relation  to  neighbouring  serpentines/' 


Erxerum  to  Khinh  49 

the  peak  of  Palandoken ;  the  Eyerli  Dagh,  however, 
to  the  south- west,  probably  marks  the  former  exit 
for  the  lavas  in  the  virestem  part  of  the  cirque,  for 
the  lava-streams  seen  in  sedtion  on  the  west  of  the 
gorge  dip  as  much  as  60^  N.E.v  while  those  on  the 
southern  slope  of  the  Eyerli  Dagh  are  inclined  to 
&e  south.  On  my  return  journey  to  Erzerum,  I 
entered  the  western  part  of  the  cirque  by  a  pass  on 
the  east  itide  of  the  Eyerli  peak,  the  cliffs  of  which 
Gonsist  of,  columnar  lava — ^a  compacfl  pyroxene-hom- 
blende-andesite  (No«  26),*  similar  to  the  superficial 
lavas  of  Madrak  and  the  Bingol  Dagh,  vide  infra. 

The  ^ound  of  the  cirque,  about  half  a  mile  north 
of  thj^  Eyerli  cli£Es,  consists  of  large  mounds,  grassy 
and  rounded  on  the  south  side,  rough  and  rocky  on 
the  north  side,  morainic  in  characften  They  contain 
angular  fragments  of  limestone,  marble  and  lava. 

The  western  wall  of  the  cirque  is  much  lower  than 
the  eastern,  and  is,  in  fadt,  not  a  continuous  barrier, 

*  The  rock  is  blade,  veiy  oompact  and  basaltic,  with  conchoidal 
fxacture ;  it  is  exceedingly  fine-grauied ;  only  a  very  occasional  felspar 
.lath  (np  to  s  mm.)  can  be  dtsomed  with  the  lens.  Under  the  micro- 
scope,  there  are  no  noticeable  phenocfysts,  but  some  of  the  elements 
of  the  groundmass  are  slightly  laiger.  Hornblende  occurs  thus  in 
tiny  idiomorphs,  almost  entirely  or  even  wholly  resorbed,  with  a 
narrow  border  of  secondary  augite,  and  occasionally  showing  a  dark 
brown  nudeus.  Brokzite  is  idso  present  in  very  small  idiomorphs, 
but  larger  and  more  abundant  than  the  augite  granules ;  it  is  slighUy 
pleocfaroic  (pale  yellow  to  very  pale  pink) ;  in  one  of  the  larger 
prismatic  sections  a  fittle  serpentmous  mineral  has  formed  along  a 
cross-joint  The  largier  laths  of  plagioclase  show  extinction-angles 
indicating  an  add  labradoritb  (about  AbiAni),  and  often  show 
jagged  ends.  The  GROtJNDUASs  is  a  very  dose-grained  felt  of  felspar 
microlites  (andesine)  in  fluidal  arrangement,  with  granules  of  augite 
and  bronxite,  and  magnetite  in  dust  and  granules — the  last  mineral 
predominating  in  the  more  dusky  irrq[ular  areas.  A  colourless  or  veiy 
pale  brown  glass  can  be  seen  in  the  thinner  parts  of  the  fiUdfry  it 
is  much  crowded  with  belonites  and  dust.  Tridvmite  is  present 
liniqg  a  tiny  cavity. 


5P  Geology  of 

but  both  the  walls  agree  in  consisting  essentially  of 
nigged,  rusty-brown,  olivine-pyroxene  rocks,  much 
serpentinized.  Indeed,  lava  has  only  played  a  quite 
subordinate  part  in  the  strudlure  of  this  region, 
merely  forming  the  Palandoken  and  Eyerli  peaks,  and 
the  upper  part  of  the  northern  wall,  while  it  is  wholly 
absent  from  the  rest  of  the  cirque  * 

On  this  occasion  of  my  return  to  Erzerum,  I  pro- 
ceeded across  the  very  imeven  interior  of  the  cirque 
(7724  feet  according  to  Abich),  which  here  consisted 
entirely  of  limestone  intruded  oy  serpentinous  olivine- 
pyroxene  rocks.  I  could  see  nothing  of  the  alabaster 
cliffs  mentioned  by  Abich,  who  pronounced  this  lime- 
stone to  be  Upper  Cretaceous  (Turonian),  but  without 
giving  any  evidence  for  his  conclusion.!  I  noticed 
some  large  oysters  in  the  cream-coloured  limestone, 
(which  dips  50°  S.  by  E.),  near  the  centre  of  the  cirque, 
but  unfortunately  time  did  not  permit  of  my  stopping 
to  obtain  the  specimens.  This  limestone  closely 
resembles  in  appearance  the  neighbouring  Lower 
Miocene  limestones  lying  between  Palandoken  and 
Madrak. 

The  view,  which  Abich  held,:J:  that  this  cirque  was 
a  large  explosion-crater  must  certainly  be  abandoned. 
I  consider  it  to  be  merely  the  northern  edge  of  the 
lava-covered  plateau  (between  Bingol  and  Erzerum), 

*  Bccke  has  described  several  rocks  gathered  by  Abich  from  the 
Palandoken  region  FAbich,  (7«t?/.  Forschn  in  den  kauk.  Ldndem^  ii.  329, 
it  stq,^  and  iii.  102],  viz,\ — olivine  basalt  (2);  glassy  augite-andesite, 
weathering  a  brick-red  (4);  bronzite-augite-andesite ;  augite-andesite 
tuff;  andesitic  tuff  with  water-worn  fragments;  |^lassy  quartz-augite- 
andesite;  augite  trachyte  (2);  rhyoiite;  serpentine  (olivine-enstatite 
rock)  from  the  floor  of  the  cirque ;  grey  marmoraceous  limestone  (with 
calclte  veins)  intruded  by  the  serpentine. 

t  Abich,  op.  eit.  ii.  106. 

X  Op.dt.  ii,  73-6,  781  Ss-y,  90. 


ErMerum  to  Kkiuis  51 

which  has  been  more  than  usually  hollowed  out  by 
erosion,  owing  to  its  great  difference  in  level  (3000 
feet)  with  the  Erzerum  plain  at  its  foot.  In  the 
former  colder  period,  when  there  were  glaciers  on  the 
Bingol  Dagh  (Chapter  XIII),  precipitation  was  greater 
and  the  agents  of  denudation  far  more  powerful  than 
at  the  present  day,  so  that  instead  of  the  detrital 
£an  being  increased  it  is  now  being  disseded  by  the 
stream  which  built  it  up. 

On  resuming  our  outward  journey,  we  found  the 
descent  from  the  southern  wall  of  Paland5ken  into  the 
extensive  plateau-basin  of  Tekman  to  be  rather  steep. 
The  ground  consists  of  a  white  limestone,  with 
W.S.W — EJ^.E.  strike,  but  it  has  been  much  obscured 
by  talus  from  the  lava-cap  to  the  east  of  the  pass. 
We  traversed  a  line  of  low  limestone  hills  about  a 
mile  to  the  south;  here  a  syncline  is  revealed  with 
a  dip  of  about  60^  S.S.E.  and  N.N.W.  Serpentinous 
rocks  frequently  break  through  these  limestones, 
which  are  probably  of  Lower  Miocene  (Tortonian) 
age,  vide  infra.  The  ultrabasic  rocks  have  evidently 
risen  up  through  fissures  parallel  to  the  strike  of 
the  white  or  buff  limestone,  which,  near  the  juncflion, 
has  become  altered  into  a  grey  marble  with  veins 
of  calcite.  This  alternation  or  limestone  and  ser- 
pentine in  zones  occurs  four  times  between  Palan- 
doken  and  Madrak,  and  gives  rise  to  a  tame  landscape 
of  undulating  downs.  The  serpentine,  here  as  else- 
where in  Armenia,  is  frequently  associated  with  a 
speckled  plagioclase-homblende  rock ;  they  form  large 
rounded  heights,  with  occasional  outcrops  of  bare,  red 
or  greenish  rock.  Another  large  hill  of  limestone 
showed  a  dip  of  20^  to  N.W.  Darkness  was  rapidly 
coming  on  as  we  rode  up  a  ravine  cutting  through 
a  zone  of  serpentine  hills  ;  and,  after  fording  a  small 
river   and   rounding  a  bleak  hill  of  limestone,  we 


52  Geology  of  Armenia 

finally  reached  the  Kurdish  village  of  Madrak  (7061 
feet ;  8  miles  from  the  Palandoken  forts).  It  lies  in  a 
broad,  rather  marshy  valley,  surroundea  by  limestone 
downs  and  dominated  on  the  west  by  the  precipitous 
break-off  of  a  high  lava-plateau.  The  lower  ground 
consists  of  a  brownish  limestone  (No.  27),  crowded 
with  bryozoan  fragments,  foraminifera,  occasional 
plates  and  spines  of  sea-urchins  (Cidaris)^  and  Litho- 
thamnion.  Mr.  R.  Bullen  Newton,  F.R.S.,  has  kindly 
examined  and  determined  my  specimens,  confirming 
my  opinion  of  their  Lower  Miocene  age ;  the  deposit 
is  in  jbjA  similar,  both  in  composition  and  horizon,  to 
the  Leithakalk  (Tortonian)  of  the  Vienna  basin.  The 
fossils  comprise  the  following : — 

Orbitoides  ( Lepidocyclina)  Mantelli. 

Antphistegtna  cf.  Hauerina^  d'Orb.* 

Orbttolites. 

Operculina. 

Mtliolina. 

Ceriopora  anomala^  Abich.* 
C.  cf.  palmata^  Abich. 

Ltthothamnton  ramossissitnum^  Reuss.^ 
Abich  t  has  described  a  very  similar  bryozoan  lime- 
stone, with  abundant  foraminifera,  and  Ceriopora 
intricata^  d*Archiac,  from  Mamakhatun  on  the  Frat 
below  Ashkala;  it  underlies  an  Orbitoides-limestonef 
and  overlies  an  Echinoderm-limestone.  It  is  not  im- 
probably in  continuity  with  the  limestone  of  Madrak, 
which  lies  60  miles  to  the  eastward,  on  nearly  the 
same  line  of  strike. 

The  bryozoan  limestone  of  Madrak  underlies  a 
massive  limestone,  which  shuts  in  the  valley  on  the 
north ;   the  latter  contains  ill-preserved  specimens  of 

*  These  detenninations  were  confirmed  by  Mr.  Newton. 

t  Gici,  F^rsch.  in  den  kauk.  Laniem^  ii.  129,  150,  993-241 ;  see  also 
Chapter  XXIII  on  the  Miocene  of  Armenia. 


JI^^-"■if'•- 


mT" 


iCflrttjUu-n^^ 


Erzerum  to  Khinis  53 

Thamnarcea  polymardha^  Abich,  and  dips  60^  N.;  the 
weathered  surface  or  the  rock  has  a  very  charadleristic 
pitted  and  channelled  appearance  (No.  28). 

On  the  morning  of  June  21st  we  rode  up  to  the  foot 
of  the  lava-cliff  at  the  head  of  the  valley,  and  after 
a  somewhat  laborious  climb  over  loose  fragments  of 
lava  we  reached  the  summit  of  the  lava-plateau, 
which  extends  for  a  considerable  distance  to  the  west 
and  south-west.  The  dark  grey  lava,  a  pyroxene- 
homblende-andesite  (No.  29),^  showed  a  rudely  colum- 
nar structure  with  subconchoidal  fradlure  in  the  lower 
part  of  the  cliff,  but  on  the  plateau  it  splits  up  into 
thin  slabs  like  tiles,  which  frequently  weather  a  brick- 
red.  The  lines  of  flow  are  clearly  marked  out  by 
steam-pores,  and  are  especially  prone  to  weather  into 
rusty  brown  bands.  A  ropy,  twisted  charadler  is  not 
infrequent.  Large  blocks  and  slabs  lie  strewn  over 
the  surface  of  the  plateau,  and  sometimes  occur  piled 
up  into  small  ridges,  which  may  perhaps  mark  former 
points  of  emission.  The  lava  is  so  remarkably  similar 
in  composition  and  stru(5lure  to  that  of  the  Eyerli 
Dagh  on  the  north  (p.  49),  and  to  the  superficial  lavas 

•  The  rock  is  dark  grey,  with  a  strong  purpUsh  tinge  owing  to 
the  purple  weathering-border  round  each  steam-pore.  The  minute 
steam-pores  are  ranged  along  the  parallel  lines  of  flow,  and  are 
sometimes  filled  with  a  white  mineral  (tridymite).  The  lens  only 
reveals  an  occasional  tiny  felspar.  Under  the  microscope,  the  rock  is 
highly  vesicular,  consisting  of  a  microfelt  of  felspar  microlites 
(andesine),  with  granules  of  augite  and  magnetite.  The  larger  felspar^ 
laths  give  extinction-angles  pointing  to  an  acid  labradorite  (about 
AbxAni),  sometimes  enclosing  augite  granules ;  a  few  larger  crystals 
contain  a  network  of  the  groundmass.  Hornblende  in  very  small 
idiomorphs  is  fairly  frequent ;  usually  completely  resorbed,  rarely  with 
a  narrow  border  of  secondary  augite.  Some  tiny  prisms  of  very 
pale  yellow  bronzitb,  slightly  pleochroic  (to  pale  pink),  exceed  in  size 
the  numerous  augite  granules.  There  are  a  good  many  grains  of 
MAGNETITE.  Tridymitb  is  abundant  in  the  usual  imbricated  ag- 
gregates in  nests,  or  else  lining  cavities  in  hexagonal  tablets. 


54  Geology  of  Armenia 

of  the  northern  and  western  Bingdl  plateau  (vide 
infra)^  that  they  must  have  proceeded  £rom  one  and 
the  same  magma. 

Prom  this  standpoint  (c.  8500  feet)  the  extensive 
view  to  the  east  and  south  is  bounded  on  the  north  by 
the  rugged  Aghri  Dagh  from  Palanddken  to  Kuseh 
Dagh— on  the  east  by  the  snowy  peaks  of  the  volcanic 
Ala  Dagh — on  the  south-east  by  the  bare,  furrowed, 
marble  heights  of  the  Ak  Dagh,  overtopped  by  the 
giant  cone  of  Sipan,  a  snowy  pimntom  fully  a  hundred 
miles  distant — and  on  the  south  by  the  swelling  out- 
lines of  the  Bingdl  volcanic  plateau.  The  central  and 
eastern  part  of  the  view  embraces  the  basin  of  the 
upper  Araxes,  and  shows  a  broad  zone  of  fertile,  red 
strata  near  Tashkessen,  overlying  the  bu£F  Miocene 
limestones  of  Madrak,  which  extend  far  away  with 
E.N.E.  strike.  Now  since  some  salt-pans  occur  at 
Tuzla,  a  few  miles  south  of  Tashkessen,.  it  is  probable 
that  this  red  zone  marks  the  outcrop  of  the  Upper 
Miocene  red  marls  and  sandstones  (with  gypsum  and 
rocksalt),  which  I  had  already  noticed  in  the  Frat 
valley  below  Ashkala«  The  serpentinous  rock,  whidi 
is  intrusive  in  the  Lower  Miocene  limestones,  weathers 
either  to  a  bluish  grey  or  to  a  dull  red,  and  appears  to 
be  very  barren. 

After  leaving  Madrak  our  course  lay  nearly  due 
south  towards  the  Bingdl  Dagh.  The  track  rises  over 
the  grassy  down  of  sandy  limestone  (dipfiing  to  S.E.), 
which  frames  in  the  Madrak  vall^  on  the  south.  On 
our  way  we  passed  several  conical  hills  of  limestone 
capped  by  fissile  lava;  they  are  evidently  outliers 
of  the  lava  plateau,  isolated  by  denudation. 

The  limestone  soon  changed  its  dip  to  the  N.W.» 
showing  a  small  escarpment  facing  N.E.;  but  at  Rez 
Eli  all  signs  of  sedimentary  strata  had  disappeared ; 
the  only  rock  visible  was  the  grey,  fissile  lava,  which 


Erxerum  to  Kktnis  55 

weathers  to  a  brick-red.  But,  as  we  descended  to 
Duzyurt,  on  the  Egri-chai  (a  tributary  of  the  Araxes), 
the  lava  became  black  and  scoriaceous,  forming 
rugged  clifEs ;  the  lava  beds  are  here  gently  inclined 
to  the  south-west.  The  river  above  Duzyurt  flows 
through  a  grassy  valley*,  about  a  quarter  of  a  mile 
wide;  but  below  the  little  Kurdish  village  it  rushes 
down  a  deep  and  gloomy  gorge  through  the  black 
lava.  However,  on  looking  down  the  river,  I  could 
see  that  it  soon  emerged  from  the  lava  ravine  to  cut 
its  way  through  low  hills  of  white,  nearly  horizontal 
strata,  similar  in  appearance  to  the  Pleistocene  lake- 
deposits,  which  I  found  at  Khedonim,  a  few  miles 
further  south  {vide  infra). 

After  fording  the  river,  we  rode  over  a  succession 
of  grassy  river- terraces,  about  fifty  feet  one  above  the 
other.  Finally  we  reached  the  summit  of  the  ridge, 
and  an  iminterrupted  view  of  the  snov^  Bingol  Dagh 
was  disclosed.  The  foreground  is  a  plateau  of  cal- 
careous lake-deposits,  gently  tilted  up  (about  5®)  to 
N.N.W.,  and  disseded  out  mto  broad  valleys  by  the 
upper  Araxes  and  its  tributaries.  I  could  see  these 
strata  extending  southwards  towards  Bingol  for  about 
five  miles,  and  eastwards  down  the  Araxes  valley 
for  at  least  six  or  seven  miles.  These  deposits  of 
Pleistocene  age  are  capped  by  a  lava-sheet  of  no  great 
thickness,  gently  slopmg  away  from  the  barren  hills 
in  the  south-west.  These  hills,  however,  consist  (as  I 
subsequently  discovered)  of  grey  and  white  marble, 
weathering  in  rusty-brown  patches,  and  much  fur- 
rowed by  valleys.  They  are  the  westerly  continuation 
of  the  similar  Ak  Dagh  peaks,  and  occasionally  show 
a  capping  of  lava,  probably  the  remnants  of  a  former 
lava-plateau,  which  has  been  almost  entirely  denuded 
away. 

As  we  descended  to  Kheddnun  in  the  Araxes  valley, 


56  Geology  of  Armenia 

it  became  evident  that  the  lava,  which  we  had  been 
traversing  (all  of  the  same  character  as  near  Madrak, 
P-  53)>  overlies  the  calcareous  marls  dipping  10® 
N.N.W.  Khedonun  (6713  feet;  iii  miles  feom  Ma- 
drak), lies  at  the  foot  of  the  cliff,  on  a  small  left 
tributary  of  the  Araxes.  Here  I  found  numerous, 
angular  fragments  of  a  pale  brownish-grey  con- 
glomerate (No.  30)  of  rounded  pebbles  (up  to  13  mm.) 
of  quartz  grains,  serpentine,  limestone  and  lava,  in 
a  calcareous  cement,  with  minute  shells  of  Spherium 
(Cyclas)^  and  Bithinia.  Unfortunately  I  could  not 
find  the  parent-bed  of  this  tufaceous  lake-deposit,  but 
the  fragments  had  been  brought  down  by  the  river 
from  no  great  distance. 

To  the  south  of  the  village  the  Araxes  flows  at 
the  base  of  a  cliff  of  lake  deposits,  but  extensive 
marshes  prevented  me  from  crossing  the  river  to  ex- 
amine these  strata.  However,  the  rounded  hill  lying 
between  Khedonun  and  the  Araxes  showed  that 
similar  calcareous  beds  are  interbedded  with  con- 
temporary lava  flows.  The  harder  outcrops  present 
all  the  charadleristics  of  a  travertine  (No.  31),  and 
doubtless  arose  from  the  activity  of  calcareous  springs. 
No  fossils  could  be  found  in  these  deposits.  The  dark 
purple  lava,  an  augite-homblende-trachyte  (No.  32),* 

*  The  rock  is  dull  purplish-grey,  extremely  porous  and  dndery,  with 
lenticular,  greyish-black,  more  coarsely  vesicular  streaks  (from  minute 
dimensions  up  to  2  inches  in  length  in  the  hand-specimen,  but  longer 
in  the  rock  tn  situ)^  drawn  out  along  lines  of  flow ;  it  contains  tiny 
felspars  (up  to  2  mm.),  and  a  few  angular  fragments  (up  to  5  mm.)  of 
pale  grey  stony  lava.  Under  the  microscope,  a  few  phenocrysts 
of  clear  sanidine  are  present,  together  with  some  oligoclasb,  whidi  is 
much  honeycombed  with  a  network  of  groundmass.  Augite  occurs 
in  a  few  fragmental  crystals  of  bottle-green  coloiur,  slightly  pleochroic 
(to  greenish  yellow).  An  occasional  crystal  of  hornblende  is  still 
recognisable  by  the  characteristic  outlines,  although  now  wholly  re- 
sorbed  and  opaque.  The  vesicular  streaks  consist  of  a  deh'cate 
lacework  of  steam-cavities,  the  thin  dividing-walls  being  composed  of 


»RX^ 


4 


S§^ 


I 


Erzerum  to  Khinis  57 

forms  a  marked  contrast  to  the  white  or  pale  buff 
limestone  (No.  33) ;  it  forms  a  bed  6 — 10  feet  thick,  in 
broad,  stumpy  columns,  and  was  clearly  contem- 
poraneous with  the  calcareous  deposits. 

We  resumed  our  journey  to  Khinis  by  fording  the 
Araxes  and  proceeding  southwards  up  the  broad 
alluvial  valley  of  the  right  tributary  above  the 
village.  It  is  bounded  on  ooth  sides  by  cliffs  of  the 
lake-deposits  (still  dipping  ^®  N.N.W.),  which  become 
particularly  white  near  their  summits ;  probably  this 
whiteness  is  due  to  the  contadl-adtion  of  overlying 
lava-flows,  which  have  been  for  the  greater  part 
removed  from  the  foreground  by  denudation.  A 
small  lateral  valley  clearly  showed  these  mutual 
relations  near  its  head.  The  cindery  lava,  with 
jet-black  vesicular  streaks  up  to  a  foot  in  length, 
IS  of  no  great  thickness  and  occurs  in  the  charadleris- 
tic,  stumpy  columns  already  noticed  in  the  similar 
lava  of  Khedonun  (v.  supra).  Shortly  before  reaching 
Kalaji  we  passed  over  a  broad  outcrop  of  this  lava, 
here  underlying  the  calcareous  beds ;  the  ends  of  the 
wide  columns  form  a  smooth  pavement  of  polygonal 
blocks.  Kalaji  is  pic5turesquely  situated  at  the  foot  of 
a  low  cliff  of  this  sombre  rock,  from  which  it  derives 
its  name.  A  gradual  rise  for  three  miles  over  broad 
slopes  of  this  kiva  brought  us  to  Kherbesor  (8f  miles 
from  Khedonun'),  on  the  plain  at  the  foot  of  the 

radiftl  aggregates  of  felspar  needles  (sanidine),  with  much  magnetite 
dust  These  stieaks  are  rather  shaiply  marked  off  from  the  finer- 
grained  OROUNDMASS,  which  is  a  felt  of  felspar  microlites  (sanidine), 
tiny  green  augite  needles,  and  abundant  magnetite  in  dust  and 
grainides,  with  much  interstitial,  pale  brown  glass.  There  is  a  great 
tendency  on  the  part  of  the  felspar  needles  to  form  spherulitic 
aggregates,  which  are,  however,  mudi  interfered  with  by  the  vesicular, 
porous  nature  of  the  rock.  Flow  structure  is  well  marked.  The  few 
fragments  of  more  cnrstalline  lava  are  rhyolitic  and  cryptocrystalline, 
with  lines  of  flow^  marked  out  by  magnetite  granules  and  dust. 


58  Geology  of  Armmiia 

high  northern  cliff  of  the  Bing5l  plateau  (Chapter 
XTlI).  Our  course  now  lay  to  S.E.,  keeping  about 
half  a  mile  away  from  the  foot  of  the  cliff.  Its  lower 
part  is  clearly  composed  of  black,  columnar  lava, 
separated  by  horizontal  calcareous  strata  from  a  later 
lava-flow  forming  the  summit  of  the  cUff,  which  rises 
about  1000  feet  above  the  Kherbesor  plain.  The 
small  streams  flowing  through  this  plain  to  join  the 
Araxes  are  bordered  by  low  cliffs  of  the  black,  cindery 
lava  in  stumpy  columns,  with  the  same  gentle  in- 
clination to  N.W.  as  the  underlying  calcareous  lake- 
dei>osits,  with  which  it  was  evidently  contemporary. 

As  we  proceeded  eastwards  we  passed  on  our  left 
a  cliff  of  columnar  lava,  forming  a  selvage  to  the 
grey  furrowed  marble  hills  of  Kurreh  Hassan  Dagh  in 
the  north-east,  whence  it  appears  to  have  flowed. 
The  columns  here  were  only  8  inches  in  diameter, 
while  the  cliff  was  about  12  feet  high.  The  rock 
weathers  to  a  brick- red,  but  internally  it  is  grey  and 
cindery,  with  jet-black  scoriaceous  streaks,  as  at 
Khedonun;  in  one  place  I  noticed  that  it  enclosed 
an  angular  fragment  of  grey  marble,  evidently  derived 
from  the  Kurreh  Hassan  hills.  Indeed,  about  a  mile 
further  east,  at  Gabeleh,  I  found  the  grey  and  white 
foliated  marble  in  situ,  dipping  50°  N.W.,  with  a 
steep  escarpment  facing  S.E.,  and  rising  like  an  island 
out  of  the  surrounding  lava. 

After  crossing  a  marsh  we  finally  reached  the  main 
mass  of  the  old  marble  (No.  34),  which  is  not  im- 
probably of  Palaeozoic  age  (see  Chapter  XV),  but  I 
could  find  no  trace  of  any  fossils.  It  has  been  much 
crushed  and  marmorized,  and  is  traversed  by  veins 
of  calcite.  The  dip  rapidly  changed  to  40^  S.£.,  and 
then  again  to  40''  N.W. 

An  intrusive  mass  of  dark  green  serpen tinous  rock 
now  appeared  to  view  at  the  head  of  the  small  valley 


Erzerum  to  Khinis  59 

up  which  we  were  riding,  forming  the  low  watershed 
7340  feet),  a  short  distance  west  of  Ali  Mur  or  Karaba 
7100  feet;  7  miles  from  Kherbesor).  The  marble 
near  the  jundtion  with  the  foliated,  serpentinous 
pyroxene-olivine  rock  (No.  3^)*  becomes  somewhat 
fissile.  This  was  the  only  distindl  instance  of  dy- 
namic metamorphism  in  any  of  the  serpentines  which 
I  gathered  during  the  whole  of  my  journey  to  Lake 
Van.  This  intrusion,  therefore,  may  belong  to  an 
earlier  period  than  these  other  serp)entines,  which 
are  of  mid-Miocene  age«  unless  indeed  the  meta- 
morphism is  due  to  the  faulting  which  (in  my 
opinion)  has  given  rise  to  the  great  Kherbesor — ^Ali 
Mur  cliff  or  fault-scarp. 

The  only  other  described  instance  of  an  Armenian 
ultrabasic  rock  exhibiting  similar  cataclastic  stru(5lure 
is  a  serpentinized  olivine- augite  rock  (Thost's  No.  3),t 

*  The  rock  is  foliated — the  'eyes'  consisting  of  large  crystalline 
grains  of  green  diallagic  augite  (up  to  25  mm.),  in  a  sparing  matrix 
of  dark  green  serpentine.  Under  the  microscope,  the  augite  (chrome 
pioPSiDs)  is  seen  to  occupy  alK>ut  three-quarters  of  the  slide,  occurring 
in  large,  pale  green  crystadlin^  grains,  often  with  parallel  schillerizing 
rodlets,  and  occasionally  enclosing ,  olivine,  which  has  been  wholly 
or  partly  serpentinixed.  Olivine  is  rather  subordinate,  and  has  been 
partly  sdtereid  into  yellow  serpentine,  with  the  usual  separation  of 
magnetite.  Enstatits  is  also  present  in  a  few  rather  small  grains. 
No  spinels  are  visible.  The  rock  exhibits  an  interesting  catiurlastic 
structure.  Dusky  streaks  of  opaque  particles  (white  and  doudy  under 
reflected  light)  occur  in  wavy,  parallel  series,  traversing  all  constituents 
irrespectively.  The  larger  diopside  grains  have  been  broken  up  and 
rounded  at  the  angles;  in  the  interspaces  thus  formed  a  mosaic  of 
secondary  augite  has  developed  in  small,  clear,  colourless  grains 
with  undulose  extinction  and  secondary,  polysynthetic  twinning,  which 
has  a  rather  washed-out  appearance.  The  original  cleavages  of  the 
diopside  can  often  be  still  traced  in  the  secondary  augite,  and  the 
dusky  streaks  have  here  become  partially  cleared ;  flecks  of  a  clear, 
talcose  mineral  also  occur  in  the  secondary  augite. 

t  Thost,  Abka$tdl.  SeHikemb.  Naturf.  Gts.  Frankfurt  a/M,  XVIII, 
iL  241. 


6p  Geology  of  Armenia 

from  the  Kalaman — ^Kalakent  pass  near  the  S.E.  angle 
of  Lake  Gokclxa,  coUedled  by  Valentini^  and  described 
by  Thost  (l.c.). 

The  Kherbesor  cliff  had  now  decreased  in  height 
and  forms  the  southern  edge  of  the  Ali  Mur  valley. 
Here  it  is  composed  of  a  grey,  stony,  fissile  lava»  a 
pyroxene-homblende-andesite  (Nos,  36  &  ^7),!  similar 
in  composition  to  the  lavas  of  Palandoken  (p.  49), 
and  of  Madrak  (p.  53),  and  also  to  the  superficial  lava 
of  the  south-eastern  Bingol  plateau  (Chapter  XIII, 
No.  170). 

About  a  quarter  of  an  hour  east  of  Ali  Mur,  this 
lava  is  covered  by  the  black,  cindery,  trachytic  lava, 
with  broad,  stumpy  columns,  which  I  had  noticed  all 
the  way  from  Khedonun  to  be  overlying  and  also 

*  BendU.  Senckmb,  Naturf.  Ges.  Frankfurt  a/M,  1891,  159. 

t  The  rock  is  pale  grey  (with  a  slight  greenish  tinge),  finely  mottled, 
and  veiy  compact.  The  mottling  is  due  to  an  alternation  of  felspathic 
and  augitic  areas  (less  than  a  mm.).  Dendritic  manganese  oxide 
has  formed  on  the  pale  yellowish  grey  dividing-planes  of  the  rock. 
Under  the  microscope,  phenocrysts  are  all  veiy  small  and  few  in 
number.  Plagioclase  (axdesine  about  Ab4An3)  occurs  in  small, 
scattered  idiomorphs,  usually  honeycombed  with  a  network  of  very 
pale  brown  glass  and  magnetite  granules;  sometimes  zoned.  A  few 
clear,  fresh  crystals  of  sanidinb  are  also  present.  Hornblende  is 
fairly  frequent,  for  the  most  part  wholly  resorbed  and  surroimded 
by  stains  of  iron  oxide,  or  elM  replaced  by  an  aggregate  of  augite 
and  magnetite  granules  with  opal,  and  with  more  abundant  magnetite 
granules  round  the  margin.  Augite  is  in  very  pale  green  idiomorphs, 
quite  fresh,  sometimes  showing  twinning  (both  simple  and  lamellar^, 
and  occasionally  enclosing  a  network  of  groundmass.  Bronzite  is 
also  present,  nearly  colourless  and  slightly  pleochroic  (pale  green  to 
pale  pink),  a  little  older  than  the  augite.  Both  pyroxenes  are  quite 
fresh  and  very  abundant  but  smaller  than  the  hornblende.  The 
GROUNDMASS  is  a  fine  microfdt  of  small  felspar-laths  (andesine),  with 
augite  and  magnetite  granules,  and  some  interstitial,  pale  brown  or 
colourless  glass.  Flow-structure  is  well  marked  in  the  superficial 
specimen  by  lighter  and  darker  streaks,  the  latter  containing  relatively 
more  magnetite  granules.    Tridyhite  is  fairly  frequent  in  small  nests. 


ErMerum  to  Khinis  6i 

interbedded  with  the  lake-deposits.  Here,  at  any 
rate,  it  must  have  been  emitted  at  a  comparatively 
recent  date,  for  the  lava-flows  conform  to  the  present 
relief  of  the  ground,  sinking  down  into  the  valleys 
and  partially  filling  them  up.  The  slope  and  di- 
rection of  the  flows  rendered  it  evident  also  that  the 
point  of  emission  of  this  cindery  lava  must  have  been 
situated  in  the  Kurreh  Hassan  hills,  north  of  Ali  Mur 
(cp,  pp.  55  &  57).  Patches  of  the  lava,  nowhere 
exceeding  20  feet  in  thickness,  often  occur  isolated, 
and  separated  by  denudation  from  the  main  mass, 
lying  on  the  slopes  of  the  grey  marble  on  the  north 
side  of  the  Ali  Mur  valley. 

The  little  river  flows  to  N.E.  to  join  the  Araxes,  but 
we  soon  turned  off  in  a  S.S.E.  direction  up  a  small 
tributary  to  the  col  joining  the  Ak  Dagh  to  the 
Bingol  plateau.  Immediately  after  leaving  Ali  Mur 
we  passed  over  ridges  of  dark  bluish- grey  marble, 
dipping  40°  N.N.W.;  it  has  been  greatly  cracked  and 
brecciated,  and  is  traversed  throughout  by  veins  of 
calcite.  The  mountains  on  the  left  consist  of  the 
same  banded,  grey  marble,  frequently  capped  by 
remnants  of  lava-sheets.  As  we  rode  up  the  gentle 
slopes,  the  marble  became  whiter,  weathering  to  a 
fine  amber.  Silvery  graphitic  schists  (No.  38)  con- 
formably underlie  the  marble;  some  thin  bands  of 
white  marble  are  interbedded  in  the  schists,  which 
become  greatly  contorted  and  puckered.  Owing  to 
their  dip  to  N.W.,  they  continue  to  nearly  the  summit 
of  the  col  (7500  feet),  which  forms  part  of  the 
Araxes — Murad  watershed.  Here  we  obtained  a  fine 
view  of  the  Bingol  Dagh  with  its  gentle  eastern 
slopes  (only  5°)  extending  down  to  the  plain  of 
Khinis,  and  framing  it  in  on  the  south  by  meeting 
the  even  gentler  contours  (3°)  of  Khamur  Pet. 

Immediately  west  of  our  position  the  marble  rises 


r 


&2  Geology  of  Armenia 

up  into  the  Karakaya  ridge,  which  derives  its  name 
from  its  cap  of  black  lava.  It  appeared  to  me  highly 
probable  that  this  lofty  marble  cliff  is  essentially  a 
rault-scarpy  and  that  the  plain  of  Khinis,  like  that 
of  Erzerum,  is  a  block  which  has  undergone  relative 
depression.  This  view  receives  some  support  from 
the  similar  discovery  of  schistose  rocks  which  was 
made  by  Ainsworth*  in  1842  on  the  south  side  of 
the  plain,  while  crossing  the  opposite  col  between 
Bingol  and  Khamur,  near  Bashkent.  He  found  this 
ridge  to  be  composed  of  "  mica-schist  and  clay-schist, 
overlaid  by  conide  and  ferriferous  limestones  "  [doubt- 
less Lower  Miocene],  "conglomerates  of  the  saliferous 
red  sandstones,  various  coloured  sandstones,  sands, 
marls  and  gypsum.  The  last  lava  eruptions  have 
broken  through  the  latter,"  which  are  of  Upper  Mio- 
cene (Sarmatian)  age.  Miocene  limestone  is  known  to 
reach  a  fuller  development  in  the  Zimek  Dagh,  the 
easterly  continuation  of  Khamur  (see  Chapters  VI 
and  XXIII). 

The  plain  of  Khinis  is  also  analogous  to  that  of 
Erzerum  in  having  been  partially  filled  up  with  recent 
lake-deposits,  and  with  lava  from  the  neighbouring 
volcanoes,  which  built  themselves  up  along  the  lines 
of  fradlure. 

Our  descent  from  the  dividing  col  was  commenced 
by  following  a  stream  flowing  to  the  S.E.  The 
valley  had  been  filled  by  a  smooth  lava  flow  from 
Bingol;  it  is  brown  and  cindery,  injdDlumns  10 — 20 
feet  in  diameter,  with  jet-black  vesicular  streaks  (up 
to  a  foot  in  length),  as  at  Khedonun.  Soon  we  came 
again  to  the  bluish- grey  marbles  (extending  far  east- 
wards), and  the  descent  over  the  bare,  polished  rock 
was  very  slippery   and  dangerous  for  horses.    The 

*  Travels  and  Reseanfus  in  Asia  Minor^  Mtsopotanda^  Chaldea  and 
Armenia^  vL  385. 


EtMerum  to  Khinis  63 

valleys  had  rapidly  become  steep  and  gloomy  canons, 
the  upper  himdred  feet  or  so  consisting  of  lavas  from 
Bing5l,  the  lower  part  deeply  carved  out  of  the 
marble,  which  has  been  intruded  by  serpentines. 
Before  reaching  Kilissadereh  I  noticed  three  distindl 
intrusions  of  such  ultrabasic  rocks,  which  weather  to 
a  russet-brown.  The  superficial,  cindery  lava  occurs 
in  broad,  stumpy  columns,  resembling  Cyclopean 
masonry,  along  the  edges  of  the  canon. 

On  the  east  side  of  the  ccnfluence  of  our  river 
with  the  Bingol  Su  a  conspicuous  terrace  has  been 
formed  by  calcareous  springs.  The  brownish  layers 
of  the  travertine  are  horizontal  near  the  base  ci 
the  terrace,  but  bend  slightly  over  along  the  summit ; 
the  rock  (No.  39)  shows  the  usual  cavernous  texture. 
Here  the  JUingol  Su  is  no  longer  an  impetuous  torrent, 
pent  up  in  a  deep  canon,  but  a  placid,  meandering 
stream,  winding  lazily  through  alluvial  flats  covered 
with  undergrowth.  The  soil  is  reddish  owing  to 
the  decomposition  of  serpentinous  rocks,  which  are 
intrusive  into  the  grey  marbles  of  the  steep  north 
bank.  On  the  other  hand,  the  south  bank  (with 
gentler  slopes)  of  the  widening  valley  is  formed  by 
lavas  from  Bingol ;  the  lower  beds  are  compact  and 
stony,  but  the  topmost  still  consists  of  the  black, 
cindery  lava  with  broad  columns. 

Soon  after  fording  across  to  the  southern  bank 
we  were  obliged  to  leave  the  river-bed  and  to  ascend 
to  the  lava-plain,  owing  to  the  river  suddenly  enter- 
ing a  short  ^orge,  cut  through  a  ridge  of  rugged 
serpentine,  with  nearly  vertical  jointing.  The  river, 
after  leaving  this  ravine  and  coming  into  the  open, 
does  not  enjoy  its  freedom  for  long,  but  soon  turns 
back  northward  to  cut  an  equally  short  gorge  through 
the  same  zone  of  olivine*pyroxene  rock  (No  40),* 

*  The  rock  is  motUed  with  nisset-brown  and  green ;  schilleriziQg 


64  Geology  of  Armoms 

much  S6rpentini2sed.  The  river  finally  quits  the 
r^ion  of  the  marbles  and  serpentines,  whidi  recede  to 
the  RE.  into  the  Ak  Dagh. 

We  descended  again  to  the  river*bed,  and  tiie 
landscape  becomes  tame,  for  the  flat-topped  cli& 
consist  of  soft,  greenish  white,  lacustnne  marls, 
with  occasional  bands  of  limestone*  These  strata 
are  nearly  horizontal,  dipping  at  most  xo^  S.E.  The 
alluvial  river-bed  increased  to  about  a  quarter  of  a 
mile  in  width.  On  the  right  bank  a  columnar  lava 
put  in  an  appearance,  overlying  the  mark,  which 
have  been  baked  by  them.  Later  on  the  lava  also 
occurred  on  the  Uit  bank.  Soon  we  came  to  the 
lonely  ruined  church  of  Kilissadereh  (formerly  Medjin- 

Sert)^  situated  cm  a  nearly  detached  hill  at  the  con- 
uence  of  the  Bingol  Su  wi&  a  tributary  from  the 
south. 

Heve  we  ascended  to  the  surface  of  the  plain  of 
Khinis,  leaving  the  church  00  our  left.  The  topmost 
bed  ooosists  of  ^bout  ao  feet  of  rolled  gravel.  Khinis 
lies  hidden  in  a  ravine  about  a  mile  distant.  Our 
camp,  however,  was  pitched  on  the  plain  (^690  feet ; 
18  miles  from  Ali  Mur),  to  the  north  of  this  ancient 
town,  which  is  picturesquely  situated  at  the  bottom 

crystaKnc  gnias  of  greenish,  diaflagir  avfke  (up  to  $  nun.)  lie  ia 
a  mesh  of  brownish-red  serpentine,  mnch  stimed  wiUi  iron  oxide. 
The  rock  is  slightly  magnetic  li*).  Vndet  the  microicope,  ouvink 
is  seen  to  form  the  bulk  of  tke  lott;  a  considerable  amount  has 
escaped  serpentinization  and  remains  dear  and  unaltered  in  die 
meshes  of  the  yellow  serpentine^  iriudi  is  sometimes  rendered 
nearly  opaque  by  iron  oxide.  Auom  (cBaoiis*»ioP8ii>s)  occurs  in 
veiy  pale  green,  allotriomoiphic  ttainsi  sometimes  twinned;  and 
occasionally  enclosing  parallel  todtels  (naifower  and  less  frequent 
than  in  the  Rop  serpentine,  p.  40},  as  well  as  magnetite  granules 
and  patches  of  clear  yellow  serpentme  ^Rcr  olivine.  Enstatitx  b 
sparingly  present  in  smaller  cnrStallilM  ptka.  Magmxtits  occurs 
in  scattmd  grains  and  in  secondary  sMms  along  reins  of  serpentine. 
No  spittds  were  visible. 


EtMnufk  to  Khinis  65 

of)  a  canon  carved  out  of  cindery  lava,  an  augite* 
IiornDlende-trachyte  (No.  43).*  The  rock  presents  the 
same  general  chara<fteristics  as  the  cindery  lava  with 
broad  columns,  which  I  had  noticed  all  the  way  from 
kheddnun  (pp.  56-58,  60-63^.  Since  these  vesicular 
lavas  only  occur  interbeddea  or  in  close  association 
with  lake-dqposits,  and  (as  I  afterwards  discovered, 
see  Chapter  XIII)  only  on  the  outer  borders  of  the 
Bingol  Dagh,  it  seems  to  me  probable  that  their  loose 
texture  is  due  to  the  steam  generated  during  their 
flow  iiit>  the  lake,  which  then  surrounded  this  vol- 
cana  The  lon^  distances  traversed  by  these  cindery 
lavas,  and  their  slight  thickn^^s  show  that  they 
poss^sed  considerable  fluidity.  The  large  size  of 
their  columns  tends  also  to  show  that  they  were  slow 
in  cooling.    Abirh  t  has  described  similar  charadleris- 

*  The  roric  is  ^vnrpllsh  grey,  cindery  and  poroiUi  with  oocitioiial, 
more  cosMtdf  TeacaJar,  gi^hh  black  streaks  along  iines  of  flow; 
h  ooatains  some  Urge  gUusy  crystals  (up  to  xo  mm.)  of  imstriated 
fclsiMtf,  but  mostly  smaller, 'and  also  many  angular  fragments  of 
pale  grey,  stony  lava.  It  is  more  compact  than  tihe  KhedSnun  lata. 
Vnd^  tiie  microscope,  nhonocrysts  occur  sparingly,  and  consist  of 
SAMii  IMS  tdiomorphs  and  fragments,  with  some  fin^  striated  ouoo> 
CLASKp  stiQ  fewer  crystals  and  fragments  of  pale  green  auoite,  and 
some  ima!^  whoDy  resoibed  HoairaLBKDi  crystals,  showiog  the 
characterstic  outlines.  The  vesicular  streaks  are  seen  to  be  a 
lacework  (00  a  smaQer  scde  than  in  die  Khedonun  lava)  of  steam- 
cavtties,  the  thin  walls  consisting  of  radial  and  spherulitic  aggregates 
of  felspar  needles  (sanidine),  with  much  opaque  dust  In  the  outer 
wall  of  these  lenticular  str^ks  the  felspar  needles  show  a  centripetal 
growth,  the  needles  nuliating  towuds  the  interior  of  the  streak. 
The  rest  of  the  finer-grained  OROUNDiiAsa  is  minutely  vesicular  and 
wtrtokff  with  a  good  deal  of  pale  brown  or  colourless  glass  crowded 
widi  bdonites  and  dust,  with  spherulitic  patches  and  scattered  felspar 
laths  (sanidine),  green  augite  gnnules  and  magnetite  granules.  SouUl 
trachytic  lava-fr^[ments  occur  here  and  there;  they  consist  of  a 
compact,  crystolhne  felt  of  the  three  last  constituents;  one  soch 
iragment  contains  a  phcnooyst  of  green  auoiti. 

t  GttbgUdk  AmhuHgm  m  dm  katik.  £dtidm$t  ti.  jj.  a 


6S  Geology  of  Armmh 

tics  in  the  lavas  near  the  base  of  AlagOt,  which  when 
adive  also  rose  abpve  the  waters  of  a  lake^  in  which 
Dreyssensia  polymorpha  flourished. 

About  200  yards  west  of  our  camp,  near  the  head 
of  the  small  left  tributary  whidi  joins  the  Khixiis 
Stt  just  below  the  ruined  castle,  I  was  able  to  es« 
tablish  the  following  sueoessiox} : — 

X.  Bfown  nnd,  with  shdl^fiagmentt    about  zo  feet 

2.  Greenish  idiite  marl  (No.  43^,  full  of  small  shdls  otDnyssmA 

p^lymarfha^  Ballas,  and  Btihima^  sp.  andct.    about  40  feet 

3.  Sleam  of  nonstone      « about  4  inches 

4<  Hard  fstty  dajr  •••        •••       «••        •••  18  ^ 

5*  Seam  of  ironstone      •••        •••        •«•        •••  about  4     n 

6.  Gravel,  sometimes  cemented  by  iron  oxide  into  a  pudding- 

stone.    In  tiiis  bed  I  found  a  Mode  of  limonite,  with  wdl 

marked  concentric  coats  (No.  41). 

This  whole  series,  amounting  to  50  or  60  feet  in 
thickness,  overlies  the  grey,  columnar  lava  of  the 
I^inis  ravine. 

A  small  gravel-pit,  situated  about  50  yards  north  of 
our  camp,  exhibited  the  following  succession^  at  a 
slightly  higher  level  than  the  above : — 

I.  Sanfty  loam     •••        ...        •«>        •»•        •»•        •••  s«5  feet 

3«  Seam  of  uronstone      »..  •••        •••        •••  s  mches 

3.  Grard  of  lava  pebbles,  some  pumiceous,  up  to  3  inches  in 
diametet       •••        •••        •••       ■%•*       •••        •••  8  mches 

4«  Pale  buff  sand*.  •        •••        •••        •«•        ••«        •••  S     »» 

5.  Very  fine,  laminated  grasvd  of  lava  pdiUes,  about  a  quarter 

of  an  inch  in  diameter       ...        •-        •••  x8  inches 

6.  Pale  buff  sand,  base  pot  vinUe       «*•        •»•        ...      a  feet 

It  must  have  been  in  this  series  of  recent  lake- 
deposits  that  Colonel  J.  M.  Giel^  during  the  Crimean 
war,  found  remains  of  a  fossil  elephant,  now  in  the 
British  Museum.  The  locality  is  described  as  '*  close 
to  Sharvoon,  near  Khanoos'*  ^sKhinis),  but  this 
name  does  not  occur  in  any  existing  map.  Owing 
|o  noticeable  peculiarities  FaJoMier  named  this  species 


Erxerum  to  Khtnis  67 

JElephds   armeniacus^    His  account   of   the  discovery 
runs  as  follows  * : — 

*^  Major  R»  Jones  Gsrdeiii  F.G.S.,  .  .  .  soon  after  .  .  .  proceeded 
to  the  locality  to  make  ftirdier  esqplorations.  The  remains  indicated 
that  the  skeleton  of  the  animal  had  lain  in  the  cliff  of  a  ravine,  about 
s5  feet  in  bright^  the  section  consisting  of  alternate  beds  of  day  and 
ilnviatile  sand»  the  hitter  charged  with  fragments  of  Vrasuna  [evident- 
ly idendcal  with  the  above-mentioned  Dr^tsmsia  folymarfha\.  llie 
Dones  were^  a  very  friable  condition,  and  the  skuU  crushed  and 
llecompbsed ;  but  Major  Garden  was  i^le  to  exhume  some  portions 
of  tusks,  six  and  a  half  inches  in  diameter,  which  in  desiccitlon 
cmmt^ed  to  ^eces.  The  spedmens  presented  by  Colond  Giels  to  the 
national  collection  consist  oS  two  last  upper  mohura  in  fine  preservation 
and  a  portion  of  a  lower  molar,  all  apparenthr  of  the  same  individual 
These  molars  strike  a  practical  eye  at  the  mst  glance  as  presenting 
something  Intennediate  between  the  Mammoth  and  the  existing 
Indian  Elefdiant  .  .  ,  but  more  nearly  allied  to  the  latter  ...  I 
now  entertain  a  strong  suspidon  that  remains  of  Ei^hat  Mrmtniacm  or 
of  a  form  dosdy  alUed^to  it  ooeor  in  Italy." 

Unfortunately  I  was  not  aware  of  this  occurrence 
until  after  my  return  to  England,  for  this  locality 
would  probably  repay  a  detailed  search*  Remains 
of  EUphas  primigenius  have  been  found  in  lake* 
deposits  ci  tne  same  age  at  Alexandropol,  in  Russian 
Anneaia.t 

^  AldtmiotpgUmI  Mmmn^  H«  S46-S49. 

f  Poh%  Vtt^k.  NaturMa.  Vtr.,  Bonn/x886,  SitwrngOtr.  48;  and 
iee  Chapter  IDOV^mi  the  IPfiooene  and  Pleistocette  of  Armenia. 


CHAPTER  VI 

KHINXS  TO  TUTAKH 

We  left  Khinis  at  ncx>n  on  June  26th,  and  our  course 
lay  eastwards  ovfer  the  dusty  plain  of  white,  lacustrine 
marl's,  keeping  the  valliey  of  the  Khinis  river  on  our 
right.  Lava  doesv  not  occur  on*  the  surface  of  the 
plain  east  of  Khinis ;  ^^the  cliffs  of  a  ravine  oq  our 
right  showed  about  150  fe6t  of  nearly  horizontal^ 
c^careous  marls,  overlying  the  dark  cindery  lava  of 
Khinis.  After  an  hour's  ride  we  descended  to  the 
Bingol  Su,  which  we  had  last  seen  at  Kilissaderehl 
Paik  lies  on  the  opposite  bank,  the  low  dwelUngs 
nestling  among  willows  and  oak-scrub;  ;here  the 
alluvial  flats  are  nearly  a  mile  in  width/  At  this 
spot  the  lava-flows  disappear  completely  from  sight 
beneath  the  calcareous  marls  and  clays  of  the  lake* 
4eposits. 

On  approaching  Chevirmeh,  the  south  bank  of  the 
river  shows  a  cliff  of  about  j^o  feet  of  buff  sand, 
resting  on  dark  brown  clavs,  which  throw  oj^t  springs. 
The  strata  dip  only  2^  S.E.  The  Armenian  villagis 
is  situated  at  the  confluence  of  the  Tekhtap  and 
Bingol  rivers;  abimdant  pebbles  of  white  marble 
have  been  brought  down  from  the  Ak  Dagh  by  the 
former  stream. 

At  Yenikoi  we  forded  another  left  tributary  of  the 


Khinis  to  Tutakh  69 

Bingol  Su,  and.  left  the  dusty  lake-marls  for  wide 
terraces  of  travertine  (No.  44),  forming  a  selvage,  3 — 5 
miles  in  width,  along  the  south  foot  of  the  Ak  Dagh. 
The  adlivity  of  the  calcareous  springs  in  former  times 
must  have  been  very  great.  This  travertine  continued 
as  far  as  Dedeviren,  where  the  lake-marls  appeared 
again  for  a  short  distance  from  beneath  the  travertine. 
During  our  ride  over  the  plain  I  could  see  that  the 
marbles  of  the  Ak  Dagh  have  been  folded  rather 
strongly: — the  dip  in  the  southernmost  ridge  was 
50°  S.SJE.,  changing  northwards  to  60°  N.N.W.,  and 
still  further  north  to  40°  S.S.E.  But  these  ob- 
servations only  apply  to  the  most  southern  part  of 
the  mass.  It  was  also  evident,  as  we  proceeded,  that 
t^e  Ak  Dagh  consists  essentially  pf  a  series  of  parallel 
ndges,  with  W.S.W. — E.N.E.  strike,  closely  ranged 
together  in  &:helon. 

As  the  Khinis  plain  gradually  narrows  eastwards, 
it  is  composed  of  travertine  in  the  northern  area, 
and  of  lake-marls  in  the  southern  portion.  At  this 
stage  of  our  ride  the  Zimek  Dagh,  which  bounds 
the  plain  on  the  south,  had  decreased  in  height,  so 
as  to  permit  of  a  clear  view  of  the  jagged  summit 
of  the  volcanic  Bilejan  Dagh  behind  it. 

On  approaching  Gunduz  (8  miles  from  Dedeviren), 
the  marble  heights  of  the  Ak  Dagh  gave  place  to 
much  lower,  rounded  hills  of  hornblende-felspar  rocks, 
yielding  a  greenish  grey  soil.  Although  our  course 
at  first  lay  about  two  miles  to  the  south  of  these 
heights,  the  travertine  plain  was  strewn  with  angular 
blocks  of  the  hornblende-felspar  rocks,  either  dark 
green  and  fine-grained,  or  else  mottled  and  coarser 
in  texture. .  One  of  these  blocks  at  Gunduz  consisted 
of  a  dark  green  hornblende-schist  (No.  45),*  in  layers 

*  The  rock  is  very  fine-grained,  with  a  lustrous  surface,  owing  to 


70  Geology  of  Armenia 

about  half  an  inch  thick.  It  shows  much  similarity 
to  the  homt)lende-felspar  rock  of  Dashkessan  (east 
of  Lake  Gokcha),  described  by  Thost,*  fhis  No,  17). 
Alteration  has  proceeded  considerably  lurther  than 
in  the  Palandoken  epidiorite  (p.  48). 

Here  we  made  a  short  digression  to  the  south  of  our 
route,  in  order  to  visit  Karachoban  on  the  Bingol 
Su,  three  miles  from  Gunduz.  We  rode  due  south, 
down  a  marshy  valley  passing  through  Sarveli.  On 
looking  westwards,  I  could  see  that  the  travertine 
terraces  along  the  foot  of  the  Ak  Dagh  break  off  in 
low  cliffs  to  the  lake-marls  of  the  plain  en  the  south. 
These  lacustrine  beds  also  form  the  low  cliffs  on  the 
south  side  of  the  Bingol  Su,  cloaking  the  flanks  of 
the  Zimek  Dagh.  The  alluvial  Valley  is  here  about 
a  mile  wide,  considerably  out  of  proportion  to  thfe 
present  size  of  the  river. 

Returning  to  Gunduz,  we  proceeded  to  cross  the 
low  line  of  hills,  which  frame  in  the  plain  of  Khinis 
on  the  east.  The  col  (6265  feet)  is  composed  of  a 
hornblende-schist,  similar  to  the  Gunduz  specimen; 
the  soil,  however,  is  here  a  reddish  brown.  In  the 
midst  of  these  dark  green  rocks  we  came  to  a  broad, 

the  minute,  glancing  hornblende  crystals,  which  seem  to  sUghtly 
exceed  in  amotint  the  equally  small,  crystalline  grains  of  fekpar  (less 
than  0.5  mm.).  Under  the  microscope,  the  hornblende  is  green, 
quite  fresh  and  free  from  inclusions ;  allotriomorphic  yet  with  a  ten- 
dency to  prismatic  habit ;  sometimes  twinned ;  compact,  not  fibrous, 
but  the  characteristic  cleavages  are  rarely  distinct,  probably  owing 
to  the  small  size  of  the  grains ;  pleochroism  strong  :  a — ^pale  {[reenish 
straw,  ^r-sap  green,  ^—bluish  green,  c>d>  >a.  The  remainder  of 
the  mosaic  is  formed  by  water-dear  granules  of  secondary  felspar 
(albite),  rarely  with  any  trace  of  turbid  primary  felspar,  which, 
however,  is  indeterminable;  often  containing  hornblende  granules. 
iLMENrrs  is  also  present,  but  no  longer  with  characteristic  outlines ; 
often  bordered  by  granular  lsucoxbne. 

•  Abhandl.  Senckenk.  Naiurf.  Ges.  Frankfurt  a/M,  XVIII,  11.164. 


1 


Khtnis  to  Tutakh  71 

rounded  hill  of  a  pinkish  brown  granitite  (micro- 
granite,  No.  46),*  which  was  evidently  intrusive  in 
the  basic  crystalline  rocks.  Kochf  noticed  an  in- 
trusion of  granite  into  limestone  (of  Miocene  age), 
at  a  spot  a  little  further  north,  between  Gok  Su 
and  Sarveli.  The  granite  )delds  a  pale  brown,  sandy 
soil,  in  contrast  to  the  greenish  grey  or  reddish  brown 
of  the  dark  green,  homblendic  rocks. 

As  we  gradually  descended  to  Gopal  we  traversed 
an  extensive  outcrop  of  a  coarse-grained  quartz- 
diorite  (No.  47),$  which  yields  a  rich,  clayey  soil,  with 

*  The  rock  is  fine-grained ;  the  most  abundant  constituent  is  quartz 
(up  to  I  mm.),  together  with  pale  pink,  nearly  colourless  felspar; 
flaikes  of  biotite  (up  to  i  mm.)  are  sparingly  present,  with  a  few  specks 
of  magnetite.  Under  the  microscope,  ihe  orthoclasb  sometimes 
shows  Carlsbad  twinning,  with  a  central  core  of  dusty  decomposition- 
products;  in  part  contemporaneous  with  the  quartz,  of  which  it 
encloses  granules,  but  there  is  no  definite  pegmatitic  structure.  A  fair 
amount  of  ougoclasb  is  present,  showing  an  approach  to  idiomor- 
phism;  it  is  slightly  dusty  with  decomposition-products.  Biotite 
occurs  in  scattered,  small,  dark  brown  flakes,  showing  in  some  cases 
incipient  alteration,  changing  to  a  green  colour  at  the  edges,  and 
finally  being  replaced  by  dilorite ;  it  encloses  apatite  and  magnetite. 
The  latter  also  occurs  separately  in  a  few  grains.  Quartz  is  very 
abundant  in  interstitial  grains,  with  roughly  parallel  strings  of  very 
fine  inclusions.  The  order  of  crystallization  seems  to  have  been  : 
I.  Apatite  and  magnetite;  2.  Biotite;  3.  Plagioclase;  4.  Orthoclase; 
5.  Quartz,  in  part  contemporary  with  the  last. 

t  J^eise  impontischen  GMrgi^  ii.  354. 

X  The  rock  contains  lustrous  hornblende  crystals  (up  to  7  mm.), 
in  a  matrix  of  dull  white  felspar,  which  forms  about  three-quarters 
of  the  rock ;  there  is  a  little  interstitial  quartz.  Under  the  micro- 
scope, the  felspar  is  exclusively  plagioclase  (andesine  about 
Ab4Aji3),  allotriomorphic  excepting  towards  the  quartz,  to  which  it 
presents  well  marked  crystal-outlines ;  fairly  fresh,  but  with  a  central 
network  of  cracks,  slighdy  dusty  with  decomposition-products ;  zoning 
frequent;  the  albite  twinning  is  rather  close  and  fine,  associated 
with  Carlsbad  and  occasiona^y  with  pericline.  Hornblende  occurs 
in  large,  green,  prismatic  sections,  sometimes  with  pyramidal  ter- 
minations;   in  some  cases  bordered  by  an  incomplete  wreath  of 


1 


72  Geology  of  Armenia 

marshy  patches  between  the  rounded  hills.  The 
broad  lava  plain  of  Gopal  now  lay  open  to  our  view, 
extending  to  north  and  north-east.  The  eastern  spurs 
of  the  /Jc  Dagh  sink  down  to  E.N.E.,  and  do  not 
extend  east  of  Karakevish  and  Gok  Su. 

Gopal  (5643  feet ;  ^  miles  from  Gunduz)  is  situated 
on  a  considerable  tributary  of  the  Bingol  Su;  the 
alluvial  valley  is  here  about  i^  mile  in  width.  The 
river  has  cut  down  its  bed  through  the  plain  so  as 
to  form  cliffs  about  150  feet  high.  The  lower  part 
of  the  cliff  is  composed  of  a  yellow,  imfossiliferous 
sandstone,  in  beds  of  varying  thickness,  dipping  20® 
N.N.W,,  and  overlaid  by  a  thick  sheet  of  an  iron-grey 
ophitic  olivine-basalt  (No.  48).*    Columnar  struAure 

magnetite  granules;  encloses  apatite,  frequent  grains  of  magnetite, 
small  flakes  of  biotite  (partly  chloritized),  and  small  idiomorphs  of 
plagioclase,  with  winch  its  period  of  growth  evidently  overlapped; 
pleochroism  in  greenish  tones,  a — ^pate  greenish  yellow,  4^— dark  sap 
green  with  a  brownish  tinge,  c — ^very  dark  sap  green  with  a  bluish 
tinge,  c>b>a.  Biotits  is  only  sparingly  present,  in  small  yellow 
flakes,  sometimes  greenish  and  partly  chloritized,  nowhere  quite  fresh. 
Apatite,  magnetite,  and  ilmenite  are  accessory.  Leucoxens 
frequently  occurs  bordering  the  ilmenite  or  entirely  replacing  it  in 
grama  of  a  yellowish  tinge  and  slightly  pleochroic  (to  brownish  yellow). 
A  fair  amount  of  interstitial  grains  of  quartz  is  also  present. 
The  order  of  aystallization  seems  to  have  been :  i.  Apatite  and 
magnetite ;  a.  Biotite ;  3.  Hornblende ;  4.  Plagioclase,  partly  over* 
lapping  the  last ;  5.  Quartz.  Several  cracks  traverse  the  slide ;  along 
some  of  them  a  slight  displacement  has  occurred,  attended  by  the 
formation  of  secondary  hornblende,  quartz  and  magnetite. 

*  The  rock  is  compact,  with  a  mottled  appearance,  due  to  alternate 
augitic  and  felspathic  areas  (up  to  2  mm.)  A  few  small  steam-cavities 
(up  to  3  mm.)  are  present  m  the  specimen.  Under  the  microscope* 
OLIVINE  occurs  in  frequent  idiomorphs,  sinking  down  to  rather  sniall 
dimensions;  clear  and  fresh,  but  often  with  marginal  orange  stains 
of  iron  oxide,  and  occasionally  showing  a  lateral  intrusion  of  the 
groundmass.  Augite  in  large,  pinkish  brown,  ophitic  plates,  varying 
m  depth  of  tint,  sometimes  showing  signs  of  bending  and  strain* 
doubtless  owing  to  some  diffierential  movement  prior  to  final  con? 
solidation ;  it  encloses  laths  of  plagioclase,  with  fluidal  arrangement ; 


Khinis  to  Tutdkh  73 

is  frequent  in  tlie  lower  part  of  the  lava  sheet,  and 
is  o(xasionaIly  accompanied  by  spheroidal  weathering. 
Nearer  the  surface  the  basalt  becomes  scoriaceous, 
with  well-marked  flow-strudhire,  and  breaks  up  into 
fragments  like  tiles.  This  lava  does  not  extend  far 
west  of  Gopal. 

The  low  hills  between  this  place  and  the  Ak  Dagb 
are  composed  of  a  white  OrWtofties-limestone,  dipping 
40^  N,N.W.,  and  apparently  overlying  the  sandstone. 
Koch  *  has  recorded  the  occurrence  or  a  similar  lime- 
stone, "praAically  consisting  only  of  foraminifera/' 
a  few  miles  further  up  this  valley,  from  Gok  Su  as 
far  as  the  pass  over  the  Kazbel  Gedik,  and  I  have 
BO  doubt  that  this  limestone  was  contemporaneous 
and  in  continuity  with  that  of  Madrak  on  the  west 
(p.  52),  and  with  that  of  Al  Khes  on  the  north-east 
(vide  tnfra)y  both  of  Lower  Miocene  age. 

During  the  morning  of  June  28th  we  again  made 
a  detour  to  the  south,  in  order  to  trace  the  course 
of  the  Gopal  river  to  its  confluence  with  the  Bingol 
Su.  On  riding  up  to  the  surface  of  the  lava  plain, 
it  became  evident  that  the  long,  white  downs,  which 
boimd  it  on  the  east,  are  formed  of  the  same  Miocene 
limestone.  The  basalt  of  the  eastern  part  of  the  plain 
sinks  beneath  horizontal  marls,  similar  to  the  lake- 
deposits  of  the  Khinis  plain ;  these  are  in  turn  covered 
up  by  travertine,  in  terraces  along  the  base  of  the 
limestone  downs. 

As  we  rode  southwards  over  the  narrowing  plain, 

our  view  to  the  west  was  shut  in  by  the  line  of 

low  hills  of   granite  and  hornblende-felspar   rocks, 

the  larger  febpais  are  moulded  on  olivine,  and  show  extinction-angles 
pointing  to  an  add  labra2X>rits,  about  AbiAni.  Magnetite  in 
granules  and  needles  is  abundant,  and  together  with  augite  grains 
forms  the  interstitial  nuitter. 

*  I^eiu  impofUischim  GMrgi^  ii.  348. 


74 


1 

t 

Geology  of  Armenia  | 


which  we  had  crossed   on  the  previous  day.    The  -n 

southern  limit  of  these  igneous  rocks  is  formed  by  — ' 

a  high  escarpment  of  grey  limestone,  the  Gozel  Baba 
Dagh,  at  the  foot  of  which  the  Bingol  Su  flows.    The  _J 

strata  dip  45*^  N.W.;  the  bold  mass  was  evidently 
once  continuous  with  the  Zimek  Dagh,  and  owes  JJ 

its  present  isolation  to  erosion  by  the  Bingol  Su. 
This  river,  in  passing  from  the  lacustrine  plain  of  H 

Khinis  to  that  of  Upper  Bulanik,  has  cut  a  deep, 
rocky  gorge  in  a  zigzag  through  the  zone  of  Miocene  --^ 

limestone,  which  stretches  from  the  Zimek  Dagh  to  -^ 

the  N.E,  into  the  Lala  Dagh  (Tulip  Moimtain),  where 
it  becomes  buried  again  beneath  considerable  sheets 
of  lava.  Koch  crossed  th6  eastern  part  of  the  Zimek 
Dagh  (here  called  Khamshujur),  and  found  it  to  be 
composed  of  a  "yellowish  white  limestone,  probably 
foraminiferal,  which  here  and  there  alternates  with 
a  fine-grained  sandstone.*"  In  fadl  we  soon  left  the 
basalt  of  Gopal  behind  us,  and  traversed  the  im- 
dulating  downs  of  a  white,  chalky  or  pale  buff  lime- 
stone full  of  foraminifera,  especially  of  Orbitotdes^ 
as  at  Al  Khes  (vide  infra);  a  rather  hasty  search 
revealed  nothing  larger  than  a  fragment  of  a  small 
Pe6ten.'\  The  dip  is  30"*  N.W.,  forming  a  shallow 
syncline  with  the  strata  of  the  Gozel  Baba  escarpment 
to  the  west  of  our  route.  On  tbfe  east  side  of  the 
Gopal  river,  a  higher  ridge  of  ^is  limestone,  dipping 
20°  S.S.E.,  set  a  limit  to  our  view  eastwards,  both 
here  and  on  the  Gopal  plain. 

^^hi  the  afternoon  we  resumed  our  journey  to  Tutakh, 
ascending  to  the  basalt  plain  above  Gopal.  Its  level 
extent  is  only  interrupted  by  long,  parallel  mounds 
(running  W.N.W. — E.S.E.)  of  lava  boulders,  highly 

*  Op.  di.^  ii.  36a. 

t  This  specimen  was  unfortunately  mislaid  in  transit. 


"K  1 


•W«,^,^,^=*«H 


la- 


a|&>s--s-.s.':|r:s%-s^i : 


-a.  .j(».j(.  .j(.  .j(.  .j(.  .j(.  .jj[.  .jj[.  .^. 


•*^i;ff|Hinii:|Hir 


.^■-  -   -   ■-  -Ji'-a- 


■:^f::f::f::f:^rT^r:|:"i 


■»t-^* 


I  ^ 


^  -I 


^^VkkS 


Khinis  to  Tutakh  75 

scoriaceous,  which  probably  mark  the  original  fissures 
whence  the  lava  issued.  The  largest  of  these  mounds 
appeared  to  be  about  2  miles  long,  200  yards  wide, 
and  30-40  feet  high.  To  the  east  of  our  route  the 
basalt  was  covered  by  travertine. 

We  reached  the  northern  boimdary  of  the  plain 
at  Ali  Kulek,  where  a  spur  from  the  Kazbel  Gedik 
rises  out  of  the  lava.  It  is  formed  of  the  same  white, 
chalky  limestone  (presumably  Miocene)  which  I  had 
observed  near  Gopal ;  but  it  has  suffered  very  greatly 
from  the  intrusion  of  basic  and  ultrabasic  rocks  along 
the  strike,  just  as  I  had  noticed  between  Palandoken 
and  Madrak  (p.  51).  The  contadl-metamorphism  has 
altered  the  friable  limestone  into  a  grey  or  purple, 
brecciated  marble,  traversed  by  veins  of  white  calcite. 
The  dip  of  these  strata  next  the  plain  is  40°  S.S.E., 
soon  clianging  northwards  to  35  N.N.W.,  as  seen 
in  the  southernmost  marmorized  outspur. 

Ali  Kulek  lies  on  a  gentle  slope  of  the  unaltered 
limestone,  and  the  houses  are  built  of  this  dazzling 
white,  chalky  rock.  Rashan,  the  next  village,  lies 
about  a  mile  further  north,  in  a  broad  upland  valley 
boimded  on  the  west  by  peaks  of  limestone  and 
marble,  on  the  east  by  dark  serpentinous  rocks.  The 
adhial  junction  was  concealed  by  a  terrace  of  traver- 
tine. Large  blocks  of  dark  grey  or  purple  marble, 
streaked  with  veins  of  calcite,  frequently  encumbered 
the  grassy  mountain -pastures.  On  continuing  our 
ascent  up  the  valley  to  N.W.,  the  metamorphism 
of  the  limestone  by  the  serpentinous  rocks  became 
Still  more  marked. 

A  specimen  (No.  49)  of  almost  unaltered  limestone 
is  pale  buff  and  concretionary,  with  splintery  fradlure. 
The  microscope  reveals  remains  of  Lithothamnion^ 
together  with  bryozoa  and  foraminifera  (MilioUna^ 
and  Operculinajj  similar  to  those  occurring  in  the 


76  Geology  of  Armenia 

undoubtedly  Miocene  Orbitoides'limestone  of  Al  Khes, 
about  12  miles  to  the  east  [vide  infra].  My  specimen 
also  contains  a  well-rounded  pebole,  an  inch  and  a 
quarter  in  diameter,  of  a  dark  green,  much  altered 
olivine-diabase,  of  a  kind  which  I  nowhere  found 
in  situ  during  our  journey. 

As  the  valley  narrowed,  the  dip  of  the  limestone 
changed  to  S.W.,  and  became  nearly  vertical.  The 
head  of  the  valley  is  dominated  bv  a  high  cliff,  the 
break-off  of  an  extensive  sheet  of  lava,  which  has 
overwhelmed  the  limestone  downs  and  the  rounded 
hills  of  intrusive  gabbro  and  serpentine.  A  zigzag 
path  up  this  cliff  led  us  up  to  the  plateau  of  lava, 
which  rises  in  terraces  to  the  N.W.  As  on.  the  Gopal 
plain,  ridges  (running  from  W.N.W.  to  E.S.E.)  of 
scoriaceous  blocks  probably  indicate  the  original 
fissures  of  emission.  The  rock  (No.  50)  ^  is  an  ophitic 
olivine-basalt,  scoriaceous  in  places.  The  blocks  have 
a  black,  varnish-like  surface,  and  the  rock  weathers 
to  a  rich  brown,  sandy  soil.  The  similarity  of  the 
lava    to    the    Gopal    basalt    (especially    under    the 

*  The  rock  is  compact,  a  dlEurk  green  with  a  slight  greenish  tinge, 
and  with  a  mottled  appearance  due  to  alternate  augitic  and  felspathic 
areas  (up  to  i  mm.);  some  still  smaller  yellow-green  particles  (olivine) 
and  spedks  of  magnetite  can  be  discerned  with  the  lens.  Under  the 
microscope,  olivins  occurs  in  frequent  idiomorphs  of  a  fair  size, 
and  also  in  smaller  grains,  often  in  aggregates  as  if  a  laiger  oystal 
had  broken  up  during  the  final  consolidation  of  the  lava;  it  is 
clear  and  fresh,  except  for  stains  of  iron  oxide.  Augite  occurs 
in  pinkish  brown,  ophitic  plates;  quite  fresh,  and  only  seldom 
enclosing  magnetite;  it  contains  plagioclase  (labradorite  about 
Ab5An6),  mostly  in  laths  with  fluidal  arrangement,  but  a  large  tabular 
phenocryst  occurs,  showing  faint  zoning,  and  with  netlike  inclusions  of 
magnetite  and  augite.  Magnetite  is  fairly  abundant,  with  a  ten- 
dency to  congregate  in  patches;  it  is  particularly  associated  with 
the  olivine ;  it  has  weathered  in  places  to  hydrated  iron  oxide.  The 
sparing  interstitial  matter  consists  of  grains  pf  augite  and  magnetite. 
A  few  small  irregular  steam-cavities  are  present,  sometimes  filled 
by  a  ZEOLITE  (natroute),  in  a  dusky  fan-aggr^ate. 


Khinis  to  Tutakh  77 

microscope)  suggests  that  both  lavas  have  proceeded 
from  the  same  magma. 

Small  peaks  of  boldly- jointed  hypersthene-gabbro 
(No.  51)*  rise  here  and  there  like  nunatakr  out  of 
the  desolate  lava  sheet.  Dark  segregation-bands  are 
sometimes  visible.  The  weathered  surface  of  the  rock 
presents  a  pitted  appearance  owing  to  the  ready 
decomposition  of  the  serpentinized  olivine,  and  the 
rock  disintegrates  into  isolated  felspar  grains.    This 

*  About  two-thirds  of  the  coarsely  OTitalline  rock  oontistB  of  a  pale 
bluish  grey,  opalescent,  finely  striated  felspar,  in  crjrstalline  grains  (up 
to  6  mm.) ;  the  rest  is  composed  of  serpentine  grains  (up  to  3  mm.), 
and  small,  darii  green  specks  (hornblende),  whuch  frequently  enclose 
particles  with  a  bronxe  lustre  (hypersthene).  Under  Uie  microscope 
the  PLAGiocLASK  (bytowmitb  about  AbiAn3)  occurs  in  abundant 
aJlotriomorphic  grains  ;  albite  twinning  sometimes  combined  wiUi  peri- 
dine  ;  dusty  decomposition>poducts  occur  aloitf[  the  numerous  cracks, 
radiating  6x>m  the  serpentinized  olivine  of  which  it  sometimes  encloses 
small  particles.  Olivine  is  present  in  rather  small  iprains,  partly 
or  wholly  altered  to  serpentine,  with  separated  magnetite  along  the 
cracks ;  but,  whe^er  fresh  or  altered,  it  is  always  separated  from 
the  felspar  by  a  reaction-rim  of  green  hornblende,  and  sometimes 
there  is  an  inner  rim  of  hypersthene.  HvPUtsTHENK  is  rather  more 
abundant  than  the  olivine,  occurring  in  pale  pink,  rounded  grains 
showing  the  usual  pleochroism  (a — salmon  pink,  ^--Hreddish  yellow, 
c — pale  bluish  green).  Augitb  in  pale  green  grains  is  less  frequent. 
Both  pyroxenes  are  associated  together  and  may  be  either  with  or 
without  a  reaction-rim  of  green  hornblende.  Some  of  the  horn- 
blende is  present  in  pale  brown,  compact  grains,  with  strong  pleo- 
chroism (a — very  pale  greenish  straw,  ^--pale  purplish  brown, 
r— purplish  or  reddish  browi^);  usually  containing  a  core  of  hyper- 
sthene, as  if  derived  from  it;  when  occurring  as  a  reaction-rim  it 
is  pale  green,  with  the  following  pleochroism :  a — very  pale  greenish 
yellow,  ^ — ^yellow-^een,  c — ^bluish  'green.  Magnetite  is  only  sparing- 
ly present,  sometimes  with  a  reaction-rim.  A  very  few  grains  of 
a  blackish  brown  spinel,  probably  picotite,  are  present.  There 
appear  to  be  some  indications  of  a  slight  amount  of  dynamic  meta- 
morphism,  m.,  (i)  the  reaction-rims  of  green  hornblende;  (2)  some 
secondaiy  twinning  and  undulose  extinction  in  some  of  the  felspars, 
which  are  all  much  cracked ;  (3)  a  granulation  of  hypersthene  and 
augite  along  a  plane  of  weakness,  with  the  formation,  locally,  of  a 
little  secondary  felspar. 


78  Geology  of  Armenia 

coarsely  crystalline,  basic  rock  seems  to  occupy  a 
central  position  in  the  zone  of  serpentines. 

We  soon  left  the  lava  plateau  to  descend  into  broad 
open  country — the  district  of  Elmali-dereh  (valley  of 
apples),  so  named  from  the  scattered  trees  of  Pyrus 
salicifolia  and  P.  elceagnifolia^  with  silvery  leaves.  A 
backward  glance  showed  that  the  lava  sheet  forms 
a  cliff  about  30  feet  high,  with  rudely  columnar 
strudlure;  we  were  now  traversing  an  outcrop  of  a 
massive  serpentine  (No.  52),^  which  had  been  in 
part  concealed  by  the  basalt. , 

Here,  at  a  Kurdish  encampment,  we  found  we 
had  come  too  far  north,  and  from  this  point  we 
turned  away  to  the  east,  crossing  a  broad  valley 
running  from  north  to  south.  Isolated,  rugged  peaks 
of  grey  marble  rise  like  islands  out  of  the  enveloping 
serpentine,  and  show  a  dip  of  40®  S.S.E.,  changing 
northwards  to  20®  N.N.W.  Blocks  of  serpentinous 
marble,  which  had  rolled  down  from  the  contadl-zone, 
showed  an  intimate  penetration  of  marble  by  the 
serpentine.  At  one  place  an  intrusive  mass  of  a  much 
weathered   and  altered  diabase  (No.  53)!  has  ap- 

*  The  dark  green  rock  is  much  veined  by  paler  seipentine  with 
strings  of  magnetite.  The  rock  is  slightly  magnetic  (4'').  Lustrous, 
pale  brown,  crystalline  grains  of  bastite  (up  to  6  mm.)  are  fairly 
frequent.  Under  the  microscope,  the  bastite  occurs  in  fibrous, 
lenticular  grains,  often  traversed  by  thin  veins  of  serpentine.  The 
serpentine  of  the  groundmass  is  clearly  after  olivine,  with  the 
characteristic  mesh-structure;  the  meshes  of  the  network  are  nearly 
isotropic,  and  paler  in  colour  than  the  network.  Some  tiny  granules 
of  dark  brown  chromite  are  present.  The  rock  was  originally  an 
olivine-enstatite-rock,  and  closely  resembles  the  serpentines  from 
Palandoken  (No.  24,  p.  48),  and  from  the  south  coast  of  Lake 
Van  (No.  134,  Chapter  VIII). 

t  The  rock  is  compact  and  crystalline,  and  is  traversed  by  veins 
of  epidote.  Abundant,  greenish  white  felspars  (up  to  2  mm.)  and 
less  numerous,  very  dark  green,  chloritic  patches  (up  to  2  mm.) 
lie  in  a  finer-grained,  greenish  grey  groundmass.    Under  the  micro- 


Khinis  to  Tutakh  79 

parently  risen  up  through  the  serpentine.  A  very 
similar  rock,  also  associated  with  serpentine,  has  been 
described  by  Becke^  from  the  Shatin  Dagh,  from 
above  Kagizman,  22  miles  north  of  this  locality. 

The  relatively  lower  country  of  the  serpentine 
district  ceases  at  Merg^  Khel,  a  small  village  at  the 
N.E.  end  of  an  alluvial  plain.  It  lies  at  the  foot 
of  a  cliff  of  dark  green  serpentine  intrusive  into  a 
limestone  (presumably  Lower  Miocene),  dipping  60® 
N.N.W.,  which  it  has  altered  to  a  grey  marble.  This 
cliff  is  capped  by  a  lava  sheet.  On  looking  up  the 
narrow  valley  to  the  N.E.,  the  grey  marble  was  seen 
to  predominate,  but  both  banks  are  capped  by  lava. 
We  proceeded  for  a  short  distance  along  this  valley, 
and  then  turned  eastwards  up  a  lateral  ravine, 
which  revealed  a  shallow  syncline  of  white,  chalky, 
imfossiliferous  limestone. 

Finally  we  emerged  once  again  on  the  summit 
of  a  wide,  open  plateau  of  basalt,  greatly  encumbered 
by  large  blocks  and  boulders.  Here,  as  elsewhere 
on  the  lava  plateaux  of  Armenia,  a  tall,  yellow  fennel 
(Prangos  foeniculacea)  grows  abundantly  and  is  the 
diaraderistic  plant  of  these  sombre,  rocky  steppes. 
Koch's  description  f  of  the  basalt  steppe  of  Karayazi, 

scope,  the  plagioclasb  (andesinb  about  AbjAna)  occurs  in  idio- 
morphs  of  prismatic  habit,  quite  turbid  with  decomposition-products, 
which  often  wholly  obscure  the  albite  twinning ;  it  frequently  contains 
grains  of  epidote,  especially  centrally.  Augite  is  fairly  abundant; 
allotriomorphic  and  nearly  colourless,  but  greatly  obscured  by  epidote, 
and  in  some  cases  altered  to  chlorite,  whilst  still  ophiticadly  en- 
closing small  crystals  of  plagioclase.  Magnetite  occurs  in  a  few 
iTTeguktf  grains  bordered  by  limonite.  Iron  pyrites  is  also  accessory. 
Epidote  is  abundantly  present^  so  to  speak  saturating  thi  rock, 
and  also  forming  narrow  veins. 

*  Abichy  Geohgischi  Forschungen  in  den  kauk.  Landem^  ii.  329. 
t  Rase  imponthchm  GeUrge^  ii.  345-347. 


8o  Geology  of  Armenia 

N.W.  of  the  Kazbel  Gedik,  shows  that  on  both  sided 
of  this  hill-ridge  there  has  been  a  vast  outpouring 
of  basalt  by  nssure-eruptions  comparable  to  those 
of  Iceland. 

As  we  rode  along  I  found  a  half  of  what  appeared 
to  be  a  volcanic  bomb  of  an  ophitic  olivine-basalt 
(No.  54).*  The  longitudinal  sedtion  is  slightly  el- 
liptical (12x8  inches),  and  clearly  shows  a  gradual 
transition  from  a  smooth,  fine-grained  cortex  to  a 
coarsely  crystalline  core.  The  microscope  shows  that 
the  cortex  is  very  similar  to  the  basalt  of  the  Gopal 
plain  and  of  the  Rashan  plateau. 

A  troublesome  but  gradual  descent  over  the  lava 
terraces,  strewn  with  large  boulders,  brought  us  to 
Kanjean,  situated  on  a  considerable  stream  flowing 
E.S.E.;  the  village  lies  at  the  southern  base  of  a  high, 
rounded  hill  of  serpentine,  intrusive^  into  Miocene 
limestone. 

From  Kanjean  we  pursued  our  course  to  S.E.; 
before  us  the  straight  lines  of  the  lava  plateau  framed 
in  the  wide,  open  valley  along  the  eastern  horizoQ. 
Before  ascending  again  to  this  plateau  we  passed  over 
several  low  ridges  of  a  concretionary  limestone  (No. 

*  The  spedxnen  is  taken  from  the  cortex,  in  which  olivine  occurs  in 
frequent  crystals  (up  to  2  nun.),  stained  with  iron  oxide;  augite, 
on  the  other  hand,  forms  wdl-developed  crystals  in  the  core,  pro- 
jecting freely  into  small  cavities,  which  have  been  partially  filled  up  by 
aragonite  in  fan-like,  radiating  crystals.  The  section  was  taken 
from  the  cortex.  Under  die  microscope,  olivine  is  present  in  fre- 
quent laige  idiomorphs ;  very  pale  yellow,  and  often  with  the  margin 
deeply  stained  orange  brown  by  iron  oxide,  and  occasionally  intruded 
by  groundmass ;  sometimes  in  aggregates,  and  sinking  down  to  rather 
smsdl  dimensions';  frequently  containing  minute  brown  crystals  of 
picoTiTK.  Augite  occurs  in  large,  pinkish  brown,  ophitic  plates* 
enclosing  laths  of  plagioclase  (an  acid  labradoritb  about  AbiAni), 
with  fluidal  arrangement  Grains  of  magnetite  and  augite  are  inter- 
stitial. Irregular  steam-cavities  are  not  ii^requent,  and  are  sometimes 
lined  by  a  zeolite. 


"H 


.1 


t^    i      ■      It 


visit'* 


V 


/ 


K*' 


Khinis  to  Tutakh  8i 

55))  dipping  20^  S.S.£m  it  is  full  of  organic  fragments 
^chiefly  Bryozoa)  in  a  pale  buff  matrix  with  a 
pinki^  tinge,  somewhat  similar  to  the  Rashan 
limestone  (p.  75),  of  which  it  probably  represents 
the  fr^  and  unaltered  form.  The  microscope  shows 
it  to  contain : — 

Numerous  Bryozoa  (mainly  BeteroporidiB) 

Ceriopora  anomala^  Abich. 

Amphistegina  cf.  Hauerina^  d'Orb. 

Litkothamnion  ramosstssitnunij  Reuss. 
These  species  are  identical  with  those  which  I  have 
already  recorded  in  the  Lower  Miocene  (Tortonian) 
of  Madrak  (p.  5a). 

On  reachmg  the  summit  of  the  basalt  plateau  I 
could  see  that  the  grassy,  undulating  plain  stretches 
far  away  to  the  north  and  south,  and  is  only  inter- 
rupted by  long,  low  ridges  (with  N.N.W. — S.S.E.  axis) 
of  scoriaceous  lava-blocks,  just  as  I  had  already 
noticed  in  the  similar  lava  plains  to  the  S.W.  (pp.  75 
&  j6).  This  is  doubtless  part  of  the  great  lava-steppe 
of  Torlu,  which  Koch  traversed  a  little  further  south, 
between  Gargalik  and  Karajviren,  where  he  found  the 
black  lava  similarly  overlying  a  "  Tertiary  limestone, 
with  abundant  foraminifera  and  fossils  like  serpulae."^ 
The  canyons^  which  have  been  carved  out  of  this 
lava-plateau  by  the  tributaries  of  the  Murad,  reveal 
the  underlying  Miocene  limestone,  dipping  45°  S.  by 
W.,  intruded  by  serpentine.  Finally  we  descended 
into  the  deep  valley  of  the  Kersik,  the  largest  river 
of  this  distriA,  and,  after  fording  the  swift  stream,  we 
camped  at  Al  Khes  (5810  feet),  on  a  terrace  halfway 
up  the  left  bank.  The  river  here  comes  from  the 
N.W.,  and  makes  an  abrupt  bend  to  the  south, 
cutting  a  fine  natural  section  through  the  limestone 
and  serpentine.    The  latter  reaches  a  higher   level 

*  JUu  impofUisehcH  GtUrgt^  vL  437. 


82  Geology  of  Artnenia 

on  the  right  than  on  the  left  bank.    The  succession 
down  to  the  river-bed  is  as  follows : — 

i«  Olivine-basalt        3o-S<>  feet 

2.  A  hemicrystaUine,   buff  limestone,  with  an  occasional  thin 

weather-crust  of  pale  brown  calcite,  as  at  Madrak ;  no  fossils 
visible.  It  weathers  in  a  cavernous  manner  along  vertical 
joints,  with  the  same  chan^cteristic  appearance  as  in  die 
Miocene  limestone  of  Adeljivas  (Lake  Van)  and  of  the  Gozd 
Baba  gorge  of  the  Bingol  Su about  30  feet 

3.  A  softer,  white  foraminiferal  limestone  (No.   56),   showing 

under  the  microscope  the  following  fossils : — 
Orbitoides  {Lepidocydina). 
AmphisUgina  cf,  Hauerina^  d'Orb. 
Ceriopara  anotnaUi^  Abicb. 
Biyozoa  {JSeteropmda), 
Lithothamnicn  ramosHsdmum^  Reuss.    about  80  feet 

4.  This  limestone  becomes  gradually  tliinly  bedded,  and  the 

foraminifera  less  abundant,   but  the   OrUtddcs  is  still  a 
marked  constituent  of  the  rock about  60  feet 

5.  Contact-zone.    Here  the  limestone  has  been  altered  to  a  grey 

marble,  with  veins  of  white  calcite*     •••        about  20  feet 

6.  Dark  green  seri>entine.     »••        ...        -«       about  300  feet 

Total  thickness  about  540  feet 

The  fossils  m  this  Orfcttoides-limestone  are  identical 
with  those  of  the  Madrak  and  Kanjean  limestones 
CpP-  52  &  81),  and  point  to  the  same  horizon  of  the 
Lower  Miocene  (Tortonian).  In  this  central  region, 
however,  the  conditions  appear  to  have  been  un- 
favourable to  the  existence  of  molluscs  and  echino- 
derms;  but  the  northern  and  southern  distridls  of 
this  Miocene  belt,  e^.,  at  Ashkala  and  Lake  Urmi 
respe(5lively,  show  a  profusion  of  these  forms  on  the 
same  horizon  (see  Chapter  XXIII). 

On  looking  up  the  Kersik  valley  to  the  N.W.,  a 
high  cliff  of  the  same  limestone  was  conspicuous 
on  the  left  bank,  dipping  20°  N.  by  E.  The  lava 
terraces  capping  the  right  bank  seemed  to  be  absent 
from  the  left  side  of  9ie  valley;  but,  on  reaching 
the  summit,  I  could  see  that  the  limestone  to  the 


Khinis  U  Tutakh  83 

^outh  of  our  route  to  Tutakh  is  soon  covered  by  th6 
lava  sheet  rendered  conspicuous  by  golden  fennel. 

Before  reaching  Bishi  we  crossed  two  low  limestone 
ridges;  the  dip  of  the  strata  in  the  former  is  20^ 
S.S.Wm  and  in  the  latter  oq^  N.N.E.  Since  leaving 
Gqpal,  the  strike  had  gradually  changed  from  E.N.E. 
to  E.S.E.,  showing  that  the  mstridl  of  Elmali-dereh 
is  in  the  region  of  the  bending  xoimd  of  th^  mountain- 
folds  of  the  Armenian  plateaxu  The  rather  shallow, 
longitudinal  valleys  widen^  outwards  towards  the 
Murady  and  have  been  partially  filled  with  a  black, 
peaty  earth.  Remains  of  the  lava-capping  still  occur 
here  and  there,  especially  between  Bishi  and  Kheveh. 
It  was  only  at  the  latter^  place  that  we  entered  the 
broad  plain  of  Tutakh,  and  first  obtained  a  full  view 
of  Sipan,  interrupted  only  to  a  sli^t  d^ee  by  the 
relatively  low  dome  of  the  Kartevin  Dagh.  The  sharp 
crest  of  Bilejan  was  visible  in  the  south-west.  The 
plain  extends  far  to  the  east  and  south-east  of  Sipan. 
The  soil  is  a  rich  dark  brown,  derived  from  the 
decomposition  of  the  basaltic  lava,  which  has  evident- 
ly flowed  to  the  south-east,  down  the  valleys,  which 
had  previously  been  carved  out  of  the  gently  sloping 
plain. 

The  last  trace  of  the  underl3dng  Miocene  strata 
(dipping  about  30*^  S.S.W^  was  revealed  in  a  low 
river-cliff,  a  little  west  of  Kali.  Our  route  lay  some 
distance  south  of  this  outcrop,  but  a  rolled  block 
of  the  pinkish,  crystalline  limestone  (No.  57)  contains 
abundant  foramimfera,  chiefly  Amphtstegina, 

Just  before  reaching  Buruzko  I  noticed  that  the 
lava  flows  are  overlaid  by  white  and  grey,  sandy  tuffs, 
showing  a  very  slight  dip  (5*^)  to  N.N.E.  The  river 
at  Buruzko  was  the  merest  trickle  at  this  time  of  year 
rJune  30th),  but  in  early  spring  the  valley  must  be 
tilled  by  a  considerable  volume  of  water,  for  we  met 


' 


84  Geology  of  Armenia 

with  large  numbers  of  water*wom  boulders  as  we 
ascended  again  to  the  level  of  the  plain.  The  con* 
tinuity  of  the  basaltic  lava  was  occasionally  inter* 
nipted  by  white,  dusty  sands,  which  gradually 
predominated  as  we  proceeded  eastwaids.  A  deep 
ravine,  which  we  crossed  shortly  before  Tadekan* 
revealed  a  considerable  thickness  of  volcanic  breccias 
and  sands,  nearly  horizontal  (3^  S.E.) ;  no  lava  flows 
were  visible  in  the  cliff *se(%ion.  The  deposits  change 
rapidly  and  frequently  in  chara<fter,  from  fine  sands  to 
coarse  breccias.  The  thinner,  laminated  beds  are  in 
places  contorted  and  bent  down  bv  large,  angular 
blocks  of  scoriaceous  lava  in  the  overlying  bed — a  fadi 
which  clearly  points  to  violent  volcanic  eruption 
having  taken  place  contemporaneously,  and  at  no 
great  distance.  On  the  east  side  of  the  marshy  stream 
a  small  fault  was  visible  with  a  downthtow  of  3  feet 
to  the  south.  I  could  not  detedt  any  fossils  in  these 
strata  during  our  rather  hasty  traverse  of  the  ravine 
in  the  failing  light,  and  it  was  already  dark  when 
we  reached  Tutakh  on  the  Murad. 


CHAPTER  VII 


TUTAKH  TO  AKHLAT 


TuTAKH  (5240  feet)  boasts  of  a  few  stone  houses  built 
of  a  pink  trachytic  lava  of  dndery  texture,  with 
conspicuous  flow-strucfture,  well  defined  by  scoriaceous 
streaks ;  glassy  felspars  and  dark  green  augite  crystals 
are  fairly  numerous.  This  rock  does  not  occur  in  situ 
near  the  village ;  it  had  probably  been  brought  from 
the  western  spurs  of  the  Ala  Dagh. 

Tutakh  lies  on  an  old  alluvial  terrace  of  the  Murad, 
which  has  now  cut  its  bed  down  to  a  lower  base-level. 
A  low  cliff,  forming  the  right  bank  north  of  the 
village,  exposed  about  20  feet  of  gravels  and  sands, 
in  which  current-bedding  was  well  shown.  The 
sedlion  lies  at  present  about  50  feet  above  the  present 
level  of  the  river : — 

I.  Gravels  and  sands,  current-bedded     about  10  feet 

a.  Buff  sands,  with  a  layer  of  small  pebbles  of  lava  cemented 

by  iron  oxide  ...        about  12  feet 

3.  Buff  sandy  clay,  base  not  seen. 

The  pebbles  and  sand  are  entirely  derived  from 
lavas,  and  I  could  fiiid  no  trace  of  any  fossils.  It 
is  not  improbable  that  these  old  alluvial  beds  were 
formed  by  the  Murad  before  it  had  cut  a  gorge  for 
itself  through  the  basalt  ridge  which  bounds  the  small 
plain  of  Tutakh  on  the  south.    At  the  present  time 


86  Geology  of  Armenia 

the  Murad,  like  most  of  the  plateau-rivers,  seems  to 
have  nearly  lost  the  power  of  deepening  its  bed,  for 
there  is  only  a  difference  in  level  of  66  feet  between 
Tutakh  ana  Melaskert — ^a  distance  of  32  miles  in  a 
straight  line. 

On  continuing  our  journey  down  the  Murad  valley 
as  far  as  Mels^kert,  we  forded  the  river,  which, 
according  to  our  measurements,  is  here  100  yards  in 
width.  Three  river- terraces  are  noticeable  on  the 
steep  left  bank,  at  heights  of  about  15,  40,  and  50  feet 
above  the  present  level  of  the  Murad ;  the  two  upper 
terraces  are  cut  out  of  basalt,  with  grassy  cliffs. 

At  Gultepeh  we  turned  slightly  aside  from  the  river, 
rising  about  200  feet  up  to  another  lava-terrace.  We 
soon  crossed  a  ravine  carved  out  of  the  basalt,  through 
which  the  Murad  also  cuts  about  a  mile  further  west. 
Here  the  olivine-basalt  (No.  58)  *  forms  columns  up  to 

*  The  rock  is  greenish  black,  very  compact  and  fine-grained,  with  a 
very  few  steam-vesicles  (up  to  6  mm.),  now  filled  with  calcite.  Some 
striated  felspars  (up  to  3  mm.)  and  fairly  numerous  augites  (up  to 
I  mm.)  are  visible,  with  a  good  many  Uttle  brown  specks  (altered 
olivines).  Under  the  microscope,  the  plagioclase  (a  medium  labra- 
DORiTE  about  Ab5An6)  occurs  in  large,  tabular  idiomorphs,  so  much 
honeycombed  with  a  network  of  groundmass  that  only  a  narrow 
border  of  pure  felspar  is  Idt,  and  they  are  partially  replaced  by 
analdme;  idiomorphs  of  prismatic  habit,  dear  and  free  from  in- 
elusions  are  more  numerous.  Augitb  is  present  in  laige  idiomorphs, 
very  pale  green,  almost  colourless,  sometimes  greener  centally;  it 
is  evidently  undeigoing  indpient  decomposition,  often  with  a  honey- 
combed appearance,  with  tiny  granules  of  epidote  and  serpentinous 
mineral,  especially  along  broad  zones  bordering  cracks.  Olivine 
idiomorphs  are  less  abundant  and  usually  smaller  than  the  augite, 
down  to  quite  small  dimensions ;  partly  serpeiitinized  along  cracks  and 
round  the  margins ;  no  inclusions.  The  groxtndmass  consists  of  a  felt 
of  felspar  laths  (labradorite  about  Ab5An6)  with  ill-defined  fluidal 
arrangement,  with  granules  of  augite  and  magnetite ;  the  latter  also 
occurs  (but  more  rarely)  in  skeletal  crystals  in  die  "  gridiron  "  arrange- 
ment ;  a  little  interstitial,  dusty  brown  glass  is  visible  in  the  thinnest 
parts  of  the  slide.  The  whole  slide  is  saturated  with  a  yellow-green, 
radially-fibrous  serpentinous  mineral  and  witl^  abundant  analcime,  in 


Tutakh  to  AhUat  87 

2  feet  in  diameter;  it  is  occasionally  amygdaloidal. 
The  lava  is  overlaid  by  white  calcareous  sands,  not 
improbably  lake-deposits  of  recent  age,  reaching  a 
thickness  of  about  40  feet  and  forming  the  summit 
of  the  high-plain,  which  is  covered  with  well-rounded 
pebbles. 

Near  Baindir  the  sands  become  light  brown  and 
compacted  into  a  sandstone,  but  for  nearly  an  hour 
afterwards  we  traversed  imdulating  limestone  downs. 
At  times  we  rose  up  to  the  surface  of  the  dissecSted 
plain,  but  usually  our  course  lay  along  winding, 
dry,  grassy  valleys,  with  no  exposures.  Whenever 
we  had  an  unrestridted  view  the  conical  dome  of 
Kuseh  I>agh  rose  high  above  the  rugged  Kilich  Gedik 
on  the  north;  but  on  the  west  tne  Ak  Dagh  was 
hidden  behind  an  extensive  basalt  plateau,  sloping 
gently  to  the  N.E.  and  forming  the  steppe  of  Torlu. 
Ota  the  south,  the  volcanic  masses  of  Sipan,  Kartevin, 
Bilejan  and  Khamur  were  conspicuous  landmarks. 

Before  descending  to  the  wide  alluvial  plain  of 
the  Murad  we  passed  a  warm  mineral  spring,  welling 
up  to  a  height  of  6  inches  above  the  level  of  its 
basin,  which  was  about  8  feet  in  width.  The 
w^ter  was  tepid,  mawkish,  and  slightly  salt  to 
the  taste.  The  spring  is  sitjuated  on  a  level  terrace 
of  calcareous  travertme  (No.  59),  extending  south- 
wards for  about  300  yards.  The  rock  is  dull  greyish 
white  and  rather  compadl,  without  any  traces  of 
organisms. 

aggrq[ates  of  dystaUine  granules,  often  lining  oavitiei  and  (as  already 
mentioned)  partially /replacing  fel^Mur.  In  some  cases  a  cavi^  his 
been  lined  with  the  serpentinous  mineral,  and  filled  with  caldte.  The 
occurrence  of  anaktme  is  similar  to  that  observed  by  Thost  (Abkamdl. 
Smckntb.  Naiurf.  Ga.  Frankfurt  a/M,  XVIII,  ii.  257)  in  some  of  the 
olivine-basalts  of  the  Bast  Karabagh  near  Shusha,  and  to  that  in 
tfaMB  teschenites  from  the  Caucasus,  described  by  Tsdiermak  and  later 
bf  Rohrbach. 


88  Geology  of  Armenia 

A  short  ride  over  dark  brown,  volcanic  sands 
brought  us  to  Gargalik  (12^  miles  from  Tutakh)« 
Thence  we  passed  over  another  terrace  of  travertine 
of  cavernous  texture,  containing  frequent  fragments 
of  lava;  the  dip  was  5''  N.N.W.  These  deposits 
overlie  dark  volcanic  sands,  which  occasionally  came 
to  light  on  the  surface.  It  was  not  until  we  had 
passed  Reshi  that  we  adlually  left  this  terrace  to 
begin  our  descent  to  the  Murad  plain,  which  lay 
about  100  feet  beneath  us.  The  brown  sands  overlie 
a  conglomerate,  consisting  of  rounded  pebbles  of 
red  lava,  cemented  together  by  iron  oxide.  Still 
lower  down  the  conglomerate  becomes  coarser,  with 
rounded  blocks  up  to  a  foot  in  diameter.  The 
lower  half  of  the  cliff  consists  of  grey  clay,  causing 
frequent  landslips  of  the  overlying  conglomerates 
and  sands.  The  beds  are  pradlically  horizontal, 
without  any  trace  of  fossils.  However,  they  are 
probably  lake-deposits,  contemporary  with  the  marls 
and  clays  with  Dreyssensia  polymorpka  of  the  Khims 
plain  (p.  66).  On  looking  up  a  lateral  valley  near 
Kesan  I  could  see  a  lava-flow  from  Kartevin  near 
the  head  of  the  valley,  about  a  mile  distant. 

We  reached  the  marshy,  alluvial  plain  of  the  Murad 
near  Bani,  a  little  below  the  confluence  with  the 
Kersik  river,  which  we  had  last  seen  at  Al  Khe& 
Bani  lies  at  the  northern  foot  of  a  large,  round  hill, 
jutting  into  the  river-plain;  to  avoid  a  detour  we 
rode  right  over  the  uopes  of  grey  clay.  On  the 
opposite  bank  a  conspicuous  volcanic  cone  guards 
the  entry  to  the  Kersik  valley,  whilst  the  symmetricel 
dome  01  Kuseh  Dagh,  nearly  50  miles  distant,  still 
dominates  the  landscape  in  the  north. 

Here  we  left  all  stratified  deposits,  which,  however, 
appear  to  be  continued  on  the  west  bank  of  tixt 
Murad.    They   were   ccmcealed   by   a   lava-flow   ol 


Tutdkh  to  Ahhlat  89 

pyroxene-homblende-andesite  (No.  60)  *  from  the 
Kartevin  Dagh.  Its  lavas  sink  down  to  the  river, 
but  do  not  cross  it.  The  Kurdish  village  Piraraasean 
was  reached  after  fording  a  marshy,  tributary  stream. 
An  Armenian  bridge  of  two  arches  (now  in  ruins) 
bears  witness  to  the  relapse  of  this  country  into  a 
state  of  barbarism.  Although  this  bridge  probably 
dates  back  to  at  least  the  twelfth  century,  yet  the 
pink,  trachytic  andesite,  of  which  it  was  built, 
remains  as  fresh  in  appearance  as  if  quarried  yester- 
day. From  Koch's  narrative  it  is  evident  that  it  is  up 
this  lateral  valley  that  the  "ruins  of  a  castle  are 
situated,  whence  the  Katervin  [Kartevin]  Dagh  de- 
rives its  name,"t  Th^  low  spurs  on  the  south  side 
of  the  valley  are  composed  of  a  grey  lava — a  hpm- 
blende-pyroxene-andesite,  with  some  biotite  (No.  61)  J 

*  The  rock  is  pale  pink,  trachytic  in  texture,  with  rough,  irregular 
fracture.  It  contains  abundant  striated  felspars  (up  to  3  mm.),  and 
less  numerous  dark  green  pyroxenes  (up  to  2  mm.).  There  is  a  brisk 
effervescence  with  add.  Under  the  microscope,  the  plagioclase 
(anbesine  about  Ab5An3)  is  present  in  large  idiomorphs,  frequently 
honeycombed  with  a  network,  mainly  peripheral,  of  dusty  glass ;  often 
zoned,  with  a  narrow  margin  of  more  acid  felspar ;  mostly  quite  fresh, 
but  sometimes  with  dusty  calcite  and  white  inica  along  the  cracks ; 
endoses  augite  and  hypersthene ;  a  smaller  generation  is  also  present, 
with  prismatic  habit.  Hypersthene  in  pale  pinkish  brown  idiomorphs 
is  fairly  frequent,  with  marked  pleochroism  (a — pinkish  brown, 
d — yellow  brown,  ^— green,  all  in  pale  tints) ;  mostly  with  an  opaque, 
hematitic  border.  Augite  is  less  abundant,  but  fresher,  in  very  pale 
green,  almost  colourless  idiomorphs ;  occasionally  in  groups.  Horn- 
blende occurs  in  brown  idiomorphs,  deeply  resorbed,  down  to  rather 
small  dimensions;  those  which  still  show  a  core  of  unaltered  horn- 
blende show  strong  pleochroism  (straw  to  umber-brown) ;  the  larger 
crystals  contain  apatite.  '  Magnetite  is  accessory.  The  groundmass 
is  an  exceedingly  fine  microfelt  of  felspar  (oligoclase)  microlites,  augite 
granules,  and  hematitic  dust,  with  some  colourless,  interstitial  glass. 
Owing  to  weathering,  dusty  calcite  is  abundantly  present  in  patches. 

t  J^ase  mpantischen  Gtbirge^  ii.  434. 

X  The  rock  is,  on  the  whole,  rather  similar  to  No.  60,  but  less 


9P  Geology  of  Armenia 

—with  rudely  colimmar  stnidhire.  The  alluvial  plain 
through  which  the  Muxad  meanders  is  2-3  miles  in 
width  at  the  spot  where  we  turned  off  eastwards  to 
camp  at  Hassima. 

The  morning  of  July  3rd  was  spent  in  an  ascent 
of  one  of  the  highest  peaks  of  the  Kartevin  Dagh, 
at  the  foot  of  which  Hassuna  nestles.  The  lava  of 
the  lower  slopes  just  above  the  hamlet  is  a  blacky 
glassy   olivine-basalt    (No.    62),*    which    frequently 

altered.    It  is  pale  grey  and  poroas,  with  trachytic  tescture  and  xought 
irregular  fracture.    It  contains  abundant  striated  felspars  (up  to  5 
mm.),  dark  green  pyroxenes  (up  to  a  mm.),  and  an  occasional  plate 
of  biotite.    Under  the  microscope,  the  plagioclask  (amdxsine  about 
Ab5An3)  is  present  in  large,  tabular  idiomorphs,  sometimes  in  groups ; 
frequently  with  a  peripheral  zone  of  dust  or  of  a  network  of  dus^ 
brown  glass  with  transparent,  adcular  crystallites;  often  zoned;  a 
smaller  generation  is  also  present,  with  prismatic  habit.      Aucrrt 
is  fairly  frequent,    in   very   pale   green,  rather  small  idiomorphs. 
Bronzits  is  sparingly  present  in  smaller,  very  pale  green  prisms,  with 
the  characteristic  cross-jointing;    only  slightly  pleochroic  (to  pa]e 
pink).    A  few  crystals  of  biotitb  occur  in  the  slide;   they  are  the 
largest  phenooysts  ^nd  have  suffered  much  corrosion;  they  contain 
small  crystab  of  apatite.    HoRMBLSxa)X  is  also  accessory,  in  small 
idiomorphs^  rendered  quite  opaque  by  resoibtion,  and  then  consisting 
of  magnetite  with  a  nanow  border  of  hematite;  in  one  case  there 
is  a  core  of  quartz,  a  zone  of  brown,  unaltered  hornblende,  and  f  nally 
a  black,  granular  resorbtion*border.    Apatite  and  magnetits  axe 
accessoiy.    The  grouncmass  is  an   exceedingly  £ne  miaofelt  of 
felspar  (oligodase)  microlites,  numerous  tiny  grains  of  augite  (stained 
by  iron  oxide),  and  a  fair  amount  of  majgnetite  granules,  with  seme 
pale  brown  or  colourless,  interstitial  glass.    There  are  many  smaU, 
irregular  steam-cavities. 

*  The  rock  is.  compact,  with  rather  pitdiy  lustre;  it  contain^ 
numerous  felspars  (up  to  3  mm.),  and  smaller  olivines  (up  to  x  mm.). 
Under  the  microscope,  the  FLAOioczjiSS  (a  basic  labradorite  about 
Ab3An4)  is  present  in  large  idiomOiphs,  mostly  of  prismatic  habit ; 
occasionally  pericline  twini^ing  is  visiblei  in  addition  to  tixe  usual 
albite  and  Carlsbad  systems;  fitih  and  free  from  inclusions,  but 
sometimes  corroded  and  penetrated  by  the  groundmass ;  often  faintly 
zoned.  Ouvikx  is  very  abundant  in  rather  small,  colourless  idio* 
morphs,  sometimes  also  penetiitcsd  by  groundmass.    Augite  is  only 


I 


Tutakk  to  Akhlat  91 

presents  a  pasty  appearance.  The  inclination  of 
the  hillside  is  quite  a  low  angle»  for  we  were  able 
to  ride  up  a  S.W. — ^N.E.  ridge,  just  south  of  Hassuna, 
up  to  the  last  300  feet  of  the  steep  and  rocky  peak. 

The  summit  is  composed  of  a  glassy  pyroxene- 
biotite-andesite  (No.  63),*  evidently  the  surwice-rock 
of  one  of  the  numerous  dikes  which  seam  the  ravines 
with  a  general  orientation  of  S.S.E. — N.N.W.  The 
Kartevin  Dagh  is  evidently  an  old  volcano,  which  has 
been  deeply  dissedled  by  long-continued  denudation 
into    many   ridges   and   peaks,    separated   by  deep 

sparingly  present  in  very  pale  green,  almost  colourless  idiomorphs; 
occasionally  twinned,  sometimes  repeatedly;  encloses  little,  green, 
hexagonal  crystals  (probably  serpentinised  olivines).  The  grouKd- 
iCASS  is  a  microfelt  of  tiny  felspar  laths  (andesine  about  Ab5An3), 
numerous  minute  crystals  and  granules  of  augite,  and  abundant 
needles  and  crystals  of  magnetite,  with  much  brown,  interstitial  glass, 
traversed  by  adcular  crystallites,  which  sometimes  show  a  stellate 
arrangement.  \ 

*  The  rock  is  coarse-grained,  owing  to  the  large,  white  felspars 
(up  to  8  mm.),  which  form  about  three-quarters  of  the  rock.  Small 
pyroxenes  (up  to  2  mm.)  and  more  seldom  a  tablet  of  biotite  (less  than 
I  mm.)  can  be  seen  with  the  lens.  The  matrix  is  a  black  glass,  which 
here  and  there  has  altered  to  a  pinkish  brown.  Under  the  micro- 
scope, the  PLAGiocLASi  (andbsine  about  Ab4An3)  is  present  in  laigi, 
tabular  idiomorphs,  often  in  radiate  groups;  much  cracked;  often 
with  a  peripheral  zone  of  brownish  dust,  with  a  narrow  margin  of  clear 
felspar ;  encloses  grains  of  augite  and  magnetite,  and  small  inclusions 
of  brown  glass  (with  bubble  and  adcular  crystallite),  often  ranged 
along  definite  lines.  Auom  in  pale  bottle-green  idiomorphs; 
encloses  crystallites  and  magnetite  grains.  Hypersthene  idiomorphs 
are  smaller  and  less  abundant  than  the  augite,  with  which  they  are 
often  associated;  usually  quite  fresh,  but  sometimes  with  a  slight 
Tesorbtion-border  (orange-brown  by  reflected  light);  markedly  pleo- 
chroic:  a — ^pale  pinkish  brown,  ^ — greenish  yellow,  c — ^pale  bluish 
green ;  frequently  full  of  crystallites.  Biotitb  is  still  less  frequent, 
in  very  small  individuals,  sometimes  deeply  resorbed.  Magnetite 
it  fairly  abundant.  The  groundmass  is  a  water-dear  glass,  with 
dusky  streaks  of  densely  crowded  belonites,  microlites  of  felspar, 
and  granules  of  augite,  biotite  and  magnetite,  ranged  along  contorted 
jines  of  tow.    The  adcular  crystallites  often  form  stellate  groups. 


93  Geology  of  Armenia 

ravines.  Our  peak  (7580  feet)  was  still  about  xooo 
feet  below  the  adlual  summit,  which  lay  further  N.E. 
We  returned  to  Hassima  down  a  steep,  grassy  valley, 
in  which  few  outcrops  of  the  black  lava  were  visible. 

A  short  ride  over  the  alluvial  plain  of  the  Murad 
brought  us  to  a  ford  over  the  .  Patnotz  river,  just 
above  its  confluence  with  the  Murad.  The  river  is 
bounded  on  the  south  by  low,  white  cliffs  of  hori- 
zontal, lacustrine  marls  with  Dreyssensia  polymoYpha. 
These  beds  evidently  extend  further  west,  ror  preasely 
similar  white  cli£Es  occur  also  on  the  north  bank  of 
the  Murad,'  after  its  great  westerly  bend  below 
Melaskert ;  and  I  found  deposits  of  the  same  age 
lower  down  the  Murad  valley,  from  Gop  to  Char- 
bahur  (see  Chapters  XI  and  XII). 

Our  camp  was  pitched  among  the  rums  of  the 
old  citadel  of  Melaskert— a  dty  in  the  last  stages 
of  decay,  dating  back  to  the  Vannic  king  Menuas, 
t.  800  B.C.  The  only  visible  record  of  tlutt  distant 
pre- Armenian  period  was  a  large,  rectangular  block  of 
diorite,  with  a  long  cuneiform  inscription,  lyings 
outside  the  city-wall,  near  the  east  gate.  This  block 
was  probably  broup;ht  from  the  west  coast  of  Lake 
Van,  for  I  found  a  similar  diorite  forming  rugged  cliffs 
between  Akhlat  and  Tunus.  The  walls  of  the  citadel 
and  of  the  city  have  been  built  of  the  black  lava  of 
the  plain — a  scoriaceous  olivine-bronzite-basalt  (No. 

64).* 

*  The  rock  is  fuQ  of  round  steam-holes  of  ntried  stse  (up  to  16 
mm.);  it  is  strongly  magnetic  (xo*"),  and  oMitains  stiiatea  fel^iars 
(up  to  IS  vamX  and  small  black  partideSi  sometimes  with  a  bronze 
lustre  (bronzite).  Under  the  microscope,  the  plagioclass  (labsa* 
t>  >RiTE  about  AbxAnx)  forms  numerous,  large  idiomoiphs  of  prismatic 
habit,  frequently  honeycombed  with  a  network  of  black  g^biss ;  endoses 
(divines ;  thd  largest  crystals  have  frequently  been  intntdeid  by  the 
poundmass.  Ouvine  is  present  in  abundant  small  tdiomorphs,  often 
xn  groups,  and  indeed  very  grqarious.    Brohzxts  is  in  somewhat 


Ttadlik  td  Akhlat:  93 

our  visit,  a  violent  earthquake  in  May  1903 
is  reported  to  have  totally  destroyed  the  city,  together 
with  33  of  the  surrounding  villages. 

During  the  morning  of  July  5*  I  made  a  survey 
of  the  aty- walls,*  and  in  the  sdftemoon  we  rode  to 
the  ruined  bridge  oyer  the  Murad,  2^  miles  distant. 
About  il  mile  from  Melaskert  we  crossed  a  ravine 
in  the  bJack  lava  of  the  plain  by  the  Chatun-kopri 
(= women's  bridge),  an  Armenian  strudhire  of  a  single 
pointed  arch,  the  upper  half  of  pink  trachytic  lava, 
the  lower  half  of  black  basalt.  The  Murad  bridge, 
about  three-quarters  of  a  mile  further  on,  on  the  road 
to  Erzerum,  was  originally  composed  of  ;2  pointed 
arches,  but  only  the  fifth  and  sixth  remain  perfect. 
The  bridge  corresponds  both  in  style  and  building 
material  with  the  Chatun-k6pri,  and  is  doubtless  of 
the  same  age,  viz.^  not  later  than  the  twelfth  century. 

On  returning  to  Melaskert,  it  seemed  to  me  not 
improbable   that    the  Murad,    at    the    time  of  the 

larger  crystals,  and  is  frequently  associated  with  the  olivine ;  very  pale 
green,  almost  colourless;  only  slightly  pleochroic  (to  pale  pink); 
tfeavages  distinct;  sometimes  a  laiger  crystal  is  surrounded  by  a 
kind  of  wreath  of  smaller  broniites  and  olivines ;  the  largest  crystals 
are  associated  with  inegular  grains  of  magnbtitx.  Augite  does  not 
occur  in  phenocrysts.  The  groundmass  is  a  felt  of  abundant  laths 
of  plagiodase  (andesine  about  AbsAni)  with  fluidal  axrangemcnt, 
Dumerous  small  olivines,  a  few  granules  and  tiny  prisms  of  augite, 
and  abundant  granules  and  skeletal  rods  of  magnetite,  sometimes 
in  "gridiron''  anangement;  the  last,  together  with  belonites,  6«r, 
render  the  dark  brown,  interstitial  glass  more  opaque  in  places.  The 
rock  is  very  similar  to  the  basalt  (No.  145)  bordering  the  Jil  Gol, 
at  the  S.W.  base  of  Sipan,  but  has  a  more  porphyritic  character  (see 
Chapter  X).  It  is  also  allied  to  some  of  the  basalts  and  pyroxene- 
andeskes  of  the  Central  Karabagh  plateau,  near  Gerjiisy,  described 
by  Thost  (his  Nos.  61-64,  Ahhandl,  Smckb.  Niiturf,  Ges.  Frankfurt 
a/M.,  XVIII,  ii.  256),  in  which  rhombic  pyroxene  is  associated  with 
9exy  small  olivines, 

^  Lynch,  Armenia^  ii,,  plan  facing  p.  370. 


94  Geology  of  Armenia 

■ 

founding  of  the  ancient  city,  flowed  close  under  fts 
wallS)  for  I  could  see  a  distindl,  dry  channel  in  the 
alluvial  plain,  running  from  the  west  wall  to  the  S,W. 
past  a  mosque,  along  the  foot  of  the  Java-sheet; 
but  at  the  present  day  the  river  is  about  two  miles 
distant  from  Melaskert. 

Within  the  precindts.  of  the  citadel  I  found  a  few 
irre^lar  blocls  of  limestone  (No.  65),  crowded  with 
shells  of  Dreyssensia  polymorpka^  Pallas;  fihey  had 
probably  been  brought  from  no  great  distance. 

We  resumed  oux^joumey  by  proceeding  southwards 
over  the  vast  lava  plain.  Near  Melasl^rt  the  basalt 
is  overlaid  by  a  thin  bed  of  white  stutrl^  but  even 
this  slight  mdication  of  sedimentary  strata  soon 
disappeared,  and  the  billowy  .plain  consisted  entirely 
of  the  black,  scoriaceous  basalt,  with  occasioned 
wart-like  moiuds  of  lava-blocks,  which  mscy  perhaps 
indicate  the  original  points  of  emission.  Ine  soil  is 
brown,  sandy  and  very  fertile. 

Kara  Ali,  a  Circassian  colony^  lies  at  the  southern 
border  of  the  plain.  Here  we  commenced  to  ascend 
the  high  downs  extending  between  Sipan  and  Bilejaiu 
The  lower  part  of  our  gradu^  ascent  lay  over  rounded 
hillocks  of  calcareous  sandstone,  with  numerous  shells 
of  Dreyssensia  polytnorpha. 

We  camped  at  Demian  (6690  feet ;  |o  miles  from 
Melaskert,  and  300  feet  below  the  pass).  The  village 
lies  at  the  foot  of  a  high,  white  escarpment,  whidi 
becomes  abruptly  vertical  further  east.  The  dip  is 
30^  S.S.E.,  and  the  downward  succession  is  as 
follows  :— 

z.  Escaipment  of  white  UmeitOQ^  whh  Dr^fWuU  folymarpha 
and  Mdanapm^  ip.  indet,  m  ttt  tmer  bedi,  wUcli  alternate 
with  grey  cakueottt  inails  cQQtaialng  the  tame  fiooils.  The 
lower  part  is  oompaot  and  uafofdllfaoas     •••    €.  xoo  feet 

9.  Yellowish  calcareous  flufttoneih  rnudi  stained  witb  iron  oxides 


\ 


\j 


HA\ 


HJ(M 


SEA-L 


v^ 


U 


•  * 

i 


] 


KAMJCAN 

^  TAMKAN 


-/4 


^-i?i 


TUTAKM 


r 

NURAB 
HIVE  II 


KARreN^tN   3>A©H 


t  t 


r 

TIRAMASCAN 


5  ■  ^ti 


rcf 


s 


\      * 


< 


Tutakh  to  Akkht  95 

and   sometimes   presenting  tke  character  of  lithographic 

stone;  with  obscure  worm*tracks       c.  20  feet 

3«  Yellowish   calcareous  shales,  forming  nearly  level  ground  at 
the  foot  of  the  escarpment      c.  80  feet 

4,  Grey  day..,'       •••        •••        •••        ...        •••      c.  60  feet 

5.  Hard  band  of  white,  traTertinons  limestone 5  feet 

6»  Grey  clay  •••        ..*        •••        ...        •••        •••        .so  feet 

7.  White  limestone  similar  to  No.  5  5  feet 

&  Grey  day r.  50  feet 

9*  White  limestcMie  similar  to  No.  5.    Demian  is  situated  upon 

this  bed  •••        •••        •••'       •••        •••        •••      ^  30  feet 

10.  Calcareous  sandstones,  with  frequent  shells  of  JDreyssensia 
polytnorpha\  grass-grown,  no  sections  exposed;  base  con- 
cealed by  the  lava  of  the  Mekskert  plain. 

The  pass  (6870  feet)  lies  a  little  south  of  Demian, 
through  a  slight  depression  in  the  clifif.  The  white 
limestone  is  here  crowded  with  Dreyssensia  poly- 
marpka^  but,  as  we  descended  into  the  braad  valley 
of  Akhviran,  the  unfossiliferous  beds  of  the  foregoing 
table  re-appeared,  dipping  25®  N.N.W.  The  Demian 
ridge  thus  marks  a  S3mcline. 

The  Akhviran  rivulet  (flowing  eastwards)  has  cut 
through  a  thick,  grey  river-gravel,  full  of  boulders  and 
pebbles  of  limestone,  pointing  to  the  former  greater 
size  and  power  of  the  present  insignificant  stream. 
As  soon  as  we  commenced  to  ride  up  the  south  side 
of  the  valley,  I  foimd  the  ground  to  consist  of  a 
sandstone,  containing  pebbles  and  rounded  fragments 
of  grey  and  red  marble  veined  with  calcite,  and  I 
noticea  also  a  pebble  of  limestone  with  a  species  of 
NummuUtes  the  size  of  lentils.  On  stratigraphical 
grounds  it  is  evidently  the  same  basal  Miocene 
conglomerate,  which  I  found  well  exposed  in  the  low 
clifEs  of  Akhlat  (vide  infra).  Here  the  dip  was  not 
apparent  owing  to  the  absence  of  any  section ;  but  on 
our  return  journey  I  met  with  this  sandstone  near 
Shebu,  a  few  miles  further  east,  in  the  same  valley. 
There    it    was    greenish    and    fine-grained,    derived 


96  Geology  of  Armenia 

mainly  from  diabasic  and  serpentinous  rocks,  and 
dipping  25^  S.S.E.  (see  Chapter  X).  Now  the  lime- 
stone, which  crops  out  a  little  higher  up  the  south 
side  of  the  valley,  shows  a  dip  of  30^  S.S.E.  Hence 
it  can  hardly  be  doubted  that  this  sandstone  and 
conglomerate  conformably  .underlie  the  limestone 
which  forms  the  steep  ridge  and  adlual  watershed 
(7690  feet)  of  these  high  downs.  This  limestone  (No. 
65)  is  greyish  brown  and  travertinous  in  appearance, 
somewhat  resembling  an  agglomeration  of  Serpulae 
tubes,  but  I  could  find  no  fossils. 

On  proceeding  S.W.-wards  to  Khanik  (on  a  stream 
flowing  west  to  Lake  Nazik);  the  dip  of  the  limestone 
increased  to  70®  S.S.E.  It  had  become  white  and 
crystalline,  cropping  out  in  bold,  rugged  masses  on 
the  south  side  of  the  valley.  In  this  Circassian 
villager  small  mosque  was  oeing  built  of  a  black, 
cindery  lava,  similar  to  that  of  Khinis  (p.  65).  Soon 
after  leaving  the  settlement  we  passed  over  a  fairly 
level  stretch  of  this  lava ;  I  imagine  that  it  must  have 
proceeded  from  some  volcanic  peak  further  west,  in 
the  neighbourhood  of  Lake  Nazik,  but  our  path  lay 
halfway  down  the  side  of  a  deep  ravine,  which 
effectually  restridled  our  field  of  view, 

The  river  flows  to  S.S.W,,  through  a  white,  marly 
limestone,  dipping  36^  N.N.W.,  forming  grassy  slopes 
with  few  exposures.  Owing  to  the  dangerous  repute 
of  this  zone  of  country  our  progress  was  somewhat 
hurried,  although  our  escort  had  been  increased  to  23 
zaptiehsy  and  it  was  not  possible  to  dismount  in  order 
to  examine  the  strata.  A  white,  conglomeratic 
escarpment,  about  20  feet  high,  on  the  opposite  side 
of  the  ravine,  showed  that  tiie  dip  had  changed  to 
S.S.E.,  and  the  angle  continued  to  increase  in  steep- 
ness. 

We  soon  turned  up  a  tributary  valley  in  a  southerly 


Tutakh  to  Akhlat  97 

direcftion.  The  strata  here  have  undergone  a  good 
deal  of  folding,  being  pinched  up  into  a  sharp 
syncline  and  anticline,  and  again  into  an  equally 
sharp  syncline  (85^  S.SJE.— 70**  N.N.W.).  This  fold 
was  rendered  more  conspicuous  by  a  band  of  coarse 
conglomeratic  limestone,  interbedded  in  yellow,  ar- 
gillaceous sandstone.  Our  course  then  turned  west- 
wards, continually  ascending ;  the  dip  of  these  sandy 
beds  diminished  to  40®  N.N.W.,  and  we  soon  reached 
the  new  guardhouse  on  the  summit  (7560  feet).  The 
building-stone  was  the  same  black,  cindery  lava 
which  I  had  noticed  near  Khanik. 

Here  we  were  standing  on  the  southern  edge  of 
the  block  of  Miocene  limestone,  which  breaks  off  in 
high  cliffs  to  Lake  Van,  near  Adeljivas ;  it  formis 
a  conspicuous,  white  escarpment  facing  the  south.  In 
places  it  has  beep  altered  to  a  grey  marble  by  the 
mtrusion  of  pyroxene-porphyrites,  both  at  this  spot, 
and  at  numerous  places  alon^  the  coast  (see  Chapter 
X).  The  dip  on  the  summit,  just  south  of  the  guard- 
house, is  30**  N.  by  W. 

A  little  distance  down  the  hillside  the  limestone 
(No.  66)  is  praAically  unaltered ;  it  is  a  hard,  cream- 
coloured  rock,  full  of  white  patches  (about  6  mm.  in 
diameter)  of  Ltthothamnion.  Under  the  microscope 
I  was  able  to  recognize  the  following  organisms  of 
Lower  Miocene  (Tortonian)  age : — 

Amphistegina  cf.  aauerina^  d*Orb. 

Heterostegina. 

Miliolina. 

Textularia. 

Bryozoa  indet.  (Hefero^aridce),  as  at  Madrak. 

Ceriopora  anomala^  Abich. 

Ltthothamnion  ramossissimunij  Reuss,  (very 
abundant).  It  will  be  noticed  that  the  species  are 
identical  with  those  which  I  have  recorded  at  Madrak 

H 


98  Geology  of  Armenia 

(p.  52),  Kan  jean  (Pi  8i)f  and  Al  Khes  (p.  82),  but  there 
is  an  absence  of  Orbitoides  ( Leptdocyclina). 

At  the  foot  of  the  escarpment  of  this  Lithothamnion- 
limestonet  about  1000  feet  below  the  summit^  the 
scenery  became  tame  and  uninteresting,  consisting 
of  brown  sands  and  purple  sandstones,  dipping  2cr 
N.N.W.,  with  frequent  bands  of  conglomerate  con- 
taining abundant  rounded  pebbles  and  boulders  (up 
to  2  feet)  of  rocks  alien  to  the  neighbourhood — red 
and  grey  marble  veined  with  calcite,  serpentine, 
diabase,  and  red  porphyrites.  It  is  evidently  the 
basal  conglomerate,  \\hich  was  formed  when  the 
Miocene  sea  invaded  Armenia.  As  we  approached 
Erkizan-^rone  of  the  distrids  of  Akhlat — the  dip  of 
these  fragmental  deposits  changed  gradually  to  4^ 
S.S.E.,  and  they  form  a  low  cliff,  alK)ut  6  feet  hign, 
along  this  part  of  the  coast  of  Lake  Van.  Between 
Erkizan  and  the  mouth  of  the  Akhlat  river  two 
shallow  folds  are  visible  (dip  not  exceeding  20^  to 
N.N.W.  and  S.S.E.),  finally  sinking  inta  the  lake  with 
a  dip  of  5®  S.S.E.  It  is  worthy  of  note  that  these 
conglomerates  contain  no  fragments  of  mica-schist, 
wMch  enters  so  largely  into  the  structure  of  the 
Taurus,  only  15  miles  distant. 

In  the  ravine  of  Takht-i-Suliman  fOld  Akhlat) 
the  purple  conglomerates  dip  10**  N.N.W.,  and  form 
the  oed  of  the  river.  They  are  overlaid  by  tufe 
and  cindery  basalt  from  the  Nimrud  Dagh  (see  also 
Chapter  IX).  This  valley  was  in  existence  previously 
to  the  emission  of  the  lava,  which  flowed  down  it, 
filling  it  up.  But  the  river  re^asserted  itself,  and 
cut  down  its  bed  to  a  lower  level  than  before,  iso- 
lating a  portion  of  the  lava-flow  to  form  the  imposing 
rock  on  which  the  citadel  was  built 

The  conglomerates  are  only  exposed  for  a  short 
distance  inland  from  the  shore,   vaxying  from  200 


^ 


-  \ 


k.AKe   VAN 


"^fniVrt^tt/* 


^^^^^«*^^^,. . 


Tutakh  to  Akhlat:  99 

to  400  yards,  and  form  t}ie  margin  to  the  coastline 
from  the  mouth  of  the  Akhlat  river  as  far  as  Tonus. 
They  disappear  inland  beneath  the  dusty  pumice 
sanos  of  the  Nimrud  Daglu  As  might  be  anticipated 
from  the  nature  of  these  zed  sandstones^  I  was 
unable  to  find  any  trace  of  fossils. 


CHAPTER  VIII 

THE  TAURIC  HEIGHTS 

The  little  promontory  of  Tadvan  is  of  particular 
interest  with  regard  to  the  strudhire  of  the  Armeniaa 
plateau,  for  it  rises  like  an  island  out  of  the  en« 
veloping  lavas  of  Nimrud,  and  forms  an  unexpjedled 
and  isolated  outlier  of  the  schistose  Tauric  Heights, 
which  bound  Lake  Van  and  the  plain  of  Mush  on 
the  south. 

The  Tadvan  promontory  consists  of  two  parts :  the 
castle-hill  is  the  larger,  and  is  connedled  by  a  narrow, 
sandy  isthmus  with  a  much  lower  and  smaller  mass, 
jutting  out  into  deep  watet.  An  anticlinal  axis  runs 
obliquely  through  the  castle-hill,  and  a  synclinal  axis 
through  the  small  outer  peninsula  (to  E.S.E.),  both 
running  W.  5°  S.— E.  5^  N. 

First  of  all  I  examined  the  castle*hill.  The  down- 
ward succession  on  the  south  side,  near  the  rude  jetty, 
down  to  the  water's  edge,  is  as  follows : — 

I.  Duk  gr^  marUe,  coarsely  foliated,  dipping  do^  N.  5^  W.^ 
and  full  of  yeini  of  rhombohedxal  calcite.  It  forms  the 
summit  of  the  hill  and  the  apex  of  the  antidine  ..  c.  25  feet 

9.  Pale  quartz-mica-schist       c.  so  feet 

3.  Pale  green,  chloritic  mica-schist  in  thinner  laminae;   quarts 

in  relatively  large  folia      a  25  feet 

4.  Dark  ^ey,  foliated  marble  (Nos.  124,  125),  becoming  more 

massive  at  its  base ;  with  white  veins  of  calcite  ...  r. 30  feet 


« 


The  Tauric  Heights  '•••'.-.  loi 


The  low  isthmus  between  the  castle-hill  and*'tlie 
much  smaller  outer  peninsula  is  covered  by  a  rec&ilt 
sliore-deposit,  about  2  feet  thick,  composed  of  sm'all;;.^ 
well-rolled  pebbles  of  basaltic  lava^  nodules  of  pumid^-\ 
and  rounded  fragments  of  mica-schist  and  marble — **/ '- 
the  same  materials  which  compose  the  present  beach  * 
at  Tadvan — in  a  calcareous  cement,  forming  a  natural 
concrete.    Whilst  practically  horizontal  on  the  isth- 
mus, this  deposit,  along  the  shore,  slopes  towards 
the  lake  at  the  same  angle  as  the  modem  beach. 

The  synclinal  axis  only  just  cuts  the  northern  shore 
of  the  smaller  hill.  A  low,  rugged  cliff  forms  the 
extremity  of  this  mass ;  it  consists  of  a  silvery,  mica- 
ceous limestbne-schist  (No.  122),**  dipping  35**  N.  s^ 
W.,  much  traversed  by  veins  of  rhomboh^ral  calcite 
(dog's  tooth  spar).  It  is  overlaid  conformably  by 
massive,  dark  grey  marble.  Higher  beds  occur  on 
the  north  side  of  the  hill,  which  yields  altogether 
the  following  succession  down  to  the  water's  edge, 
each  bed  being  about  10  feet  in  thickness : — 

I.  Pale  mica-scfatst  with  irregular  layers  and  '  eyes '  of  quartz. 
9.  Fissile,   black  marble  (No.    123),   with   obscure  markings, 

*  The  rock  is  veiy  thinly  foliated;  the  white  mica  is  in  minute 
flakes,  mingled  with  occasional  larger  scales  of  dark  green  chlorite 
(up  to  I  mm.) ;  the  rode  is  stained  in  places  with  iron  oxide  Under 
the  microscope,  CALcrrx  in  irregular  elongated,  crystalline  grains  forms 
the  bulk  of  die  rock ;  part  of  a  vein  of  large  calcite  crystals  just  confies 
into  the  slide.  A  I-ilspar  is  present  in  numerous  irregular  grains, 
often  rounded,  usually  lenticular  along  schistose  planes  forming  *  eyes'; 
sometimes  simply  twinned;  encloses  calcite»  quartCi  and  sparing, 
adcular  crystallites  similar  to  those  in  the  felspars  of  the  mica-schut 
from  Gozel-dereh  (No.  xa8,  v,  infra).  Quartz  is  quite  subordinate, 
ii  small  grains,  zatber  inegularly  distributed,  but  sometimes  in  a 
drawn-out  mosaic  Biotite  is  sparingly  present  unaltered,  but  it 
has  apparently  been  replaced  by  chi^orite,  m  scattered  flakes.  Sma  1 
fl..ke8  of  wHrrs  mica  sweep  round  the  felspars  and  give  the  rock  its 
schistosity,  but  are  rather  sparingly  developed.  LiMOKrrE  occurs  in 
scattered  granules  and  specks. 


••' 


t 


•  • 


I02   fvV/  Geology  of  Armenia 

'  *  -  perhaps  fuooidal. 
''..3.  Folialedt  grey  marble,  with  scattered  mica-flakes  along  the 

foliation-planes;  passing  into 
>*'•'  4.  Massive,  dark  grey  marble. 

'•.-'    5.  Silvery,  micaceous  limestone-schist  (Na  laa),  with  veins  of 
rhombohedral  caldte. 

The  rocks  present  a  slightly  different  charadler  from 
those  of  the  castle-hill,  and  probably  represent  a 
higher  horizon,  if — as  it  seems  to  me —  these  outliers 
at  Tadvan  of  the  Tauric  Heights  are  merely  the 
summits  of  blocks,  which  have  simk  down  by  step- 
faults  to  a  lower  level,  and  have  then  been  surrounded 
by  the  lavas  of  Nimrud  and  the  waters  of  Lake  Van. 

By  the  kind  invitation  of  Colonel  fthen  Captain) 
G.  S.  Elliot,  then  Consul  at  Van,  and  ot  Mr.  Monahan, 
then  Consul  at  Bitlis,  a  short  excursion  by  boat,  in 
their  company,  from  Akhlat  to  Gdzel-dereh,  in  the 
S.W.  comer  of  the  lake,  enabled  me  to  examine  a 
marginal  portion  of  the  Tauric  Range.  Owin^  to 
the  wind  falling,  it  took  us  nearly  8  hours  to  sail  only 
20  miles  to  the  little  crater  of  Sheikh  Ora,  on  the 
south  coast.  A  slight  eastward  deviation  was  ren- 
dered necessary  by  the  long,  shoaling  promontory 
of  Zighag,  which  appeared  to  be  entirely  volcanic, 
with  low,  grey  cliflb  and  yellow  sands.    We  passed 

Quantities  of  pumice-nodules,  the  size  of  walnuts, 
oating  on  the  water;  they  have  a  brownish  in- 
crustation, and  were  doubtless  washed  out  of  beds 
of  similar  nodules,  which  I  observed  near  Kizvag,  and 
elsewhere  on  the  west  coast  of  Lake  Van,  in  the 
neighbourhood  of  the  Nfmrud  Dagh  (see  Chapter  IX). 
As  we  drew  nearer  and  nearer  to  the  southern  shore, 
I  could  distinguish  three  old  shore  lines,  at  about  15, 
Ko  and  100  feet  respedtively  above  the  present  level; 
the  highest  beach-line  is  the  most  conspicuous. 


«.  ♦•^    •. 


At 


-S-OivM-a 


'.^."•■»"»:^: 


t 


The  Tauric  Heights  103 

The  little  crater  of  Sheikh  Ora  lies  open  to  the  lake 
by  a  breach  in  its  wall  on  the  N.W.,  and  we  sailed 
slowly  through  the  strait  into  the  inner  lake,  which 
is  about  I  mile  in  diameter  at  water-level. 

The  highest  edge  of  the  crater,  viz.^  on  the  east 
and  south,  rises  to  about  500  feet  above  the  lake. 
The  north  and  north-west  part  of  the  wall  clearly 
owes  its  lower  height  to  the  time  when  the  lake  stood 
at  the  100  foot  b^ch-level.  It  may,  I  think,  be  also 
deduced  that  the  lake  did  not  remain  very  long  at 
this  height,  or  else  this  crater  of  volcanic  ash  would 
have  been  entirely  reduced  and  washed  away  by  the 
waves. 

It  is  probable  that  the  final  explosion  destroyed 
more  of  the  southern  wall  than  the  northern  and 
eastern,  for  in  these  the  anticlinal  structure  is  clearly 
visible,  whilst  in  the  steeper  southern  and  western 
sides  the  dip  is  exclusively  outwards,  away  from  the 
centre.  The  adlion  of  the  waves  has  doubtless  had 
some  share  in  widening  the  basin,  for  it  is  not  stridlly 
circular,  but  slightly  elliptical — the  long  axis  passing 
from  S.E.  to  N.W.,  through  the  breach  in  the  wall. 

The  crater  consists  entirely  of  a  very  basic  tuff  (No. 
120),*  in  well-marked  layers,  varying  from  about  4 

*  The  rock  is  dull  brown,  friable  and  dndery,  full  of  black,  porous 
lapilli,  which  weather  a  rusty  brown.  One  of  Uie  specimens  is  slightly 
magnetic  (4'').  A  small,  rounded  fragment  of  grey  limestone  is 
present.  Under  the  microscope,  the  angular  lapilli  consist  of  a 
black  glass ;  they  are  mostly  elongated,  and  very  vesicular,  the  vesicles 
being  often  filled  with  calcite ;  the  lapilli  contain  small,  fresh  idio- 
morphs  of  olivine,  very  free  from  inclusions  except  for  occasional 
intrusions  by  the  glassy  groundmass.  Angular  fragments  of  a  yellow- 
brown  GLASS  (sometimes  with  a  greenish  tinge)  are  frequent,  often 
pumiceous,  with  a  narrow,  clearer,  feebly  anisotropic  selvage ;  drawn- 
out  bubbles,  streaks  of  black  glass,  fresh  olivine  crystals,  and  (in  one 
or  two  cases)  a  few  narrow  laths  of  plagioclase  (ijibradorite  about 
AbiAni)  occur  in  sdme  of  the  larger  glass-fragments.  One  fragment 
is  more  crystalline:,  consisting  of  numerous  tiny  felspar-laths  (labra- 


I04  Geology  of  Armenia 

feet  to  an  inch  in  thickness.  Some  blocks  of  grey  and 
white,  banded  marble  occur  in  this  tuff. 

This  little  volcano  is  situated  on  the  great  line  of 
faulti  whichi  accordinp^  to  my  observations,  runs  for 
over  loo  miles,  definmg  the  southern  boundary  of 
Lake  Van  and  of  the  Mush  plain. 

The  little  village  of  Sheikh  Ora  lies  on  the  eastern 
shore  of  the  basin,  where  the  water  is  fringed  by  a 
group  of  willows,  about  20  years  old,  some  of  which 
are  now  immersed  in  the  bitter  water,  and  have  died 
in  consequence.  The  rise  in  level  cannot  have  oc- 
cnirred  long  ago,  for  the  dead  wiUows  hardly  differ 
in  size  from  their  still  living  companions.  A  gigantic 
mulberry  tree  of  great  antiquity  is  also  (1898)  about 
to  succumb  to  the  recent  encroachment  01  the  waters, 
for  one  half  of  its  lobster-red  toots  is  now  sub- 
merged. This  circumstance  clearly  shows  that  the 
present  rise  of  the  lake  is  the  highest  that  has  taken 

{>lace  during  the  life  of  this  tree,  f.e.,  perhaps  for  as 
ong  as  800  years. 

On  proceeding  along  the  coast  to  Gozel-dereh, 
where  the  camp  had  been  pitched,  we  kept  close  in 
shore,  for  the  water  was  deep,  but  it  soon  became 
too  dark  to  see  the  rocky  coast  distindlly.  But  at 
first  I  was  able  to  make  out  that  the  comparatively 
low  cliffs  were  composed  of  banded,  grey  and  white 
marbles  overlying  mica-schists.  Several  faults  were 
apparent,  and  the  strata  dipped  about  60^,  first  to 
S.S.E.,  and  then  N.N.W.;  an  anticlinal  axis  apparent- 
ly nms  along  this  part  of  the  shore.  The  boiat  was 
beached  in  darkness  on  the  narrow  spit  of  sand  at 

dorite),  olivines,  and  magnetite  granules  in  a  dusty  brown  glass.  The 
GROUKDMASS  consists  of  txDj  glflis-splinters  and  dust,  with  a  good  deal 
of  disseminated  caldte.  Sq^ate  crystals  of  olivme,  rarely  serpen* 
tinized,  wholly  or  in  part,  also  lie  in  the  groundmass.  A  small 
aggtegate  of  quartz  grains  exoq^ionally  occurs,  filling  up  a  cavity. 


The  Tauric  Heights  105 


the  mouth  of  the  Gozel-derdi  river,  and  after  crossing 
the  shallow  lagoon  we  climbed  up  a  rocky  cliff  to 
the  tents. 

When  daylight  came  I  found  that  the  camp  was 
situated  on  a  terrace  of  travertine,  which  was  part 
of  the  old  beach  at  the  xoo  foot  level.  The  terrace 
was  strewn  with  rolled  pebbles  of  red  and  grey 
marble. 

Here  again  the  recent  encroachment  of  the  lake 
has  made  itself  manifest  by  the  formation  of  lagoons 
behind  a  narrow  spit  of  beach  across  the  mouth  of  the 
valley.  This  state  of  things  is  particularly  noticeable 
wherever  a  river  enters  the  lake,  e.^.,  at  (xozel-dereh, 
Akhlat,  Sor,  and  Adeljivas,  but  it  ako  occurs  when  an 
indentation  is  flooded  by  the  rising  waters  of  the  lake, 
so  that  the  continuity  of  the  coastline  is  maintained 
from  headland  to  headland,  e.^.,  at  Tadvan,  Kizvag, 
and  Sor.  The  narrow  spit  is  clearly  formed  by  a 
coastal  current,  sweeping  round  from  east  to  west 
alon^  the  south  coast  from  Cape  Devehboyim,  and 
bending  round  northward  alon§[  the  western  coast 
of  Lake  Van ;  that  this  direction  is  constant  is  proved 
by  the  position  of  the  mouths  of  the  various  rivers : — 
vix.^  on  the  west  side  of  the  ^spit  at  Gdzel*dereh  on 
the  south  coast,  but  on  the  north  side  of  the  spit 
at  Akhlat,  Sor,  and  Adeljivas  on  the  west  coast. 

Further  observations  on  this  presumed  current  are 
greatly  to  be  desired ;  it  cannot  be  due  to  wind,  for 
it  doubles  on  itself  as  it  rounds  the  S.W.  comer  of 
the  lake.  It  must  therefore  be  mainly,  if  not  al- 
together, due  to  the  direction  which  the  waters  of 
the  Bendimahi  river  take  on  entering  the  lake,  im- 
pinging on  the  low  promontories  at  the  base  of  Sipan, 
and  thereby  crossing  over  to  strike  upon  the  south 
side  of  the  lake,  and  then  eddying  round  the  west 
coast.    Furthermore,  all  the  considerable  rivers  enter 


io6  Geology  of  Armenia 

the  lake  on  the  east  coast,  and  both  the  Mermid-chai 
and  the  Khosh^ab  (north  and  south  of  Van  respedive* 
ly)  bring  down  a  larger  body  of  water  than  the  Bendi* 
mahi  Su»  and  would  assist  the  aiftion  of  the  current 

The  following  day  (July  9^1^)  was  spent  in  making  a 
short  excursion  to  a  peak  on  the  left  bank  of  the 
Gdzd^derch,  in  the  company  of  Colonel  Elliot,  whose 
intimate  knowledge  of  the  topography  was  invaluable. 
At  first  we  walked  up  the  ti^td  side  of  the  valley, 
descending  to  the  alluvial  flats,  where  a  right  tn- 
butary  is  spanned  by  an  old  ibmenian  bridge,  built  of 
the  lava  which  has  partly  filled  up  the  Gdzel-dereh. 
This  basalt  is  grey  and  cindery,  weathering  to  a  pale 
brown,  and  has  formed  coHmms,  4-5  feet  in  diameter. 
It  flowed  down  the  Gocelndereh,  from  the  S.W.,  at 
a  comparatively  recent  period^  probably  from  the 
Gholtig  Meidan,  just  above  the  head  of  this  valley,  for 
Earl  Percy  *  describes  this  mountain  as  *'  exadtly  like 
the  crater  of  an  extin(%  volcano,  a  basin  rimmed 
on  every  side  by  low,  pointed  bluffs/'  Subsequently 
the  river  has  bieen  able  to  cut  down  about  w  feet 
through  the  lava,  forming  pidhuresque  rapids  and 
watexfalls  in  its  attempt  to  return  to  its  former  bed. 
An  isolated  block  of  the  columnar  lava  still  occurs 
on  the  1^  bank,  at  the  mouth  of  the  river.  It  is  veiy 
similar  to  the  basalt  which  flowed  down  the  Akhlat 
river  feom  the  Nimrud  Dagh  (p.  98)  and  was  x>robably 
contemporaneous. 

Soon  we  climbed  up  to  another  flat-topped  terrace 
of  travertine  TNo.  131),  full  of  impressions  of  walnut 
leaves.  Another  part  of  the  travertine  (No.  r33) 
contained  numerous  fragments  of  mica-sdiist,  up  to 
6  inches  in  length,  and  further  on  angular  blocks 
of  quartz  and  grey  marble,  derived  from  the  surround- 
ing heights,  were  embedded* 

*  migkbmdi  p/AjMc  J^a^key.    LMdcn,  1901. 


•i-    -s*^^    -i- 


ip^^^iraituv'-lau*!,  / 


•v  . 


J 


Tie  TanrU  Hmghts  107 

We  again  crofiBed  a  small  right  tributary,  cutting 
through  mka^schist  Black»  dndery  lava— the  over- 
flow nom  the  main  valley*-still  fills  the  bed  of  the 
stream.  The  mica-schist  weathers  easily  into  a  dark 
brown  sand  Here  we  descended  to  the  Gozel-dereh 
river,  past  a  cave  hollowed  out  of  the  soft,  cindery 
lava,  in  which  the  columns  have  undergone  an  ir- 
regular modification,  radiating  from  different  centres. 
The  lava  is  dark  grey  when  fresh,  with  finely  scoria* 
ceous  streaks,  and  easily  worked — in  fadl  presenting 
similar  features  to  the  later  basalts  of  Nimrud  Dagh, 
especially  to  the  great  lava-streams  which  flooded 
the  plain  of  Mush  (see  Chwter  IX).  Included  frag- 
ments of  quartz  were  not  in&equent,  evidently  caught 
up  by  the  lava  in  flowing  doym  the  narrow  valley. 

We  crossed  the  mountain-torrent,  and  began  to 
ascend  a  conspicuous  peak,  abdut  20cx>  feet  above  the 
lake.  The  lower  slopes  are  composed  of  grey  marble 
(mudi  veined  with  calcite),  dipping  80^  S.S.E.  The 
mica-schist,  which  we  had  traversed  on  the  opposite 
bank,  underlies  this  grey  marble.  The  ridge,  up 
which  we  were  climbing,  happened  to  nearly  coincide 
with  the  apex  of  an  anticline,  for  on  the  north  side 
the  dip  had  changed  to  50^  N.N.W.,  but  exposures 
were  few.  About  half  way  up,  the  ^rey  marble  is 
overlaid  by  mica-schist  (No.  128),*  dippmg  5o®N.N.W. 

^  The  rock  ii  ohreqr  brown,  rather  finely  foliated,  with  frequent 
patns  of  irisite  felspar  (up  to  2  mm.),  giving  a  spotted  appearance 
to  the  cross-section  I  small  chlorite  scales  are  associated  with  the 
white  mica.  Under  the  microscope,  the  fblspar  occurs  mostly  in 
rounded  grains,  yet  sometimes  angular  j  occasionally  in  groups  of  a  or 
j|,  forming  '  eyes,*  round  which  the  mica  flakes  sweep ;  twinnii^  only 
umpl»  when  present;  cloudy,  bluish  grey,  sinuous  streaks  in  the 
fdqwzs  contmue  lines  through  the  enveloping  mica,  and  consi^  of 
numbers  of  tiny  adcuhr  crystaDites,  (mostly  hairlike  even  with  a 
1/4  inch),  crowded  and  felted  t<^gethcr,  wiUi  rather  hi^  refractive 
index  and  strong  polarisation ;  along  these  sinuous  lines  some  dear 
secondary  felspar  has  developed.     Qt^ARxs  occurs  in  drawn-out 


io8  Geology  of  Armenia 

Continuing  our  ascent,  I  found  the  mica-schist 
becoming  more  quartzose,  and  passing  upwards  into  a 
brownish,  micaceous  (juartzite  (No.  129),  with  only 
occasional  streaks  of  mica-schist.    This  rock  is  in  turn  ^f 

overlaid  by  a  pure  white  marble,  forming  the  a(5lual  t^ 

summit  of  the  peak.  ^ 

On  looking  southwards,  I  could  see  that  the  rugged 
heights  of  marble  and  schists  extend  for  fully  10  miles 
to  the  horizon ;  another  anticlinal  ridge  occurs  a  little 
to  the  south  of  our  peak,  with  a  dip  of  about  50°  to 
N.N.W^  and  S.S.E.    The  view  from  this  spot  shows  f^ 

the  strong  contrast  existing  between  the  tumultuous,  * 

serrated  border-ranges  of  metamorphic  rocks  on  the  ! 

south  and  the  smooth,  gentle  contours  of  the  Ar-  1 

menian  plateau  on  the  north — ^an  alternation  of 
swelling  limestone- downs  with  lava-flows  from  ex-  |^ 

tinc%  volcanoes. 

We  returned  to  camp  in  as  diredl  a  line  as  possible, 
keeping  some  distance  north  of  the  line  of  our  ascent. 
Our  course  lay  entirely  over  mica-schist,  dipping  50® 
N.N.W.,  and  weathering  to  a  thick,  brown  sand, 
which  is  much  more  fertile  and  overgrown  than  the  ; 

marble.    On  nearing  the  Gozel-dereh  river  I  noticed  ^ 

that  the  mica-schist  is  unconformably  overlaid  by 
a  nearly  horizontal  bed  over  10  feet  in  thickness, 
of  pumice  nodules;  the  base  of  the  bed  was  not 
risible.  Near  Tukh  these  beds  are  overlaid  by  finely 
stratified,  ^ey  volcanic  ash — an  occurrence  similar 
to  that  which  I  subsequently  observed  in  the  Akhlat 

mosaics  of  small  angular  granules.  MAGNErm  is  present  in  a  few 
scattered  gxains.  A  small  amount  of  moTrrs  remains  unaltered  when 
entangled  in  the  quartz^mosaic ;  elsewhere  it  has  given  rise  to 
CHLORITE,  which  occurs  in  bundles  of  short  flakes,  with  marked 
pleochroism  {O — ^straw,  JE-— green),  associated  with  the  white  mica  ; 
tills  is  abunaantly  present  in  small  flakes,  enveloping  the  quarts- 
mosaics  and  the  felspars.  It  is,  in  fact,  a  rock  wbxoi  might  have 
come  from  the  Alps.  ^  , 


4 


I  T^OswA^L^     del  £  imp^ 


t: 


v> 


V  ^  ^-'  .. 


♦    '^Jvn.H  Si«"'K«  « 


'  ^ 


,  -< 


The  Tauric  Heights  109 

gorge  (see  Chzpttt  IX). 

Ine  zone  of  marble  and  schists  extends  eastwards 
to  Ardost  and  the  Agherov  Dagh,^  and  also^  as  I 
found,  for  a  considerable  distance  westwards.  The 
southern  part  of  the  watershed  between  Lake  Van  and 
the  Bitlis-chai  is  formed^ust  west  of  Tadvan — ^by 
a  spur  of  mica-schist  (dipping  40^  N.N.W.),  extending 
from  the  base  of  a  high  peak  ot  marble  above  Urtab. 

My  next  accjuaintance  with   these  rocks  was  at 

Norshen,  13  miles  westwards,  on  the  south  side  of 

the  plain  of  Mush.    The  village  is  situated  on  terraces 

of  basalt,  flanking  the  mountain-side,  oompMed  of 

mica-schist,  dippin^^  about  50^  N.N.W.,  with  seams  of 

quartz.    Here  again  it  yields  a  fertile,  sandy  soil, 

much   overgrown   with   oak-scrub.     Further   south, 

the  higher,  barren  mountains  are  clearly  composed 

of  grey  and  white  marble.    Indeed . Ainsworth  f  states 

that  the  Zirzira  Dagh  (south  of  Norshen)  is  '*  a  giant 

limestone  precipice  on  primary  schist/'  and  that  **  Just 

west  of  Khaskoi  a  cape  of  limestone  juts  out  mto 

the  plain''  of  Mush.     Mica-schists  have  also  been 

recorded  south  and  west  of  Mush  by  Koch,$  and 

still  further  west  by  Kotschy§  at  Noreg  and  Angag 

(S.W.  of  Boghlan),  so  that  these  schists  extend  as 

tar  as  the  41st  meridian,  and  not  improbably  to  Palu, 

and  Lake  Goljik,  where  they  have  been  recorded  by 

several  travellers  (see  Chapter  XV). 

Although  I  could  £a)d  no  trace  of  any  fossils  in 
the  marble  I  am  inclined  to  consider  the  series  to 

*  See  Chapter  XV  for  a  detaBed  description  of  this  pi^rt  of  the 
Titinis. 

t  JVavds  and  JtaearAa  in  Asia  liinor^  Mesopotamia^  CkaUea  and 
Armsnidt  iL  357. 

X  Jtdu  im/oniisckin  GMrgs^  ii.  396,  399,  407. 

{  jRiiu  von  Ih^esnnt,  etc.,  Petemiaim't  IfittL  vi.  69.  Gotfaa,  i860. 


I  JO  Geology  of  Armenia 

be  pre- Devonian  in  age ;  for  further  west,  at  Belen  in 
the  Antitaurus,  between  Albistan  and  Kaisariyeh, 
Tchihatcheff  **  found  Upper  and  Middle  Devonian 
beds  overlying  "  clay-slates,  blue  limestone,  and  mica- 
schist/*  The  question  of  the  precise  age  of  the  Tauric 
schists  must,  however,  remain  in  abeyance  until  the 
geology  of  the  region  to  the  south  can  be  definitely 
ascertained.  However,  Loftust  found  Eocene  strata 
overlying  the  ancient  marble  of  the  Chelifteh  Dagh, 
between  BitKs  and  Sert.  Qn  the  north  side  of  the 
Mush  depression  the  oldest  rocks  are  Lower  Miocene 
(Tortonian),  which  I  found  in  the  Murad  gorge  (see 
Chapter  XII). 

This  block  of  ancient  metamorphic  rocks,  lying  to 
the  south  of  Lake  Van  and  the  plain  of  Mush,  has  had 
a  N.E.-S.W.  strike  originally  impressed  upon  it  when 
the  strata  were  still  sufficiently  plastic.  This  folding 
probably  took  place  at  the  close  of  the  Carboniferous 
period,  for  this  diredlion  was  (according  to  Abich  J) 
the  original  strike  of  the  Devonian  and  Carboniferous 
rocks  in  the  Ararat  and  mid-Araxes  region.  This  is 
also  the  strike  of  the  similar  Devonian  and  Carboni- 
ferous of  the  ^titaurus.  Since  the  Palaeozoic  age 
the  Tauric  block  pr  *  horst '  has,  it  seems  to  me,  re- 
sisted all  later  foldiqg,  and  when  the  Caucasus  sub- 
sequently received  its  .final,  post-Miocene  folding  by 
a  tremendous  pressure  from  the  N.E.,  the  result — so 
far  as  the  Armenian  area  was  concerned — was  to 
produce  a  series  of  N.W.-S.E.  fractures  (see  p.  lo),  and 
in  particular  the  great  fault,  with  this  diredion, 
which  bounds  the  plain  of  Mush  and  Lake  Van  on  the 
south,  cutting  obliquely  across  the  old  strike  of  the 

*  Asie  Mimkn^  GMogU^  L  679^99. 

t  Quart.  JcmrfL  Giot.  Sac.  xL  325,  and  see  Chapter  XXI. 

X  Vergfachende gnl.  GrundMugi^  etc,  ai,  8i. 


The  Tauric  Heights  i|i 

Tauric  *  horst'    Hence  the  S,W.-N.E,  ridges  appear  in 
^helon  along  the  line  of  this  fault. 

The  descriptions  of  a  few  specimens  from  the  south 
coast  of  Lake  Van.  which  Colonel  Elliot  coUedled  and 
kindly  placed  at  my  disposal,  may  conveniently  be 
appended  to  this  chapter,  viz.: — 

Nos.  124,  125.  Foliated,  dark  grey  and  white 
marble :  base  of  castle-hill,  Tadvan  (see  p.  100). 

No.  126.  Grey,  compadt  marble;  hill  above 
Urtab ;  similar  to  the  marble  of  Tadvan  and  Gozel- 
dereh. 

No,  127.  Foliated,  grQr  and  white  marble, 
slightly  micaceous ;  hill  above  Urtab. 

No.  133.  Limestone-sdiist  ;^  south  coast,  Lake 
Van.  This  specimen  probably  comes  from  the  heights 
to  the  east  of  the  plam  of  Garchigan,  for  Hommaire 
de  Hell  t  describes  them  as  consisting  of  *^  blue  lime- 
stones .  .  accompani:ed  by  greenish  micaceous  schists, 
which  often  pass  insensibly  into  the  limestone." 

No.  I33A«  Saccharoidal  marble ;  Ahkiul  Dagh 
near  Akhtamar. 

No.  133B.  Grey,  silky  mica-phyllite,  weather- 
ing brown ;  road  between  Alaman  and  Tukh. 

No.  134.    Serpentine,'];  south  coast,  Lake  Van. 

*  The  rock  consists  mainly  of  caldte,  with  flakes  of  white  mica 
and  more  sparing  chlorite  along  the  folia.  Under  the  microscope,  the 
abundant  calcxtb  is  present  in  irr^[iilar,  crystalline  grains.  The 
CHLORITE  flakes  are  larger  but  less  numerous  than  the  white  mica. 
Quartz  is  present  in  scattered  grains,  sometimes  cracked,  and  veined 
with  caldte.  A  slight  amount  of  icagnetite  is  also  present  along 
the  foliation-planes. 

t  Vtyagi  en  T^urqme  ei  en  Perse^  ir.  95. 

%  The  rock  is  very  dark  green  and  uniform  in  texture;  with  a 
smooth,  vamish-like  surface.  Under  the  microscopei  bastite  is 
present  in  fairly  numerous,  very  pale  green  grains  in  a  typical  network 
of  pale  green  sbrpsntins,  dearly  after  olivine;  a  good^xieal  of 
MAGNsnTB  has  separated  out  in  strings  along  the  meshes*    The  mag- 


112  Geology  of  Armenia 

No.  135.  Fine-grained  diabasic  rock,  with 
pyrites ;  cliff  at  Surp. 

No.  136.  Carbonaceous  shale;  no  label,  but 
probably  from  Seivan  (east  of  Van),  where  coal  is 
known  to  exist  (see  Chapter  XXI). 

No.  137.  Iron  pyrites;  Shattakh,  south  of 
Lake  Van. 

No.  138.  Rock-crystal — the  crystals  up  to  i 
inch  in  diameter;  no  label,  but  probably  from 
Shattakh. 

No.  1^9.  Specular  iron  ore;  Talori  mines, 
Sassim  district. 

netite  is  somekimet  bordered  by  limonite  dae  to  wtaihering.  CHRomn 
is  only  sparin^y  present,  in  very  smaO  grains,  bordered  by  magnetite. 
This  serpentine  must  have  originally  been  an  enstatite-oUvine  lod^ 
bat  neither  of  these  minerab  has  escaped  alteration.  It  is  very  similar 
to  the  serpentine  from  Ebnali-dereh  (p.  78). 


CHAPTER  IX 


7«E  NIMRUD  VOLCANO 


Among  the  numerous  extincft  volcanoes  of  the  Ar- 
menian plateau  Ararat  holds  the  first  place  in  majesty 
and  altitude.  To  the  geologist,  however,  the  huge 
crater  of  the  Nimrud  Dagh  offers  equal,  or  even 
greater  interest,  yet  this  mountain  is  barely  known 
to  the  world  even  by  name. 

In  1852  Loftus,*  in  passing  along  the  western  shore 
of  Lake  Van  at  the  base  of  Nimrud,  susp^ed  the 
presence  of  a  crater  on  the  summit,  and  indeed  was 
informed  that  ^*on  clambering  to  the  top  a  person 
looks  down  on  the  opposite  side  into  a  circular  de- 
pression with  a  lake  at  the  bottom."  In  recent  years 
Major  Clayton!  has  given  a  brief  description  of  a 
hasty  visit  to  the  crater,  yet  he  makes  no  mention 
of  the  large  crater-lake,  although  he  describes  the 
much  smaller  '^hot  lake.'*  But  no  attempt  at  any 
survey  had  been  made  until  our  visit  in  July  1898. 
Even  then  our  time  might  have  been  wasted  if  it 
had  not  been  for  the  forediou^ht  of  the  Vali  of  Bitlis, 
in  sending  an  escort  of  50  soldiers  to  encamp  in  the 
crater  during  our  sojourn.    Under  ordinary  circum- 

*  Quart.  Jamm.  Giol.  Soc.  xL  333. 
t  Alpine  Jaum.  Aug.,  1887,  *93*300. 


114  Geology  of  Armenia 

stances  it  is  a  haunt  of  brigands,  and  indeed  we  found 
numerous  indications  of  a  recent  occupation. 

The  crater  is  nearly  5  miles  in  diameter,  yet  when 
the  mountain  is  viewed  from  Lake  Van  or  from  the 
surrounding  plains  it  appears  merely  as  a  great  ram-  * 
part  (rising  about  4000  feet  above  the  lake^  and  no 
one  would  suspedl  the  existence  of  one  of  the  largest 
craters  in  the  world.*  The  crater-rim  varies  so  mudi 
in  level  that  the  circular  form  only  becomes  apparent 
when  it  is  overlooked  from  a  greater  height,  e.g.^  from 
Sipan  on  the  N.E. 

The  outline  of  the  crater  is  nearly  circular;  the 
long  axis  runs  E.N.E;-W.S.W.,  and  measures  close 
upon  5  miles  (8405  yards) ;  the  slightly  shorter  axis 
amounts  to  790^  yards.  These  results,  together  with 
our  plan  of  the  interior,  were  arrived  ^t  after  a  series 
of  careful  observations,  which  1  made  with  Stewart's 
Telemeter  on  measured  bases,  combined  with  compass 
readings  by  Mr.  H.  F.  B.  Lynch  from  numerous  points 
within  ^  the  crater  and  on  its  rim.  Our  genetal 
impression  at  first  was  that  the  crater  was  very 
elliptical;  this  was,  however,  an  optical  illusion, 
due  to  the  fadt  that  the  west  and  east  points  are 
the  lowest  on  the  crater-rim  (8iao  and  81^0  feet 
respedUvely),  while  the  highest  ana  most  precipitous 
wuls  lie  to  the  north  and  south  (9900  and  9434  feet 
respedUvely. 

In  regarding  Nimrud  from  the  exterior,  the  a(ftual 
crater-wall  is  seen  to  rise  rather  suddenly  from  the 

broad  and  gently  sloping  lava-plains  which  encircle 
the  volcano.    This  charaderistic  is  especially  notice- 

*  The  basin  of  the  cntter-Iake  of  Oregon  has  a  diameter  of  a  Uttk 
over  5.^  miles ;  the  crater  of  Jengger  in  Java  is  5.75  miles  across;  the 
Lago  di  Bolsena  is  an  elfiptiad  crater-lake  s8  miles  in  circumference, 
wiSk  a  larger  diameter  of  10.25  mfles.  The  aUdenu  of  Palma  and 
Tenerife  are  respectively  4.5  and  11.75  miles  ii^  diameter. 


The  Simrud  Volcano  115 

able  on  the  east  and  west^,  x>n  the  south  side  the  bold, 
trachytic  mass  of  the  Keikur  Dagh  has  adled  as  a 
dam    to    the    lavas    of    Kimrud.^     This   strikingly 

*  The  two  specimens  (Nos.  106  &  107)  gathered  by  me  from  large 
blocks  at  the  foot  of  the  ea.stern  sc  rees  of  Kerkur  afree  closely  with 
-others  (Nos.  108  &  109)  taken  by  Colond  Elliot  nom  the  summit 
— the  qiBly  place  where  the  rock  can  be  obtained  in  Hiu.  The 
first  sp^men  (No,  106)  is  a  pale  grey,  compact  avoitb-trachytb, 
showing  numerous,  dark  grey  spots  with  a  dull  crimson  nucleus 
(up  to  3  mm.);  glassy,  unstriated  febpars  (up  to  3  mniL)  also  Se 
scattered  in  tne  porous  gioundmais.  Under  the  micnMOO^»  tlie 
SANioiNB  idiomorphs  are  of  prismatic  habit,  sometimes  in  small 
groups ;  the  crystals  have  sometimes  been  broken  and  the  fragments 
slightly  separated  by  groundnuss.  Avoin  is  only  present  unaltered 
when  enclosed  in  febpari  but  it  is  dearly  represented  by  psmdomorphs 
qf  hematite  associated  with  magnetite.  The  onouNDMASS  is  a  micro- 
fdt  consbiing  mainly  of  fdsparia^s  (sanidine)  in  fluidal  anangement, 
together  with  granules  and  needles  of  augite  (stained  a  dark  brown 
by  iron  oxide,  or  indeed  rendered  quite  opaque),  and  subordinate 
inagnetite  granules.  Tkidymits  b  the  chaxactenstic  imbricated  ag- 
gregates occurs  in  several  places. 

The  second  spcdmcn  fNo.  107)  is  a  very  similar  rodk,  bnt  is  a  pale 
purplish  grey.  Under  tne  microscope  the  alteration  of  the  augite 
crystals  can  be  more  easily  traced,  for  a  nudeus  of  unaltered  green 
augite  b  sometimes  present  in  the  hematitic  pseudomorphs. 

On  the  south  side  of  Kerkur  some  low  cones  rise  out  of  the  basaltic 
covering  of  the  Rahva  plain,  and  probably  belong  to  the  Kerkur 
system*  The  rock  (No*  no)  is  a  vesicular  avgits-trachyte  ;  it 
is  light  grey,'  with  sparing,  ghssy,  unstriated  felspars  (up  to  ^  mm.), 
and  more  rarely  some  uack  specks  (augite).  The  tiny  cavities  are 
mostly  filled  with  a  whitish  brown,  mammillated  zeolite.  Under  the 
microscope,  the  sanidinb  idiomorphs  are  sometimes  in  ^ups  of  3  or 
4;  dear  on  the  whole  but  occasionally  seolitized  penpherally  and 
along  cracks.  Avoin  is  ptesent  in  small,  bright  bottle  green  crystab, 
endosmg  magnetite.  A  few  corroaed  little  crystals  of  olivine  are 
also  present,  someiriiat  resorbed,  as  in  some  of  the  Nimrud  rhyolites, 
vide  infra ;  and  in  aU  these  cases  it  has  probably  been  derived  from 
the  basalt  through  which  these  add  lavas  have  forced  a  passage. 
Large  grabs  of  iCAGimmti  with  a  border  of  hematite,  are  fairly 
numerous.  The  oaouNDMASS  is  a  microfdt  of  felspar  laths  (sanidine) 
and  augite  needles  (stained  brown  by  iron  oxide),  with  much  mag- 
netite in  granules  and  dust ;  some  interstitial,  pale  brown  or  colourless 
glass  is  offastonaWy  visible    The  whole  slide  is  strongly  impregnated 


ii6  GMogy  0/  Arm$nia 

isolated  moim^in  stands  <m  an  elliptical  base  (with  a 
N.W.-S.E.  axisX  and  is  cloaked  by  steep  screes.  The. 
summit  is  formed  by  a  number  of  conical  peaks, 
which  surround  a  grassy  upland,  accessible  only  froki 
the  north,  where  the  broad  lava-flows  of  Nimrud  were 
dammed  ud  and  forced  to  diverge  to  the  east  and 
west  of  Kencur.  The  eastern  lava-streams  created  the 
barrier  (670  feet  above  the  lake),  which  brought  Lake 
Van  into  existence.*    The  basalt  then  flowed  round 

with  s  pale  brown,  mdiaUy  flbrous  feolit^  (hatrolitb),  with  con- 
<  cntric  hnes  of  growdL 

On  dMomding  from  the  lUhvft  plain  to  the  plain  of  Mtiah  I  found  a 
somewhat  nnuiar  AVom-TiucHTn  (Na  1 11)  xisiDg  np  in  low  oonei 
from  the  tiuroundng  basalt  of  the  leoond  terrace.  It  is  a  p^ 
greentdi  grqr  rock,  very  porous  and  friable,  containing  numecodi, 
glai^,  nnstiiated  fdspan  (up  to  $  mm.),  and  a  few  black  specks 
(augite).  Under  the  mioosodpe,  the  aamDim  idiemorphs  are  of 
prismaitc  habit,  fresh  and  free  from  inclusions.  Auom  is  present 
in  a  few  bottle-areen  crystals,  usually  with  a  deep  resorbcion-bocder, 
or  even  rendsredquite  cqpa^ue,  and  cmlir  teoognisabte  fay  tfie  outlines. 
The  oiODiiDifASS  is  a  vencuhr  microfelt  of  fdqMiHaths  (sanidine). 
needles  and  granules  of  aqdte  (stained  brown  bf  iron  oxide),  and 
magnetite  in,  dust  and  granules.  Numerous,  water-clear  jiphxrulitis 
are  present,  down  to  rather  small  dimensions,  often  showing  con* 
cntric  lines  of  joowdi;  a  partade  of  fdspar  or  aqgite  frequent  acts 
atf  a  nucleus.  There  is  a  good  deal  of  otaoge  staming  of  iron  oxide, 
caused  by  weatheriBg. 

*  The  present  lakes  of  Armeiiia  hare  arisen  mainly  owiog  to  the 
obstructive  influence  of  lava-streams.  In  diis  way  a  barrier  has  bete 
formed  across  a  valley,  ponding  back  die  natinal  drainage  of  thie 
basin.  Lake  Urmia  is  the  only  notable  exception,  for  it  oocqpies 
a  shallow  depression  cf  the  plateau.  The  southern  part  of  Lake  Van 
is  evicientty  the  eastern  section  of  the  great  dqwession  wfaidi  forms  the 
plain  of  Mush,  sharply  smarating  the  senated  Tausic  heidits  from. 
the  Armenian  pbteau.  The  noruera  and  shallower  part  of  the  lake 
is  formed  by  the  submeq^ed  valley  of  die  Bendimahi  river.  The 
drowning  of  tUs  double  valley  and  its  separation  fr^m  the  Mush 
depressioo  was  caused  hf  the  obstructive  mflnenoe  of  the  lava-flows 
from  NuBTud. 

Lakes  SenapL  (Gokcha)  and  CSiaUir  simihrly  owe  dieir  origiB 
to  lava-baniers,  birt  thqr  are  each  diafaied  by  a  river  at  the  narrow 


The  Nimrud  VoUmo  117 

to  fonn  tiie  broad,  basaltic  Rahva  plain  (5600  feet 
above  the  seal  which  lies  between  Kerkur  and  the 
head  of  the  Bitlis  valley.  The  basalt  has  flowed 
down  this  valley  for  a  considerable  distance,  uid 
form;  the  castle-rock  of  Bitlis,  in  a  manner  exadly 
analogous  to  that  of  Old  Akhlat  (p.  98).  Ainsworth  ^ 
states  that  this  lava  is  **red  or  brown,  generally  a 
light,  friable,  porous  rock,  but  in  some  places  also 
cpmpEuft  and  augitic,  passing  into  bi^alt  '* — a,  descrip- 
tion, though  vague,  yet  suitable  to  the  lava  forming 
the  Rahva  plain.  Ilie  lava-streams  which  swerved 
roimd  the  western  end  of  Kerkur  contributed  to  form 
this  plain;  the  main  stream,  however^  flowed  down 
into  the  extensive  plain  of  Mush  (4630  feet  at  the  pool 
6i  Norshen).  The  low  cliff  (fault-scarp)  of  the  terrace 
which  lies  immediatdy  west  of  Kerkur  is  composed  of 
a  slaggy,  rhyolitic  augite-trachyte  (No.  ii3},t  pro- 
aid.  The  eipbuuuioo  of  tfiii  icmiewhat  unuiOAl  dKOinitiiiice  ieems 
to  be  that  the  lutonl  oodet  of  die  drowned  Ttlleyi  is  tcOl  obetnicted 
by  high  yolcanic  biirien,  and  that  the  Zaaga  tbnsi  and  Oialdir-chai 
lespecdvdjr  are  due  to  die  overflow  of  theie  aaftdial  teiervoiri  at  their 
tpperendt. 

^  Ihnfds  m  Asm  UiMr^  tie  ii.  380. 

t  The  rock  is  dark  i^reenbh  frej,  wKh  narrow  |rarplish  brown 
streaks;  comfMCt,  with  urrqpdar,  iq>lintcrjr  fracturv.  It  contains  a 
fiew  dauy,  unstriatcd  felspars  (up  to  3  nun.),  and  some  histroos,  bkck 
necks  of  aogtte  (up  to  i  nim.)»  with  an  occasional  6aginient  of 
dttfc  brown  lava.  Under  the  nucrosoope,  the  felspar  idiomorphs  are 
ttsin]^  SAKiDUrx  of  prismatic  habit,  fredi,  .and  free  from  indnsions, 
but  a  few  tabolar  OTStals  of  ouooctasi  are  present,  much  honey- 
oombed  with  tndnsions  of  dusty  i^ass,  and  abo  didosin^  grains  of 
a^gite.  Auom  is  present  xt  psle  green  idiomoiphs,  sometimes  feebly 
nieochroic;  usually  quite  fresh,  pot  sometimes  with  a  resorbdon 
border;  endoses  magnedte.  The  oaOunmiASs  it,  compact  and 
ciyptocrystalline,  with  dusky  stroda  caused  by  crowded  aystsllites 
and  margariles.  Probably  some  devitrilicadon  has  taken  p*aoe. 
Numerous  fdspar-bths  and  less  abundant  augjte-needles  are  ran^ 
akmg  die  Unes  of  flow,  which  bend  round  the  phcpioaysts,  fcmnug 


ii8  Geology  of  Armenia 

bably  belonging  to  the  older  lava-flows  of  the  Nimrud 
volcano. 

I.    The  Interior  of  the  Crater. 

The  most  striking  feature  in  the  vast  cauldron  is 
the  great  semicircular  lake  occupying  about  3/8  of  the 
area.  It  lies  in  the  western  half  of  the  crater,  and 
is  Iiemmed  in — excepting  on  the  east — by  the  steep 
precipices  and  screes  of  the  crater- wall,  towering 
nearly  2000  feet  above  the  waters  of  the  lake.  The 
sc  rees  somewhat  recall  those  of  Wastwater  in  Cum- 
berland, though  on  a  far  grander  scale,  and  sink 
down  so  abruptly  into  the  lake  that  it  would  be 
impossible  to  make  a  complete  ciicuit  round  its 
shores. 

The  lake  is  doubtless  of  great  depth,  for  its  waters 
^pear  an  intense  indigo-blue  when  seen  from  above. 
The  level  has  been  rising  of  late  years  just  as  in  Lake 
Van,  for  some  of  the  birch  trees  (about  10  years  old) 
along  the  eastern  shore  have  been  totally  submerged, 
probably  owing  to  greater  aAivity  of  mineral  springs. 
The  western  shore  is  only  accessible  by  a  small  head- 
land of  olivine-basalt.  {No.  90),*  jutting  out  into  the 

*  The  rock  is  a  dark  purple,  with  abondaat  minute  steam-pores 
elongated  along  the  lines  of  flow.  It  contains  numerous  stnated 
felspars  (up  to  10  mm.),  and  sparing  dark  green  olivines  (up  to  x 
mm.).  Under  the  microscope,  the  plaoioclasb  (labradoritb  about 
Abi Anj)  is  of  priatnaticJiabtt,  very  fresh  and  free  from  inclusions,  but 
much  corroded  by  the  groundmass  and  partially  melted  down.  The 
OLiviME  is  in  very  pale  yellow  idiomorphs,  often  in  groups  of  s  or 
3,  associated  with  granules  of  magnetite;  quite  fresh  except  for 
occasional  dark  brown  stains  of  iron  oxide.  Apatite  occurs  in  lane 
idiomorphs,  fresh  and  unstained,  slightly  more  abundant  than  the 
olivines,  to  which  they  approximate  in  size;  under  a  high  power 
minute  rod-like  inclusions  are  visible,  but  in  no  definite  arrangement. 
The  GROUNDicASs  is  a  microfelt  of  tiny  plagiodase-laths  (andesine^ 
showing  fluidal  arrangement,   numerous  augite  granules,   stained 


The  Nimrud  VoUmo  1x9 

lake,  just  below  the  western  col  ci  the  crater-walL 
This  promontory  is  part  of  a  shdf,  or  inner  crater-rim, 
which  exists  in  greater  perfection  on  the  S.W.  side, 
at  a  height  of  about  800  feet  above  the  lake.  It  is 
there  about  aoo  yards  in  width,  and  is  covered  by 
a  sheet  of  basalt,  with  the  usual  smooth,  brown, 
weathered  surface  as  on  the  crater-edge  above  it  (vide 
infra).  Along  the  middle  of  this  terrace,  and  parallel 
to  its  edge,  there  lies  a  long,  narrow  ridge,  dying 
out  at  both  ends.  It  is  evidently  a  dike  of  acid  lava, 
for  I  could  clearly  see  obsidian  glittering  in  the  sun- 
light when  I  looked  down  upon  it  from  the  crater-rim. ' 
It  probably  belongs  to  the  last  period  of  volcanic 
adhvity  of  Nimrud,  and  its  parallelism  to  the  crater- 
rim  would  suggest  that  its  extrusion  was  comiedled 
with  the  sinking  down  of  this  terrace. 

The  lake  is  bounded  on  the  east  by  the  various 
lava-streams  which  have  filled  up  the  eastern  half 
of  the  crater,  and  projed  into  the  lake  in  bold  cliSs 
about  300  feet  high. 

At  first  sight  this  eastern  half  of  the  xrrater  appeared 
to  be  nothing  but  a  tumbled  and  confused  mass  of 
lava,  yet  on  closer  examination  I  found  that  it  may 
be  divided  into  three  parts: — ^northern,  central,  and 
southern — each  with  a  greater  extension  from  east 
to  west  than  from  north  to  south. 

The  Central  Area  is  rather  larger  than  the  other 

with  iron  oxide,  and  abundant  magnetite  granules. 

In  the  olivine-basalt  collected  bjr  AUch  from  the  bottom  of  the 
Tandurek  crater,  over  ^000  feet  deep  (a  volcano  between  Sipan  and 
Ararat),  and  deticribed  by  Becke  piis  No.  47,  Abich,  Ge^.  Forsch.  in 
den  kauk.  Zandem^  iii.],  apatiti  abo  occurs  ^'  in  short,  thick  crystals, 
pyramidal  at  each  end,  mostly  intergrown  with  the  olivine."  But 
this  rock  differs  from  the  preceding  in  containing  phenocrysts  of 
augite.  In  some  of  the  bflmlts  of  the  Central  Kanbach  [Thost, 
Abhandl.  Sendmh.  Naturf.  Ges.  Frankfurt  a/M.,  XVHI,  ii.  257] 
APATITE  is  similarly  present  in  large,  dusty  sections. 


I20  Geology  of  Armenia 

two,  and  is  dominated  by  a  terraced  peak  (with  a 
nearly  level  summit,  i8o  yards  from  north  to  south), 
marking  the  point  of  emission  of  the  pasty  lava, 
which  spread  out  in  a  billowy  mass.  The  eastern 
part  of  the  central  region  has  had  its  continuity 
interrupted  by  a  subsequent  explosion,  for  the  hill 
falls  away  precipitously  in  this  diredlion  towards 
a  flat,  sandy  depression  of  pumice  dust,  with  a 
shallow  lake  in  its  centre  (see  plan).  The  higher 
ground  east  of  this  lakelet  extends  right  up  to  the 
E.  and  S.E.  part  of  the  crater-wall,  and  was  probably 
once  continuous  with  the  lava  of  the  central  peak. 
The  S.E.  region  of  the  interior  is,  however,  so  thickly 
cloaked  in  pumice  dust  as  to  obscure  the  adlual 
nature  of  the  ground. 

The  Southern  Area  has  been  formed  partly  at  the 
expense  of  the  central  region.  Here  a  large  crater 
(about  li  mile  in  diameter)  has  recently  been  formed 
and  breached  on  the  west  by  a  broad  lava-flow  ex- 
tending into  the  lake.  Only  the  eastern,  semicircular 
wall  remains  in  place ;  it  is  of  no  great  height,  and 
consists  of  pumice  dust.  The  lava  which  issued  from 
this  secondary  crater  is  a  glassy  rhyolite,  forming 
a  confused  mass  of  brown,  angular  boulders,  rough 
and  barren,  unrelieved  by  any  vegetation  to  tone 
down  their  desolate  appearance.  Tne  lava  has  been 
so  lately  emitted  that  there  has  been  no  subsequent 
shower  of  volcanic  ash  to  cover  up  its  ruggedness.  In 
all  probability  it  forms  part  of  the  latest  eruption 
of  Nimrud,  which  was  chronicled  in  144 1  {vide  infra). 

The  Northern  Area  consists  of  an  extensive  mass 
of  lava,  so  much  cloaked  in  pumice  dust  that  no  clear 
point  of  emission  is  any  longer  visible;  but  the 
mounds,  when  viewed  from  a  greater  height,  seem  to 
run  in  roughly  concentric  circles,  with  a  central  point 
ia    the   N.E.   part  of   the  crater.    This  lava-sheet 


The  Nimrud  Volcano  121 

extends  close  up  to  the  E.  and  N.E.  part  of  the  crater- 
Irall,  but  between  it  and  the  northern  precipice  there 
lies  a  smooth  depression — 30-40  yards  in  width — 
which  may  perhaps  have  been  occupied  by  an  arm 
of  the  lake  at  the  time  of  emission  of  the  lava.  A 
narrow  valley  also  separates  the  lava-sheet  of  the 
northern  area  from  that  of  the  central  region. 

Rounded  cliffs,  about  300  feet  high,  covered  with 
undergrowth,  face  the  lake.  The  surface  lava  (No. 
69)  b^mes  so  highly  spongy  and  vesicular  all  along 
the  shore  that  it  seems  to  me  probable  the  lake  may 
have  been  in  existence  when  the  lava  was  erupted; 
and  the  coarse,  open  texture  of  the  rock  would  be  due 
to  the  volumes  of  steam  generated  by  the  cantojSt 
of  the  molten  lava  with  the  water  of  the  lake. 

This  superficial,  spongy  pumice  graduallv  passes 
downwards  into  a  compact  augite-rhyolite  (Nos.  67 
&  68)* 

*  The  first  spedmen  (No.  67)  shows  naxrow,  dark  purple  bands 
altematiiig  with  broader,  pale  greenish  byers.  Some  minute,  lustrous, 
blade  spedci  (augite^  are  visible  in  the  green  bands ;  the  purple  bands 
are  even  doser-gramed.  Under  the  microscope,  the  I»nds  consist 
of  fanlike  tufts  of  felspar  (sakioihx)  aud  auoitb  needles,  radiating 
from  the  exterior  to  the  interior  of  each  band.  In  the  purple  bands 
the  augite  needles  predominate  over  the  felspar,  and  vice  versa  in 
the  green  bands.  The  only  phenooyst  is  an  irr^;ular  grain  of  biotzte 
with  well-marked  cleavage. 

In  a  more  compact  spedmen  (No.  68),  from  a  large  block  probably 
derived  by  one  of  the  numerous  explosions  from  a  greater  depth, 
the  greenish  grey  rode  shows  an  ill-ddSned  mottling,  due  to  quartzose 
and  felspathic  patches,  but  no  banding  is  present;  the  fracture  is 
subcondioidaL  It  contains  somedassy,  unstriated  felspars  (up  to 
6  mm.),  and  abundant,  minute,  black  spedcs  (augite).  Under  the 
microaoope,  the  sanidinx  idiomorphs  are  fresh,  and  free  from  in- 
dusioQS.  AuGm  is  present  In  a  few  bottle-green  idiomorphs, 
sometimes  twinned,  slightly  pleodiroic  (to  a  pale  greenish  yellow), 
and  endodng  magnetite  granules.  A  very  few,  smaD,  corroded 
crystals  of  olivine  are  also  present,  probably  caught  up  by  this  recent 
lava  on  fordng  its  way  through  the  (dd  basalt-plug  iriiich  blocked 


122  Geology  of  Armenia 

The  feirly  level  northern  area  has  been  covered 
up  by  pumice-dust,  favourable  to  the  growth  of 
vegetation.  This  mantle  of  volcanic  ash,  here  as 
elsewhere,  was  doubtless  the  result  of  violent  ex- 
plosions, to  which  the  numerous  Secondary  Craters 
bear  eloquent  testimony.  The  largest  of  these  is  the 
great,  breadied  crater  of  the  southern  area  (p.  120). 

Several,  minor  explosions  have  taken  place  in  the 
central  region.  A  deep  rift  in  the  S.W.  side  of  the 
central  p(sak  has  been  formed  as  the  result  of  one 
of  these  paroxysms  of  Nimrud.  Two  explosions  must 
have  successively  taken  place,  for  a  small,  circular 
crater  merges  into  a  larger,  elliptical  chasm.  It 
affords  an  excellent  sedtion  of  the  central  peak: — b. 
thick  layer  of  pumice-dust  overlies  a  selvage  of  ob- 
sidian, which,  on  being  traced  downwards,  becomes 
more  and  more  crowded  with  grey  or  pink  spherulites 
(in  general  not  larger  than  peas),  until  the  rock  (No. 
72)  *  is  composed  of  little  else.    Still  lower  down,  the 

op  the  crater-chimney  previously  to  the  latest  eniptions  {vUe  infta). 
The  GRouNDMAfls  is  a  micropoealitic  fdt  of  felspar-laths  in  sheaves, 
with  strings  of  minute  augite  fpranules,  and  a  good-  deal  of  quarts 
in  tiny,  irregular  granules,  occastonally  in  aggregntes. 

*  The  rock  is  a  mass  of  mutually  interfering,  dark  grey  spheruUtes, 
often  elongated,  lying  in  a  pale  pink,  homogeneous  matrix;  a  little, 
dark  green  glass  (brown  where  weathered)  occurs  sometimes  between 
the  h^er  spheruhtes.  A  laige  number  of  unstriated  felspars  (up  to 
3  mm.),  and  a  few  dark  green  augites  (up  to  a  mm.)  are  visible.  The 
rock  is  slightly  magnetic  (3"*).  Under  the  microscope,  the  felspar 
(sanidink)  sometimes  shows  sharp  ctyital-outlines,  but  occurs  mostly 
in  fragments,  either  angular  or  partially  rounded  by^  melting  down; 
sometimes  in  aggregates  of  as  many  aa  a  dosen  individiuds ;  oc- 
casionally enclosing  crystals  and  grains  of  augite,  as  well  as  inclusions 
of  the  groundmass.  Only  a  few  phenocrysts  of  jpreen  Auom  are 
present,  showing  a  broad  re8orbtion4>order  of  hematite  and  magnetite. 
The  8PHS8C7LITX8  somctimes  diow  spherical  shiinkage-cracks  near 
the  periphery,  and  not  infrequently  contain  a  fragment  of  felspar 
pt  augite  as  a  nucleus.  But,  while  tihe  spherulites  Aaw  the  ordinary 
centrifugal  structure,  the  fine-grained,  ayptooTstalline  OBoxirMDiCASS  ii 


The  Nimrud  Volcano  123. 

lock  is  a  bandedi  vesicular,  bluish  grey  augite*rhyolite 
(No.  73),*  weathering  a  pale  brown.  It  becomes 
rudely  columnar — ^the  columns  measuring  12-20  feet 
in  diameter.  The  lower  part  of  this  crater  is  con- 
cealed bv  steep  screes,  and  its  bed  is  ojbcupied  by  a 
shallow  lake  derived  from  the  melting  snows. 

On  the  western  part  of  the  terrace  encircling  the 
peak  we  found  a  small,  shallow,  elliptical  crater,  with 

composed  of  £Ems  of  fdspar-latfas,  radiating  from  the  exterior,  dusky 
outline  of  the  streaky  areas  towards  the  interior,  and  meeting  similar 
fans  from  the  opposite  side;  the  dividing-lines  are  marked  by 
hematitic  dust.  A  few  angular  fragments  of  rhyolite  are  present,  irith 
turbid  and  decomposed  felspar-lat£i.  The  groundmass  is  also  densely 
crowded  with  belonites,  granules  of  augite  (usually  stained  brown 
with  iron  oxide),  and  a  little  magnetite ;  quartz  granules  are  some- 
times present.  In  some  cases  a  Uttle  brown,  interstitial  glass  occurs 
between  the  sphenilites. 

*  The  rode  consists  of  pale  grey  and  brownish  white  streaks,  with 
many  irregular  steam-cavities  (mostly  about  la  mm.  in  diameter), 
slightly  elongated  along  the  lines  of  flow ;  the  internal  ridges  and 
the  strands  which  frequently  stretch  across  these  cavities  give  the  rock 
a  pasty  appearance,  like  dough,  to  which  it  probably  approximated 
in  consistency.  The  walls  of  these  cavities  are  finely  crystalline 
and  siliceous,  with  tin^  black  sped^s  (augite).  Glassy,  unstriated 
felspars  (up  to  6  mm.)  he  scattered  in  the  stony  groundmass,  together 
with  a  less  number  of  augites  (up  to  i  mm.),  which  are  surrounded 
by  brown  stains  of  iron  oxide.  Under  the  microscope,  the  sanidine 
idiomorphs  are  quite  fresh,  and  usually  free  from  indusions,  excepting 
for  occasional  granules  and  needles  of  augite.  Aucm  in  small, 
bottle-green  crystals  is  sparingly  present ;  slightly  pleochroic  (to  straw- 
brown),  and  occasionally  twinned;  much  resorbed,  sometimes  to 
opad^.  Magnetite  is  accessory.  The  grouiidmass  is  micropoedli- 
tic,  with  irregular,  dusty  quartzose  Ussera^  densely  crowded  with 
crystallitei  of  felspar,  belonites  and  margarites.  These  ieMatra  are 
separated  from  eadi  other  by  dearer  matenal,  marked  out  by  strings 
of  augite  needles;  in  this  base  there  occur  scattered,  small  fdspar 
laths,  abundant  granules  and  needles  of  augite  (mostly  stained  orange 
by  iron  oxide),  and  much  magnetite  dust  Under  a  hig^  power 
skdetal  augites  are  sometimes  visible.  The  steam-cavities  are  lined 
by  a  narrow  sdvage  of  tridtutb  tablets. 


124  Geology  of  Armenia 

its  long  axis  running  north  and  south ;  its  sides  and 
floor  consist  of  pumice  fragments  (No.  74)  and  dust. 

In  the  lava-sheet  just  N.W.  of  the  peak  a  nearly 
circular  pit  has  been  formed  by  an  explosion,  likewise 
without  any  effusion  of  lava.  It  is  about  200  feet 
in  depth,  and  measures  298  yards  in  its  longer 
diameter  from  north  to  south.  The  walls  of  this 
blowhole  are  absolutely  vertical,  excepting  on  the 
east,  and  even  here  I  could  not  descend  more  than 
about  50  feet.  At  this  depth  the  obsidian  selvage 
of  the  lava  was  becoming  more  and  more  crowded 
with  pink  spherulites  (No.  75) ;  *  doubtless  the  seAion 
would  have  proved  to  be  very  similar  to  that  of 
the  crater  on  the  S.W.  side  of  the  peak  (p.  122). 

In  the  northern  area,  just  north  of  the  shallow, 
elliptical  crater  on  the  east  side  of  the  central  peak 
(p.  123),  there  are  three  more  craters  in  series  along 
the  same  S.-N.  line  of  weakness,  slightly  interfering 

*  The  rock  is  a  black  obsidian,  slightly  streaky  in  places,  crowded 
with  pale  grey  or  pink  spherulites  (up  to  7  mm.),  with  mammillated 
surfaces,  sometimes  in  groups ;  the  pmk  colour  is  due  to  weathering. 
Crystals  of  unstriated  ^spar  (up  to  3  mm.)  and  dark  green  specks 
(augite)  are  present  both  in  the  glass,  and  as  nuclei  of  the  spherulites. 
Under  the  microscope,  the  sanidzns  is  fairly  frequent  m  tabular 
idiomorphs,  often  in  aggregates,  and  also  in  angular  n^gments ;  very 
fresh  and  clear,  only  occasionally  enclosing  augite ;  sometimes  zoned. 
AuoiTE  is  present  in  bottle-green  idiomorphs,  frequently  in  aggregates, 
sometimes  twinned;  slightly  {deochroic  (to  diQl  brownish  green); 
encloses  magnetite.  There  are  also  several,  much  corroded  crystals  of 
ouviNE,  probably  caught  up  by  this  recent  lava  on  forcing  its  way 
through  the  old  basalt  plug  (q>.  No.  68,  p.  X3x).  The  SPHXRULrrES 
show  the  characteristic,  radially  fibrous  structure,  but  are  turbid  and 
almost  opaque,  owing  to  wea^ering.  The  origin  of  a  spherulite  may 
sometimes  be  <ybserved : — the  glass  surrounding  a  gram  of  augite 
shows  a  dear,  anisotropic  halo,  with  a  dark  cross  under  crossed  nicob. 
The  GLASS  contains  abundant  belonites  and  longulites,  together  widi 
felspar  microlites  with  jagged  ends.  All  of  Aese  become  concentrated 
in  the  dusky  streaks,  whidi  sweep  round  die  phenocrysts,  forming 
'  eyes,'  but  extend  uninterruptedly  through  the  spherulites. 


The  Nimrud  Volcano  135 

with  each  other.  The  oldest  and  largest  lies  to  the 
north  of  the  others;  it  is  comparatively  shallow  in 
relation  to  its  width  (about  ludf  a  mile),  and  the 
crater- wall  is  low,  rising  only  slightly  above  the  level 
of  the  lava-plain,  through  which  this  crater  was 
drilled.  It  has  been  fill^  up  by  a  rough,  tumbled 
mass  of  glassy  rhyolite  (similax  to  that  of  the  southern 
area,  p.  120),  which  has  been  tmable  to  breach  the 
crater,  and  shows  a  slight  subsidence  in  the  centre. 
Moreover,  two  later  explosions  have  occurred,  each 
drilling  a  deep  hole  with  vertical  sides  through  this 
lava  on  the  N.W.  and  E.  sides  respectively  of  the 
crater  (see  plan).  On  looking  into  the  N.W.  pit  I 
noticed  in  its  lower  walls  a  rudely  columnar  strudlure 
oi  the  rhyolite. 

In  the  southern  wall  of  this  broad,  secondary  crater 
a  very  small,  shallow  crater  (about  ^  50  yards  in 
diameter^  has  been  blown  out.  Immediately  south  of 
this  hollow  a  larger,  but  incomplete,  semicircular 
crater,  open  to  the  south,  has  been  formed,  with  walls 
of  pumice  dust.  Here  again  a  rugged  mass  of 
rhyolite  and  obsidian  has  oozed  up,  precisely  similar 
to  that  in  the  larger  crater  just  described,  and  in  like 
manner  the  lava  did  not  exceed  the  limits  of  its 
crater. 

On  the  eastern  shore  of  the  large  lake  several 
upwellings  of  similar  viscous  lava  have  taken  place 
along  a  NJ^.E.-S.S.W.  line.  Firstly,  the  barrier 
between  the  large  lake  and  the  small  **  hot  '*  lake  has 
been  caused  by  a  large,  circular  boss  of  glassy  augite- 
rhyolite  (Nos.  80  and  81),*  with  a  smaller  one  just 

*  The  tnrface-nxik  (No.  80)  is  very  pale  grey,  rather  pnniceoiu, 
with  glassy,  imsfriated  felspars  (up  to  t  mm.^  and  less  nomeroiis,  tiny 
black  sp^ls  (aiigite)  in  a  matra  of  greenish  grey  fl^ass,  frothy  in 
places.  Under  ttie  microscope,  the  samidinb  occms  in  abnndant, 
tabular  idiomorphs,  often  in  groups  of  3  or  4,  quite  fresh,  and 


126  Geology  of  Amunia 

north  of  it  A  little  further  S.W.,  a  long,  peninsular 
masSy  with  two  humps,  projeds  into  the  lake,  and 
ends  in  a  group  of  little  islands  (see  plan).  All  of 
these  are  composed  of  the  same  glassy  augite-rhyolite 
just  described.  Evidently  the  lava  issued  in  a  very 
viscous  condition,  and  at  so  recent  a  date  that  these 
bare  and  rugged  masses  have  not  been  covered  up 
by  volcanic  dust,  and  must  therefore  be  classed  among 
the  latest  manifestations  of  the  adivity  of  Nimrud. 

The  so-called  "hot"  lake  must  at>one  time  have 
formed  part  of  the  large  lake,  and  became  separated 
by  the  upwelling  of  this  recent  lava,  which  probably 
belongs  to  the  1441  eruption.  The  lake  is  penta- 
gonal, and  about  700-800  yards  in  diameter.  On 
looking  down  upon  it  from  the  crater-edge,  it  shows 
a  strikmg  difference  in  colour  from  the  large  lake — ^a 
dull  yellow- green  contrasting  with  a  deep  indigo. 
Hie  colour  of  the  small  lake  is  due  not  merely  to  its 
shallowness,  but  also  to  a  Conferva^  which  luxuriates 
in  the  stagnant  water.  I  found  the  temperature  of 
the  water  to  vary  from  78®  to  81*^  F.,  at  10  a.  m.  on 

enclosing  augite.  A  few  crystals  of  ouooclabk  are  also  present. 
AuGiTS  occurs  in  small,  botde-green  oystals,  slightly  pleodiroic  (to 
olive-green),  and  enclosing  magnetite.  The  grounduass  consists 
of  water-cl^  glass,  crowded  wim  microlites,  scopulites,  and  skdetid 
crystals  of  felspar,  with  abmidant  tricfaites  and  longulites — all  showing 
a  fluidal  arrangement    Small,  irregular  steam-cavities  are  present. 

The  deeper  seated  rock  (No.  81)  is  a  well-banded  augite-rhyolite, 
with  dull*  black  bands  in  a  yellowish  grey  matrix.  The  layers  show 
a  tendency  to  separate  when  bent  or  contorted.  A  few,  glassyr 
unstriated  felspars  (up  to  2  mnh)  and  innumerable,  black,  glancing 
needles  of  augite  can  be  seen  with  the  lens.  Under  the  microscopey 
no  phenocrysts  occurred  in  the  slide;  a  micropoedlitic  structure 
is  well  developed  in  the  broader  buuls;  but  in  the  narrow  bands 
the  small  tesserae  of  the  mosaic  are  pressed  out  and  elongated.  The 
rather  clear  tesserae  CMf  nutow  fdspar*latfas  in  sheaves  are  outlined 
by  strings  of  tiny,  dark  green  augite  needles  (often  stained  brown 
by  iron  oxide),  together  witii  brMniab  duit.  In  the  dark  bands 
the  augite  needles  predominate. 


The  Nimrud  Volcano  127 

July  2y^ ;  this  heat,  however,  is  probably  due  to  the 
sun's  rays,  aided  by  rad»tion  from  the  rocky  shores 
and  from  the  bare  precipices  of  the  crater- wall,  rising 
to  2000  feet  above  the  lake.  In  a  shallow  lagoon 
on  the  eastern  shore  the  temperature  was  as  high 
as  p8^  F.  at  the  same  time  of  day.  An  intermittent 
emission  of  bubbles*  occurred  here  and  there  at 
intervals  of  a  few  seconds,  but  this  was  probably 
marsh-gas  arising  from  decaying  vegetation,  for  I 
noticed  a  precisely  similar  occurrence  in  Lake  Van,  on 
the  south  side  of  the  Kizvag  promontory.  Here  the 
rocks  of  the  sto^  bottom  were  also  covered  with 
yellowish  green  Cmfervce  in  slimy  masses,  and  there 
was  a  distindUy  marshy  smell  at  both  localities.  The 
water  of  the  ^^hot'*  lake  is  slightly  alkaline,  and 
mawkish  to  the  t^ste.  but  is  well  able  to  support  life, 
for  it  swarmed  with  freshwater  shrimps  and  larvae  of 
dragonflies,  and  a  large  frog  was  hopping  along  the 
shore.  Reeds  and  rushes  fringed  me  banks.  A 
sample  of  the  water  which  I  coUedled  was  analyzed 
by  the  late  Mr.  William  Thorp,  B.  Sc.:— 

"  Quantities  estiouited  in  parts  per  100,000. 

Chlorine  ...        «••        •••        •••  4* '5 

ouipnates**.        .••        •••        •••  ••. 

Nitrates    •••        •••        ...        .••  0.05 

Sodium  and  potasstnm  carbonates  91.13 

mMgJlttUL  •••  •••  ...  •••  ••• 

JLune       •••        •••        ...        •••  5*^ 

Iron  oondc          •.•        •••        •••  o.o8 

Silica       •••        ...        «•*       .••  13*^ 

Alumina   •••        •••        •••        •..  0*68 

Total  solids  in  solution  ...        ...  114.43 

Suspended  matter          a.  18 

*  It  was  probably  this  drcnmstanoe  that  caused  Major  Gayton 
{Alpifu  Jntrnal^  1887,  293)  to  state  that  **  in  and  round  tlus  tarn 
bubbled  up  numerous  hot  springs,  in  which  one  oould  only  just 
bear  to  keep  one's  hand.  Six  or  seven  other  tarns  lie  among  the 
hollows  at  the  bottom  of  die  crater."' 


128  Geology  of  Armenia 

The  water  is  sUg^Uy  alkaline,  bnt  the  ratio  of  the  potasaum  to 
the  sodium  could  not  be  determined.  It  was  rather  turbid  owing 
to  fine  fragments  of  v^etaUe  matter.  It  is  scarcely  conceivable  that 
it  can  possess  healing  properties."* 

The  water  of  the  large  lake  contains  much  less  solid 
matter  in  solution,  but  an  accident  (during  analysis) 
to  our  small  sample  *'  caused  the  loss  of  the  iron, 
alumina,  lime,  and  magnesia  estimations.  Some 
vegetable  matter  occurred  in  suspension"  (Z.c.).  Mr. 
Thorp's  partial  analysis,  however,  showed  that  the 
water  contains  (in  paxts  per  100,000) ; — 

Chlorme  ...        •••        ...        •••  ^''S 

Nitrates   •••        •••        ...        •••  o.o8 

Total  solids  in  solution 39-4< 

Suspended  matter          x.88 

The  only  organisms  which  I  could  see  in  the  large 
lake  were  freshwater  shrimps  and  long,  narrow 
leeches.  The  total  absence  of  pelicans,  cormorants, 
or  any  other  fish-eating  birds,  so  abundant  on  Lakes 
Nazik  and  Bulama,  is  highly  significant  in  this 
respe(%.  Partially  submerged  birdli  trees  on  the 
western  shore  of  the  "  hot "  lake  show .  that  here  too, 
as  in  the  large  lake,  there  has  been  a  recent  rise  in 
level. 

Numerous  angular  blocks  and  boulders  of  a  black 
"  breccia  "-basalt  (Nos.  91-95)! — ^^  this  term  may 

*  Lynch,  Arminia^  ii.  470. 

t  This  basic  glass  (No.  91}  is  duH  black,  compact,  and  slightly 
magnetic  (4*).  It  contains  numerous,  glassy,  striated  felspars  (up  to  a 
mm.),  lustrous,  dark  green  augites  (up  to  i  wm.\  with  an  occasional, 
minute,  ^enish  yellow  crystal  of  olivine.  Under  the  microscope 
the  rode  is  seen  to  be  a  rich  brown  glass,  with  small,  but  well-maiked 
SPHBRUUTBS,  and  with  a  streal^  iflow-structure  rouiid  the  larger 
crystals.  It  is  sometimes  nearly  c^Mujue  with  fine  dust,  and  at  other 
times  has  been  partly  devitrified  Tlie  glass  is,  however,  exceeded 
in  amount  by  the  crystals  it  contains.  Most  of  these  are  in  angular 
fragments,  but  they  are  clearly  identical  in  all  respects  with  the 


The  Nimrud  Volcano  129 

be   applied — lie   scattered  over  the  surface  of  the 


phenociyBtB  of  die  basalts  of  the  aater-iim  (vide  infra\  vit.^  pumio- 
CLA8X  (a  basic  labbadoritb  about  Ab3Aii4)  with  the  same  fine  and 
complicated  twinning^  pale  green  augitx  with  associated  magnetite, 
and  xaidy  a  pale  yellow  ou^inb  cxystal,  with  stains  of  iron  oxide 
a|i(mg  its  cracks.  The  abundant  lava-fragraents  enclosed  in  this  glass 
vary  greatly  in  sise,  but  are  fairly  uniforoi  in  character.  I  was  able  to 
cnunme  the  following : — 

(i).  A  compact,  greenish  grey  AUGtTE-TRACHrTB^  containing 
nwneionsy  i^assy,  unstriated  felspars  (up  to  3  mm.),  and  a  few  augites 
(up  to  I  mm.).  Under  the  microscope,  the  sanidinb  idiomorphs 
are  sometimes  in  groups  of  3  or  4  individuals,  often  moulded  on 
the  larger  augites ;  they  occasionally  include  part  of  the  groundmass 
as  well  as  augite  granules ;  often  with  a  peripheral  line  of  dust,  and 
fiwquent  orange  stains  of  iron  oxide  along  cracks.  Augite  crystals 
are  few  in  number,  bottle-green  in  colour,  and  slightly  pleochroic 
(to  dull  sap-green)  in  the  peripheral  part  of  the  crystals;  cl^^avages 
well  marked ;  often  with  a  nigged  resorbtion-border ;  associated  with 
ICAGVBTITB.  The  OBOUNDMASS  is  a  micrographic  felt  of  rather  stumpy 
felspar-laths  (sanidine)  and  quartz,  together  with  abundant,  irrqgular 
strings  of  ra^ed  granules  of  green  augite,  sometimes  stained  brown 
by  iron  oxide  almost  to  opacity,  and  magnetite  in  granules  and  dust. 

(ii^.  A  similar  fragment  to  (i),  but  rather  finer  in  grain.  Under 
the  nucroscope,  sakidinb  still  predominates  in  large,  tabular  idio- 
morphs, but  some  oligoclasb  with  fine  albite  twinning  is  also  present  \ 
both  sometimes  show  rounded  comen  owii^  to  a  certain  amount 
of  melting  down;  they  enclose  minute  ^pranules  of  augite.  The 
GROUNDBiASS  is  an  extremely  fine  micropcecilitic  felt  of  felspo-  micro- 
lites  (sanidine)  in  ill-defined,  qtiartsote  UsserOf  and  showing  fluidal 
anangement,  together  with  tiny  granules  of  au^te,  and  magnetite 
in  granules  and  oust.  The  rock  was  doubtless  originally  more  super- 
ficial than  (i^. 

(iii).  Similar  tp  ^),  bnt  die  fdspar  phenociysts  are  rather  turbid 
with  brownish  dust,  and  the  fel^ar^latfas  of  the  groundmass  are  larger. 
The  augite  occurs  in  strings  of  granules  and  needles,  sometimes 
stained  oran^  by  iron  oxide. 

(iv).  Similar  to  (ii).  Amoqg  the  felspar  ^henocrysts  a  fragment 
of  PLA610CLA8B  (amdbsikx  about  AbsAns)  is  also  present  The 
ORouNDXASS  18  microposdlitic  with  irregular  t^mrm — showing  felspar 
microlites  in  quartsote  areas — ^and  with  magnetite  in  dust  and  strings. 
Under  a  high  power  the  magnetite  is  seen  to  occur  in  tiny  needles, 
strongly  reoallinff  in  appearance  the  augite  microlites  and  needles 
of  the  Nimrud  niyolites  (pp.  lir-xsj  &  XS5X  but  no  au^te  is  now 


139  Geology  of  Armenia 

interior  of  fhe  oater  and  also  on  fhe  north-west 
terrace  of  the  crater-rim,  but  the  rock  occurs  nowhere 
in  situ.  It  consists  essentially  of  a  compadfci  basaltic 
glass,  containing  numerous  fragments  of  augite- 
trachjrte ;  in  f acSfc,  the  fragments  are  often  so  large  and 
abundsuit  as  to  firequentfy  exceed  the  matrix  in  mass. 
Probably  this  rock  formed  part  of  the  original  floor  of 
the  crater  previously  to  the  eruption  of  the  rhyolites 
of  the  intenor ;  and  the  blocks  have  doubtless  reached 
the  present  surface  as  the  result  of  the  explosions, 
whidi  have  formed  the  various  secondary  craters 
already  described  (pp.  122-125). 


II.    The  ExTBRioR  of  the  Crater. 

In  describing  our  circuit  of  the  crater-rim  of 
Nimrud,  it  is  convenient  to  take  for  the  starting-point 
the  highest  part  (9430  feet)  of  the  soudiem  edge.  As 
in  other  parts  of  the  volcano,  the  steepness  of  the 
external  slope  varies  diredtty  with  the  acidity  of  the 
lava;  here  the  rock  is  a  pale  grey,  banded  augite* 
rhyolite  (No.  82),*  with  a  selvage  of  gre^iish  Dlack 

pfcsent. 
In  all  these  fragments  except  (})  felspar  is  the  only  phenocryst 

*  The  rock  splinters  readily  into  slaty  fragments ;  the  layers  aie 
about  X  mm.  diick;  their  bounding-planes  are  white,  weathering 
to  a  pale  brown;  black  needles  (augite),  and  small  quartz  crystals 
rboth  up  to  o.^  mm.)  are  abundant  on  Aese  surfaces,  wherever  the 
layers  have  shghtly  8q>arated  to  form  shrinkage  cracks.  Sparing, 
glassy,  imstriat^  felspars  (up  to  3  mm.)  are  the  only  visible  pheno* 
crpts.  Some  steam-cavities  (up  to  10  mm.)  are  present.  Under  the 
microscope,  the  sanidinb  idiomorphs  are  of  prismatic  habit,  sometimes 
with  angles  rounded  by  coirosion;  occasionally  intruded  by  ground- 
mass;  encloses  augite.  Auom  phenocr^fsts  are  still  less  frequent; 
the  tiny  green  oystels  show  a  deep  resoibtion-border  of  magnetite  and 
hematite,  and  have  been  stained  yellow  by  iron  codde.  The  ground* 
MASS  is  a  micropoedlitic  mosaic  of  inegular  quartzose  tesserae,  crowded 
with   felspar  microlites;   abundant  needles  and  granules  of  mgjitit 


The  Nimrud  Volcano  131 

obsidian  (No.  83).*  This  lava-flow  forms  a  long,  arid 
slope  of  33%  extending  southwards  towards  Kerkur. 
Our  pathway  was  not  more  than  10  feet  wide,  and  on 
our  right  the  crater-rim  fell  away  abruptly  in  preci- 
pices down  to  the  large  lake,  2000  feet  below.  The 
crater-wall  appears  to  consist  wholly  of  rhyolite,  for 
Colonel  Elliot  visited  the  base  of  these  precipices,  and 
colledled  a  specimen  (No.  84)  of  a  compadt,  green 
augite-rhyolite,  with  abundant  sanidines  (up  to  3 
mm.)  and  fewer  augites  fup  to  2  mm.). 

A  few  minor  points  or  emission  rise  out  of  the  long 
southern  slope: — the  largest  is  a  beautifully  sym- 
metrical cone,  forming  a  conspicuous  landmark,  and 
still  further  down  a  group  01  three  shallow  craters 
and  two  low  bosses  could  be  discerned  from  our  lofty 
stand-point.  The  first  of  these  craters  is  breached  on 
the  N.W.,  the  second  on  the  N.£.  Still  lower  down 
a  linear  series  of  4  small  cones,  which  have  arisen 
along  this  N.-S.  line  of  weakness,  could  be  well  seen 
from  Tadvan. 

The  views  of  Nimrud  from  the  S.E.  (1.  c,  from  Tad- 
Van  and  also  from  Gozel-dereh)  clearly  show  the  large 
flow  of  rhyolite,  which  extends  from  about  the  middle 
of  the  southern  crater-rim.  It  soon  divides  into  two 
branches,  and — like  the  similar  spurs  at  the  S.E.  and 

(often  stained  dark  brown  by  iron  oxide,  or  resorbed  to  opacity)  are 
present,  chiefly  outlining  the  tesserae  in  irregular  strings.  A  fluidal, 
wavy,  banded  arrangement  is  very  distinct — the  darker  bands  bebg 
characterized  by  more  abundant  aiuite  grains  and  needles.  Sparing 
interstitial,  colourless  glass  is  visible  In  places  with  a  high  power. 

*  Obsidian  occurs  generally  as  the  selvage  of  the  rhyolites  of 
^mrud.  It  is  frequently  poxphyritic,  with  laq;e  felspars,  but  usually 
occurs  in  homogeneous,  black  or  ^eenish  black  masses,  with  fine, 
conchoidal  fracture.  Under  the  microscope,  the  colourless,  streaky 
GLASS  contains  abundant  mioolites  of  sakidins,  with  jagged  or 
battlemented  ends.  Rods  of  pale  green  or  brown  auofis  are  less 
frequent.  Under  a  high  power  the  parallel,  dusty  streaks  traversing 
the  glass  are  seen  to  be  crowded  wi^  trichites,  longulites^  &c 


134  Geology  of  ArmmUa 

its  broad,  polygonal  columns  (8-z3  feet  in  diameter) 
appearing  lilce  a  smooth,  mosaic  pavement  on  a 
gigantic  scale.  It  was  almost  impossible  to  obtain 
n&h  specimens  of  tiiis  basalt,  owing  to  the  geneml 
absence  of  accessible  diff*sedtions. 

The  gentle  slope  of  the  basalt  to  tiie  S. W.  is  only 
diversi^  by  a  group  of  4  or  5  dikes— the  largest 
being  about  10  feet  in  width— radiating  in  a  south- 
western diredion  for  about  300  yards  from  the  crater* 
rim.    The  augite*trachyte  (No.  88)*  of  these  dikes 

almost  oolonileM  ayilalt.  MAommTt  is  aooessorj.  The  orounik 
MASS  is  a  loosdj  knit,  Tiesiailar  microfdt  of  plagiodase-laths  (oligo- 
dase-aadesbe),  with  nuniitc^  interstitial  granmes  of  angite  and 
magnetite,  nuidal  struaige  is  well  maxked.  Much  pale  brown 
GLASS  is  present  with  small  aiutBVLiTBS,  round  each  of  iriiidi  a  sl^t 
Mraching  has  occnrr^;  a  few  forked  crystallites  axe  also  visible 
in  the  i^bus.  Small  frapsenti  of  an  augite-tiachyte  are  occasional^ 
pKsent,  as  in  the  **  breoda^'-basalty  No.  911  p.  isS. 

Another  specimen  (No.  97)  of  this  glosTi  dndery  basalt  from 
the  same  neighbourhood  is  more  weathered,  forming  a  friabley  ruddy 
blown,  tttff-hke  lock.  It  is  precisely  dmilar  in  all  other  respects, 
excepting  dut  ft  is  sli^t^  more  magnetic  (8*),  and  that  the  bottle* 
gieen  augite  shows  feeble  pleocfarmsm  (to  dull  ydlow-green). 

*  The  rock  is  a  pak  brownish  ydDow  mottled  with  purple— the 
yellow  tint  being  due  to  stains  of  iron  oidde.  It  is  only  very  feebly 
magnetic  (i^).  Glassy,  unstriated  felspan  (up  to  ^  mm.)  are  frequent, 
and  a  few^  tmy,  Vtuk  specks  (augife)  are  visible  with  the  lens.  Under 
the  microscope^  the  sanioins  occurs  in  two  generations  of  pheno» 
aysts: — (i)  laige  and  tabular,  (ii)  much  smaller  and  prismatic;  the 
former  sometimes  show  curious  interpenetration-twins ;  an  indistinct 
aoning  is  occasionally  present,  and  a  peripheral  line  of  fine  dust  is 
fre<fttent;  they  enclose  abundant  needles  or  rounded  prisms  of  green 
augite.  A  few,  tabular  idi<tanoiphs  ^  flaoioclasi  (oliqoclasx-amdb* 
suf  a),  with  very  fine  albite-twinning,  are  also  present  Auoin  is  not 
freouent  except  iriien  enclosed  in  fdspar ;  the  idiomorphs  are  small 
and  dull  green«  with  a  resorbtion-border ;  encloses  magnetite.  A  few 
scattered  grains  of  MAOHwnrB,  with  a  border  of  hematite,  are  also 
present.  The  gkoundmass  is  microoystalline,  consisting  of  sheaves 
of  long,  narrow  felspar-needles  (sanidine),  with  interstitial  granules  of 
augite  and  magnetite.     Accumulations  of  dust  frequently  occur  round 


The  Nimrud  Volcano  135 

has  forced  up  blocks  of  the  overlying  basalt  to  a 
height  of  about  10  feet 

To  the  north  of  the  western  col  (8140  feet)  the 
ground  rises  steeply,  forming  precipices  2000  feet 
above  the  lake.  At  the  same  time  the  crater-rim 
rapidly  widens  out  into  a  fairly  level  terrace  of  basalt 
stretching  northwards,  ^-^  mile  in  breadth.  This 
terrace  is  dominated  by  a  conspicuous,  ridged  peak 
(0  on  plan),  with  an  E.S.E-W.N.W.  axis;  when 
viewed  end-on  from  the  interior  of  the  crater,  it  has 
the  appearance  of  a  sharp  cone.  It  is  the  highest 
point  on  the  west  side  of  the  crater  (9670  feetj ;  it 
stands  back  about  200  yards  from  the  edge,  and 
presents  a  low  cliff  (evidently  a  fault-scarp),  on  its 
northern  side.  It  marks  a  secondary  point  of  eruption 
of  the  lava  (No.  98),*  an  andesitic  basalt  with  sparing 

the  larger  felspars,  which  doubtlest  acted  as  centres  of  cooKng. 
Tliere  is  much  yellow  MainiDg  by  iron  oxida^  A  few,  small,  irregular 
steam^holes  are  visible. 

*  The  rock  is  dark  purplish  grey,  and  somewhat  streaky  and 
dndery.  The  purple  tinge  results  from  the  reddened,  weathered 
lining  of  the  numerous,  minute,  irregular  steam-holes,  as  so  frequently 
happens  in  vesicular  lavas.  It  contains  abundant,  glassy  felspars 
(up  to  5  mm.),  sparing^  lustrous  au^tes  (up  to  3  mm.),  and  still  fewer 
olivine  granules.  The  rock  is  highly  magnetic  (36*).  Under  the 
microscope,  the  crystals  of  plagioclasb  (a  basic  LABiuiDOitin  about 
Ab3  An4)  occupy  more  space  than  the  groundmass ;  the  lai]^,  tabular 
idiomorphs  have  undergone  much  fractare,  and  partial  mdtu^^  down ; 
often  deeply  honeycombed  by  a  network  of  groundmass ;  thie  same 
complicated  twinning  as  in  the  similar  basalt  No.  96,  p.  133.  Auorrs 
in  pale  f;reen,  well-developed  idiomorphs  is  sparingly  present,  some- 
times twmned ;  some  corrosion  has  taken  place  except  when  enclosed, 
wholly  or  partially,  by  felspar.  Ouvmx  crystals,  somewhat  corroded, 
occur  very  seldom.  The  oeoukdicass  is  a  streaky  microfelt  of 
plagiodase  laths  (oligodase-andesine),  with  minute,  interstitial  grains 
of  augite  and  magnetite,  the  latter  predominating  in  the  darker 
streaks.  As  the  streaks  become  paler  they  show  an  increasingly 
loofier  texture,  with  spherulitic  or  fan-shaped  groups  of  felspar  laths, 
radiating  inwards  from  the  exterior  to  the  interior  of  these  lenticular 


136  Geology  of  Armenia 

olivine  (q>.  No.  96,  p.  133),  which  has  flowed  down  to 
the  plam  of  Mush  in  broad  lava  streams.  This  part 
of  the  crater-wall,  when  viewed  from  the  plain,  e.  g.^ 
from  Norshen,  projedls  as  a  colossal  bastion  from  the 
rampart  of  Nimrud. 

The  broad  N.W.  terrace  is  mudi  traversed  bv  deep 
cracks  and  fissures,  probably  due  to  earthquakes,  so 
that  it  was  not  possiole  to  nde  close  to  the  crater-rim. 
The  smooth,  brown  pavement  becomes  coated  in 
places  with  a  black  crust  or  frothy  scum  of  more 
cindery,  slaggy  basalt  (No.  99).*  The  continuity 
of  the  terrace  is  interrupted  on  the  N.E.  by  a  low 
ridge  running  from  point  a  on  the  crater-rim  north- 
wards for  about  2  miles,  and  ending  in  a  beautifully 
symmetrical  crater  (N)  about  ^  mile  in  diameter.  This 
ndge  is  due  to  the  tracfturing  of  the  basalt  sheet  by 
the  upwelling  of  a  very  recent  rhyolite,  so  viscous 
that  in  most  cases  it  has  been  unable  to  issue  from 
the  fissures  which  preceded  its  emission  (see  plan). 
The  small  crater  (N)  was  probably  originally  a 
rounded,  basalt  cone,  similar  to  several  others  to  the 
W.  and  S.W.  Its  summit  must  have  been  blown 
off   by   some   paroxysm   (perhaps  during  the    144 1 

areas.    These  fan-like  groups  iiiterfere  more  or  less  with  each  other. 

*  The  rock  is  dark  purple,  aad  minutely  vesicular.  It  contains 
abundant,  glassy,  tabuUtr,  itfiaced  felspar  crystals  (up  to  ^  ;nni.), 
and  a  few,  dark  green  augites  (up  to  i  mm.).  The  rock  is  highly 
magnetic  (32'').  Under  the  microscope,  the  same  constituents  are 
present  as  in  the  last  rock:— ^he  labkadorite  crystals  (showing 
exactly  the  same  characteristics)  also  occupy  more  space  dian  the 
groundmass;  the  pak  ydlowi  well-formed  crystals  of  oliyinb  are 
here,  however,  ratfier  more  abundant  than  the  pale  green  augitb, 
which  is  present  in  sikaller  crystals,  but  neither  is  frequent.  The 
GiouNDMAss  IS  also  similar  to  that  of  No.  98,  except  that  the  felspar 
laths  in  the  mutually  interfering,  spherulitic  and  fan-shaped  groups 
a»e  longer  and  narrower;  in  addition,  there  is  some  dusty,  pale 
brown  glass.  A  few  fragments  pf  compact,  crystalline,  trach3rtic  lava, 
similar  to  the  fngments  in  the  ''breccia "-basalt,  p.  128,  are  present. 


The  Nimrud  Volcano  i^y 

eruption),  and  the  crater  thus  formed  became  filled  by 
rhyolite  and  obsidian,  which  has  bteached  the  wall 
both  on  the  north  and  south.  It  is  the  same  lava 
as  in  the  fissures,  and  forms  a  tumbled,  barren  mass  of 
blocks. 

Qo  returning  to  the  Nimrud  crater  along  the  east 
side  of  the  ridge,  which  on  this  side  has  been  much 
fissured  by  the  upwellin^  of  rhyolite  (see  plan),  I 
noticed  two,  short  flows  of  black,  cindecy  lava.  Iney 
are  of  no  great  thickness,  yet,  owing  to  their  colour, 
they  are  conspicuous  landmarks,  even  from  Sipan, 
35  miles  distant.  A  similar  flow  of  black  cinder 
occurs  also  on  the  west  side  of  the  ridge»  issuing 
from  the  base  of  the  rhyoliiSc  mass,  whi(£  breached 
the  southern  wall  of  the  small  crater  {N).  This  black 
basalt-pumice  (Nos.  xoo-102)*  appearea  so  fresh,  not 
even  overgrown  with  an;j^  form  of  lichen,  that  it  was 
difficult  to  credit  the  testimony  of  one  of  our  zaptiehs^ 
that  this  part  of  the  mountain  had  not  altered  for 
at  any  rate  the  last  15  years.  Yet  it  is  highly  pro- 
bable that  these  features  were  the  result  of  the  IA41 
eruption.  The  restridted  flow  of  this  basalt-pumice, 
and  its  exadly  similar  composition  to  the  basalt 
through  which  the  rhyolite  forc^  a  passage,  seem 
to  me  to  indicate  that  these  shallow  cinder  flows  were 
merely  the  result  of  a  partial  re-fusion  of  the  basalt 
by  the  rhyolite. 

Whilst  the  terrace  to  the  west  of  this  ridge  is,  for 
the  most  part,  nearly  level,  its  east  side  is  steep, 

*  The  rock  is  a  his^ly  vencola^  blad^  vomice^  with  numefous, 
glaiqr,  ttriated  fdspan  (up  to  4  mm.),  and  irregular  cavities  with 
drawn-ont  threads  of  glw.  Under  the  microscope^  the  abundant 
GLASS  is  dark  brown,  pumiceous,  frequently  streaky  and  drawn  out^ 
with  some  mag^ietite  dust,  but  without  any  microlites;  The  pheno- 
oystB  are  identical  with  those  in  the  basalt  of  the  N.W.  termce,  No. 
99,  p.  136,  t/fs.,  L4BRADOIUTB,  AuoiTV,  and  ouvms,  all  with  the  same 
cha^cteristics  as  in  that  rode. 


138  Geology  of  Ammnia 

and  then  fhe  bioad»  basaltic  lava  streams  sweep  down 
with  a  slope  of  about  15^  to  Al^ilat,  in  complete 
continuity  with  the  slope  from  the  north  rim  of 
the  Nimrud  crater. 

On  reaching  the  cmter*s  edge  and  resuming  our 
circuit,  we  noticed  that  the  point  a,  which  was  a 
conspicuous  feature  on  the  rim  when  seen  &om  the 
interior,  is  in  reality  the  point  where  the  N.-S.  rid|[e 
intersect  the  Nimrud  crater.  The  surface-rock  is 
still  the  same  dark  purple,  cindery  basalt  (103),* 
and  here  I  found  numerous  fulgurites  (No.  104),  the 
walls  of  the  tubes  fup  to  12  mm.)  being  formed  of 
dark  brown  glass.  Beads  of  glass  (up  to  12  x  8  mm.), 
fused  by  the  fightning,  also  occur  at  this  point. 

The  northem  edge  of  the  crater  preserves  a  nearly 
even  level  for  a  considerable  distance,  and  consists 
of  tiie  same,  hi^ly  magnetic  basalt,  sloping  about 
15^  to  N.  and  N.E.  The  nature  of  the  deeper-seated 
part  of  the  basalt  flows  (No.  105)  t  was  revealed  by 

*  The  rock  contains  abundant  striated  felspais  (ai)  to  3  mm.),  a 
few  dark  green  augites  (up  to  s  mm.),  and  an  occasional  olivine  (i 
mm.)  in  a  minutely  vesicular,  dark  purple  groundmass,  wiA  grejjr 
streaks.  It  is  markedly  magnetic  (is^.  Under  the  microscopey  it 
presents  almost  the  same  appearance  as  the  basalt  of  tiie  N.W.  tetraoe 
(No.  99,  p.  136),  on  the  west  side  of  the  ridge,  m.,  with  phenocrysts 
of  LABBADORiTx  (about  Ab3An4),  Auom,  and  ounny  all  with 
the  same  characteristics  as  in  that  rock.  Apatitx  is  enclosed  bf 
the  olivine.  The  gbounoiuss  is  rather  opaque,  and  minutdy 
vesicular;  it  contains  a  food  deal  of  pale  brown  glass  interstitially  to 
the  fekpathic  streaks,  which  display  the  same  tendenqr  to  a  qdierumic 
and  radiate  grouping  of  the  felspar  laths  as  in  No.  99.  A  few  frag- 
ments of  a  stony,  trachytic  hva  are  also  present,  ronsisting  of  a 
oystalline  microfdt  of  plagiodase  laths,  with  strings  of  magnetite 
granules  (perhaps  in  part  resorbed  aqgite,  cp.  No.  83,  p.  131). 

t  The  rode  is  compact,  a  duD  puijpfish  grey,  widi  diin,  white 
streaks.  Only  a  very  few  minute  cavities  are  proent.  It  contains 
abundant,  (fhusy,  tabular  aystals  of  striated  febpar  (up  to  5  mm.), 
very  few  olivines  (up  to  s  mm.),  and  smaller  aiiigites.    The  rock  is 


The  Nimrud  Volcano  139 

larce  blodcs,  from  the  screes  between  the  **  hot  '*  lake 
and  the  inaccessible  precipices  above  it.  This  basalt 
is  more  compadl»  and  contains  more  olivine  than 
the  preceding,  superficial  specimens  of  the  Nimrud 
basut 

At  the  highest  point  (b ;  9903  feet)  of  the  Nimrud 
crater  the  basalt  sheet  is  interrupted  for  a  short  space 
by  the  imderlying  augite-rhyohte,  with  its  selvage 
of  obsidian.  Here,  as  on  the  south  rim  of  the  crater 
(p.  130),  the  different  nature  of  the  rock  is  associated 
with  an  increase  in  the  angle  of  slope,  mt.,  up  to  33^. 

However,  the  basaltic  sheet,  with  its  smooth,  brown 
pavement,  soon  re-appeared,  and  remained  the  surface 
rock  during  our  descent  to  the  eastern  col  (81^0  feet^, 
the  second  lowest  point  on  the  crater-rim.  Ine  cliff- 
sedion  here  shows  that  this  basalt  is  merely  the 
superficial  covering  of  the  older  rhyolite,  of  which  the 
crater-wall  is  mainly  composed.  Indeed,  its  eastern 
part  consists  entirely  of  rhyolite  and  obtidian,  and 
was  traversed  by  us  during  our  ascent  of  Nimrud 
on  coming  from  Akhlat.  Jiist  north  of  the  eastern  col 
the  basalt  sheet  is  only  15  feet  thick,  and  is  overlaid 
by  about  15  feet  of  loose,  incoherent  pumice-dust, 

obHj  very  dightly  magnetic.  Under  the  ndcrcMOope,  the  plagioclase 
crykals  (tABRiux>RiTB  abolit  Ab3 An^  present  the  same  dutfacteristics 
at  in  the  preceding  basalts,  but  the  large,  tabular  idiomorphs  have 
not  been  so  greatly  noneycombed  and  corroded ;  they  cndose  granules 
of  au^  and  magntdte.  The  olivinb  is  sHghtly  more  abuiidant  than 
the  pale  green  auoitb.  Apahtb  in  pale  brownish  crystab,  slightly 
pleochxoic,  is  accessory.  The  obooiidicass,  as  in  No.  989  is  a  stroJ^, 
bat  more  compact  and  crysttDine  microfelt  of  pbgiodase  laths 
(andesme),  with  minute,  interstitial  granules  of  augitc  And  magnetitty 
the  latter  predominathig  in  the  duskier  streaks.  Here  too  the 
felspar  laths  show  a  maiked  tendency  to  fonn  mutually  interfering, 
spherulitic  gioups,  particukriy  in  the  lighter,  and  more  crystalline 
streaks,  in  which  dieir  growth  has  extended  firom  the  margins  to 
the  centre  of  the  streak.  Both  these  fdspathic  stireaks  and  the  felspar 
.pbenocrysts  are  surrounded  bf  a  dusty  zone. 


I4P  Geology  of  Armenia 

evidently  derived  from  the  later  explosions  in  the 
crater  (see  p.  122  et  sqq.).  My  sketches  of  the  crater- 
wall  from  the  interior  clearly  show  the  contrast 
between  the  thick,  lenticular  sections  of  the  rhyolite 
flows,  which  built  up  Nimrud,  and  the  broad,  thin, 
even  layers  of  the  overlying  basalt  sheets. 

The  outer  slope  of  the  eastern  crater- wall  does  not 
exceed  17^,  and  only  amounts  to  12^  along  the  easiest 
approach.  A  laminated  deposit  of  white,  volcanic 
ash  has  in  many  places  reduced  the  original  in- 
clination of  the  rhyolite  and  obsidian,  which  are  now 
and  again  laid  bare  in  the  numerous  gullies.  It 
seems  to  me  probable— judging  from  the  low  angles 
of  slope  (17®}  in  comparison  to  those  (32°)  of  the 
crater-rim  in  the  north  and  south — that  the  ex- 
plosions, which  have  left  the  crater  in  its  present 
condition,  took  place  chiefly  in  the  eastern  and 
western  areas  of  the  original  crater  of  Nimrud. 

In  completing  our  circuit  of  the  crater- rim  we 
found  that  the  ground  rose  again  considerably  up 
to  three  rounded  peaks  (d  on  plan)  on  the  S.E.  edge. 
The  whole  of  the  S.E.  rim  is  so  much  cloaked  by 
volcanic  ash  that  it  was  not  possible  to  determine  the 
nature  of  the  underlying  rock.  A  specimen  (No.  87), 
colleded  by  Colonel  Elliot  from  a  fallen  block  at  the 
base  of  the  cliffs,  is  a  compacfl,  dark  grey  augite- 
rhyolite,  with  numerous  sanidines  (up  to  3  mm.)  and 
less  frequent,  smaller  augites.  The  cliff-secftion  shows 
that  it  is  free  fix)m  the  capping  'of  basalt,  which  is 
so  prevalent  on  the  north  and  west  sides  of  the  crater. 

A  sharp  descent  brought  us  to  the  south-east  col  (e  ; 
about  8000  feet),  where  the  path  from  Tadvan  enters 
the  crater.  From  this  point  a  steep  rise  of  over  1000 
feet  led  to  the  lofty  southern  rim,  the  starting-point 
of  our  circuit  of  the  vast  cauldron. 

The  exterior  slope  of  the  S.E.  crater-wall  down 


I 


Th0  Ntmrud  Volcano  141 

to  the  plain  of  Tadvan  is  nowhere  steeper  than  20^. 
Here  again  the  rhyolite  has  been  thickly  cloaked  in 
white  volcanic  dust;  and  the  crater- wall  has  been 
carved  out  by  temporary  torrents  (during  the  melting 
of  the  snows)  into  numerous,  deep  gullies  full  of  large 
boulders  of  lava. 

III.    Outlying  Features  of  Nimrud. 

Spurs  radiate  outwards  from  the  three  peaks  {d) 
on  the  S.E.  rim;  the  most  conspicuous  of  these  is 
a  knife-edged  spin:,  with  W -E.  axis,  rising  up  out 
of  the  broad,  littoral  plain  of  Lake  Van,  about  3 
miles  east  of  the  crater-rim.  Still  further  east,  on  the 
same  line  of  fradhire,  the  Kizvag  promontory  rises 
up  with  similar  abruptness,  and  juts  out  into  the 
lake.  It  clearly  marks  the  position  of  a  separate 
upwelling  of  rather  acid  lava  related  to  the  rhyolites 
of  the  Nimrud  crater.  The  summit  of  the  Kizvag 
ridge  shows  a  thin  selvage  of  obsidian,  passing  down 
into  augite-rhyolite  (No.  114),*  in  layers  inclined  yf 
S.S.E.    Still  lower  down  the  rhyolite  passes  into  an 

*  The  rock  is  pale  brown,  with  dark  grey  bands  closely  ranged 
together,  and  inducing  a  sla^  fracture.  Qassy,  unstriated  felspars 
(up  to  3  mm.),  with  a  few  black  specks  of  augite  (up  to  2  mm.) 
are  present  in  the  fine-grained  groundmass,  which  contains  a  few 
minute,  elongated  steam-holes.  Under  the  microscope,  the  sanidine 
occurs  in  vei^  clear  and  btah  i^omorphs  of  prismatic  habit,  some- 
times enclosing  augite  granules  and  particles  of  clear  glass.  Augitb  is 
present  in  a  few,  small,  green  crystals,  feebly  pleodiroic  (to  yellow 
green),  and  with  a  deep  resorbtion-border.  The  groundmass  is 
well  banded,  and  consists  of  an  irregular,  micropoecilitic  mosaic,  each 
quartsose  tessera  being  crowded  widi  felspar  laths,  and  outlined  by 
strings  of  augite  granules  and  needles;  the  green  au^te  has,  for 
the  most  part,  been  stained  a  dark  brown  by  iron  oxide,  even  to 
opacity.  The  dark  streaks  contain  more  augite  needles  and  magnetite 
dust  than  the  clearer  zones.  The  quartz  becomes  more  distinctly 
visible  in  places  of  looser  texture,  and  projects  in  tiny  crystals  into  the 
cavities. 


1 


142  Geology  of  AtHtenia 

augite-trachyte  (No.  115)9*  with  rudely  columnar 
strudhixe;  the  thick  columns  have  weathered  at  the 
siuface  into  rounded  points.  Near  the  base  of  the 
precipitouSt  southern  cuff  there  are  three  small  bosses/ 
in  which  the  columns  radiate  outwards,  curving  over 
at  the  top. 

It  seems  to  n)e  probable  that  when  the  lava  rose 
up  along  this  E.-W.  fissure  the  level  of  the  neigh- 
bouring ground  was  much  higher  than  at  p!resent» 
so  that  the  igneous  rock  only  became  exposed  to 
the  air  along  the  top  of  the  ridge,  where  the  glassy 
selvage  occurs.  Subsequently  the  softer  ground  (per- 
haps merely  volcanic  ash)  would  have  been  deluded 
away  by  the  waters  of  Lake  Van,  leaving  only  the 
hard  core  of  igneous  rock.  Similar  deiiuded  DOsses 
occur  in  Iceland,  e.  ^.,  the  Hljddaklettar  in  the  gorge 
of  the  Jokulsa  river,  which  has  washed  away  the 
''mould"  of  volcanic  ashes,  into  which  the  lava 
was  injecfted.f 

*  The  rock  is  dull  purple,  with  rough,  irregular  fracture.  It  is 
siightlf  magnetic  (s^).  Numerous,  iffiBasf  Cellars  (up  to  6  mm.),  and 
some  lustrous  augites  (up  to  a  mm*)  lie  m  the  greenish  black  ground* 
mass.  Small,  irregular  steam-holes  are  frequent,  and  are  surrounded 
by  a  dull  crimson  border,  which  gives  the  rock  its  purplish  tinge. 
Under  the  microscope,  die  fdspar  phenocrysts  occur  in  two  genera- 
tions : — (i)  large,  tabular  crystals,  showing  a  rounded  or  irregular  core 
of  PLAOiocLASK  (oligoclask),  suTTOunded  by  dear  sanidivs,  mA 
sharp  crystal  outline,  and  sometimes  zoned ;  the  core  frequently 
contains  augite,  and  is  often  markedly  honeycombed ;  (u)  sakidinb 
idiomorphs  of  prismatic  habit,  ve^^  free  from  indusions,  but  much 
broken  up,  owing  evidently  to  differential  movements  prior  to  final 
consolidation.  Aucrrx  occurs  in  pale  green  idiomorphs,  somewhat 
pleodiroic  {a  &  r— -pale  green,  ^ — straw-brown),  and  endosing  mag- 
netite. A  very  few  corroded  grains  of  ouvinb  are  present ;  where  one 
of  these  is  partially  enclosed  by  a  felspar  it  has  escaped  conrosion. 
The  ORoCNDMAss  is  a  foke  microfelt  of  felspar  laths  (sanidine),  with 
abundant,  pale  green  augite  microlites,  and  ma^etite  granules. 
Inegular  patches  have  been  stained  brown  by  iron  oxide. 

t  Figured  by  Tempest  AnderKm,  FatamicStudieSf  lii.  London,  X903. 


The  Nimrud  Volcano  143 

On  the  south  side  of  the  Ki^^  rid^  there 
extends  a  low  promoiitory  only  slightly  above  the 
present  level  of  Lake  Van.  It  consists  of  a  pale 
greenish  grey,  porous  augite-trachyte  (No.  116),* 
with  a  rudely  columnar  strudhire,  tm  columns  being 
about  2  feet  in  diameter.  The  lava  is  at  any  rate 
20  feet  thick,  for  this  is  about  the  level  of  the 
promontory  above  the  lake.  The  rock  weathers  to  a 
reddish  tint.  The  sporadic  occurrence  of  corroded 
olivine  in  these  Kizvag  tradiytes  as  well  as  in  those 
of  Kerkur  (p.  1%$)^  and  in  spme  of  the  Nimrud 
riiyolites  (pp.  |2i,  loj)  is  probably  due  to  the  travers- 
infi^  of  earUer  olivine-oasalts  by  the  later,  add  lavas. 

In  riding  along  the  low  coast  from  Tadvan  to 
KLzvag  the  first  ezpnosure  of  lava,  just  north  of  the 
Tadvan  peninsula,  is  a  dark  green  augite-rhyoHte 
(No.  ii7),t  overlaid  by  sheets  of  the  widely  prevalent, 
dndery  basalt,  weauering  a  rusty  brown«  The 
rh^olite  probably  belongs  to  the  older  flows  of 
Nimrud. 

The  basalt  sheets  alternate  with  thin,  laminated 
beds  of  pumice  nodules  and  dust,  which  are  some- 

^  The  rpek  if  rather  fiiable.  It  contains  abundant,  ffMty^  nn- 
flnaled  felspars  ^up  to  3  nun.),  and  nmnerous,  dark  gieoi  augites 
(m  to  I  mm.),  with  small,  pmplish  brown  patches.  There  is  a  slight 
.  eierfescence  with  add.  Under  the  microscope,  the  SAMiDtHB  is 
Nicsent  in  prismatic  idionunphs,  occasionany  m  aggregates;  very 
nesh;  encloses  augite  needles  and  small  partides  of  glass  with  a 
bubble.  Auom  oocurs  in  bottle-green  grains,  seldom  showing 
crystil-ontlines ;  slightly  pleochroic  (to  a  dull  ydiow-green) ;  some- 
times in  groups  of  4  or  5  indinduab.  A  very  few,  corroded,  little 
aystals  of  ouvnnc  axe  also  pr^ent  MAONBTnrs  is  aooessoiy.  The 
GRouirDiCASs  is  vesicular  and  micropoecilitic ;  the  ill-defined,  irregular, 
quartsose  tessens  are  roughly  Outlined  by  strings  of  ainmdmit  needles 
and  granules  of  green  augite,  sometimes  stained  Inown  by  iron  oxide. 
Some  TUDYMiTB  is  irisiblei  eq;iecially  when  projectinig  slightly  into 
cavities. 

t  The  rock  is  compact^  with  darker  streaks  and  someiAat  fissile. 


146  Geology  of  ArtMnia 

much  is  pidluresquely  situated.  This  dark,  scoria- 
ceous  tracnyte  has  been  largely  concealed  by  the  later 
basalt  sheets,  and  by  volcanic  ash.  It  probably 
issued  from  a  conspicuous,  breached  crater  (0),  about 
5  miles  to  the  S.W.  Between  this  old  crater  and 
the  crater-wall  of  Nimrud  there  is  a  bold,  rounded 
buttress  (P),  similar  to  the  one  at  the  S.£.  anete  (p. 
132).  In  the  angle  between  this  spur  (probably  of 
acid  lava)  and  the  eastern  col  (c)  of  Nimrud  there 
lies  the  little  village  Teghurt,  the  seat  of  a  disastrous 
earthquake  on  A^y  3C^,  i88i.  It  appears  also  to 
be  called  Tapavank  according  to  Nolde,^  who  gives  it 
the  height  of  8000  feet ;  it  cannot,  however,  l^  more 
than  6500  feet  above  sea-level,  for  it  is  situated  at  the 
foot  of  the  crater- wall. 

The  extensive  sheets  of  basalt  which  have  flooded 
the   country   down   to  Akhlat,   sweeping  round  to 

febpan  (up  to  6  mm.),  and  some  black  pyroxenes  (up  to  z  mm.^  with 
steel-blue  lustre.  It  is  onljr  feeblj  magnetic  (x**).  Unda  the 
microscope,  the  saniddix  idiomorphs  are  of  prismatic  habit,  veiy  htah 
and  fairly  free  from  inclusions,  but  sometimes  containing  augite 
granules  and  small  patches  of  ^roundmass.  EMSTATm  occurs 
sparingly  in  pale  yellowish  green  idiomorphs,  occasionally  somewhat 
corroded;  some  of  the  crystals  have  been  so  much  resorbed  as  to 
be  nearly  <^>aque ;  it  is  only  slig^Uy  pleochroic  (to  pale  bluish  green). 
Auorrs  in  smaller,  pale  green  oystala  occurs  stiU  more  sparingly. 
The  QROUNDMASS  consists  of  pale  grey  streaks  alternating  with  &xk 
bands.  The  former  have  resisted  weathering,  and  consist  of  a  dusty, 
devitrified,  cryptocrystalline  base,  crowded  wiA  minute  crystallites, 
and  enclosing  numerous,  small  felspar  laths  (sanidine),  and  green 
augite  prisms,  slightly  pleochroic  (to  yellow-green)— both  quite  fresh. 
The  daxk  bands,  on  the  other  hand,  are  more  crystalline,  and  mudi 
weathered;  they  are  composed  of  mutually  interfering,  spheruUtic 
groups  or  sheaves  of  felspar  laths,  with  augite  needles  and  dcdet^ 
needles  of  magnetite  (with  the  gridiron  arrangement).  The  augite  in 
these  darker  bands  has  weathered  considerably,  and  is  mudi  stained 
with  iron  oxide,  which  has  permeated  these  parts  of  the  rock.  A  few 
iiregular  cavities  occur,  occasionailf  fiHed  wuh  caldte. 

*  IUmnachInurirahU»*t  &C.,  Brunswick,  1895,  '4i- 


The  Nimrud  Volcano  147 

the  S.E.  to  form  the  coastal  plain,  evidently  issued 
from  the  north  lip  of  the  Nimrud  crater  (between 
points  a  and  b\  and  flowed  down  between  the 
rounded  spur  01  Teghurt  and  the  two,  colossal, 
northern  buttresses  {Q  &  J?),  which  are  such  conspi- 
cuous landmarks  from  Akhlat  or  from  Lake  Nazik. 
Contrary  to  what  might  be  expedled,  these  buttresses 
slope  gently  southwards  towards  Nimrud,  presenting 
steep  cliflFs  to  the  north.  This  inward  slope  may 
perhaps  be  accounted  for  by  a  settling  down  of  the 
huge,  central  cone  of  Nimrud,  when  the  eruptive 
forces  had  died  down  previously  to  the  emission  of  the 
basalt.  However,  it  is  not  improbable  that  these 
buttresses  mark  old  points  of  eruption,  and  perhaps 
owe  their  northern  cliffs  to  the  wave-adlion  of  a 
rather  shallow  lake,  of  which  Lake  Nazik  seems  to 
be  a  last  remnant.  We  traversed  the  sandy  plain 
between  these  long  buttresses  of  Nimrud  and  Lake 
Nazik  (6380  feet) ;  it  is  certainly  very  level  (6460  feet) 
and  has  all  the  appearance  of  a  lake-bed.  It  is 
separated  from  the  basin  of  Lake  Van  by  a  low 
ridge,  now  notched  by  a  dry  gully,  through  which 
the  waters  (if  the  lake  existed)  would  seem  to  have 
been  drained  away.  The  sides  of  this  ravine  showed 
nothing  but  brown,  volcanic  sands  in  practically 
horizontal  beds. 

This  northern  plain  is  separated  from  the  plain 
of  Mush  (on  the  S.W.)  by  a  lofty,  terraced  spur  of 
lava-flows,  extending  from  the  broad,  N.W.,  basalt 
terrace  of  Nimrud  for  about  7  or  8  miles.  This  wide 
spur  eventually  meets  the  lava-flows  from  the  Kekeli 
cone,  south  of  Bilejan,  and  breaks  off  to  the  plain 
of  Mush  by  a  lateral  line  of  cliffs  (about  1000 
feet  high),  which  we  ascended  by  a  zigzag  path,  on 
our  journey  from  Norshen  to  Nazik.  The  view  on 
reaching  the  summit  of  the  cliff  was  one  of  broad, 


148  Geology  of  Armenia 

swelling  downs  gradually  sloping  to  the  N.W.  The 
clifiE  is  not  a  continuous  precipice,  but  is  interrupted 
by  successive  terraces  in  steps,  just  as  at  Norshen 
on  the  opposite  (south)  side  of  the  plain.  It  is 
probably  of  the  nature  of  a  fault-scarp — the  present 
plain  of  Mush  being  an  area  of  depression  (p.  12), 
in  which  slipping-down  has  continued  subsequently 
to  the  outflow  of  the  basalts  of  Nimrud.  The  violent 
earthquakes,  which  still  occur  from  time  to  time  in 
this  area,  seem  to  indicate  that  the  subsidence  has  not 
yet  ceased. 

After  crossing  the  plain  of  Mush  from  Norshen, 
we  rose  at  first  quite  gradually  from  the  plain — the 
continuous  slope  of  the  basalt  sheet  becoming  in- 
terrupted by  occasional,  low  cliffs  facing  south.  The 
steep  ascent  began  at  Karnirash.  Here  the  track 
made  a  sharp  zigzag  up  a  cliff  of  grey  augite-rhyolite 
(with  small  quartz  crystals),  showing  a  selvage  of 
obsidian ;  and  after  reaching  a  terrace  of  brown, 
cindery  basalt  we  rode  obliquely  up  the  face  of  a 
similar  cliff  of  rhyolite  and  obsidian,  capped  by 
basalt,  which  forms  the  surface  of  the  broad,  N.W. 
spur  of  Nimrud.  I  had  previously  noticed  numerous 
fragments  of  obsidian  scattered  over  the  surface  of  the 
plain  between  Gol  Bashi  and  Karnirash. 

Similar  terraces  and  low  cliffs  (probably  fault- 
scarps)  occur  also  at  the  head  of  the  Mush  plain; 
they  are  particularly  conspicuous  along  the  western 
bases  of  Nimrud  and  Kerkur.  Here  again  they 
interrupt  the  continuity  of  the  sheets  of  basalt,  which 
flowed  down  from  Nimrud  to  flood  the  plain  of  Mush, 
and  at  the  present  day  constitutes  the  surface  rock. 
This  cindery  basalt  weathers  readily  to  its  usual, 
rusty  brown  colour,  and  has  been  worn  into  deep 
cart-ruts  along  the  main  tracks.  The  polygonal  endis 
of    the    broad    columns    (10-15    ^^^    ^^    diameter) 


The  Nimrwi  Volcano  149 

heighten  the  illusion  of  ancient,  paved  roadSi  running 
across  the  plain.  The  ravine  at  Norshen  is  about 
30  feet  deep,  and  reveals  no  other  rock  but  the 
columnar  basalt.  Patches  of  a  black,  scoriaceous, 
glassy  selvage  (No.  119),*  up  to  2  feet  in  thickness, 
sometimes  conceal  the  oasalt,  e.g.,  near  Morkh,  and 
I  noticed  the  same  black  selvage  in  patches,  6  inches 
to  I  foot  tbftck,  whilst  crossing  the  plain  between 
Norshen  and  Kamirash.  It  presents  the  appearance 
of  having  spurted  up  through  the  thin,  newly  formed 
crust  of  the  basalt,  previously  to  the  final  con- 
solidation of  the  lava-stream. 

The  furthest  westerly  point  which  I  reached  in 
the  plain  of  Mush  was  the  pool  of  Norshen  (4634  feet), 
near  the  ruined  kumbet  of  Karanlai  Agha,  about  i^ 
mile  N.W.  of  the  village.  It  is  a  circular  basin,  35 
yards  in  diameter,  and  6-8  feet  deep.  It  has  been 
excavated  in  the  brown,  cindery  basalt  of  the  plain, 
and  has  been  enclosed  on  the  north  side  by  2  rows 
of  well- trimmed  blocks  of  basalt.    The  water  wells 

*  The  rock  is  fine-grained,  and  full  of  steam-holes  (mostly  about 
6  mm.  in  diameter)  much  dongated  in  a  direction  nearly  vertical 
to  the  surface.  A  few  striated  felspars  (up  to  6  mm.),  and  some 
specks  of  olivine  can  be  distinguished,  under  the  microscope,  the 
PULGiocLABS  phenocrysts  occur  in  two  generations : — (i)  a  few,  large, 
tabular  crystals  of  labradorite  (about  AbxAns),  usually  in  groups, 
and  sometimes  penetrated  by  groundmass;  (ii)  more  numerous 
crystals  ot  andbsine  (about  Ab4An3)9  of  prbmatic  habit ;  occasionally 
twinned  on  the  Baveno  system  in  addition  to  the  usual  albite  and 
Carlsbad  plans ;  frequently  moulded  on  olivine.  Olfvine  is  the  only 
other  phenocryst,  occurring  in  fragments  and  crystals,  down  to  quite 
smadl  dimensions ;  fresh  and  water-dear  except  for  occasional  stains  of 
iron  oxide ;  sometimes  invaded  by  the  brown  glass  of  the  groundmass. 
A  few  large  crystals  of  magnetite  are  associated  with  the  larger 
olivines.  The  0(iouNDif  ass  consists  of  a  very  dark  brown,  dusty  glass 
crowded  with  skeletal  needles  of  magnetite  (sometimes  with  delicate 
c*'oss-bars),  together  with  numerous  felspar  laths  (andesine  about 
AbsAnj),  showing  fluidal  arrangement,  and  some  interstitial  augite 
gruttles. 


150  Geology  of 

up  freely  and  is  agreeable  to  the  taste ;  it  can  certain- 
ly have  no  connexion  either  with  the  mawkish  water 
of  the  Nimrud  lake  (as  averred  by  the  villagers), 
or  with  the  bitter  water  of  Lake  Van  (as  alleged 
by  Ains worth  ♦).  At  7  p.  i».,  on  July  29th,  the  tem- 
perature of  the. water  was  only  5I^  while  that  of 
the  air  was  80®  F,  The  only  water-plant  present  was 
a  species  of  Elodea ;  I  ^.Iso  noticed  a  few  freshwater 
shrimps  (Gammarus). 

Two  conspicuous  hills  (S  &  T),  each  with  a 
S.W-N.E.  axis,  rise  out  of  the  plain  north  of  Norshen. 
I  was  able  to  cursorily  examine  the  lower  slopes, 
much  seamed  with  dry  watercourses,  of  the  first  hill 
(S),  and  I  found  it  to  consist  of  a  dark  purple,  dndery, 
columnar  basalt,  similar  to  that  of  the  plain.  The 
hill  is  terraced  on  the  south  si^e,  but  slopes  gently  to 
the  north.  It  is  traversed  bv  a  broad  depression 
along  its  long  axis,  similar  to  the  f;jas  of  Iceland,  and 
especially  to  the  Almanna-gja  m  the  Thingyalla, 
figured  by  Dr.  Tempest  Anderson*!  The  more  distant 
hill  (T)  shows,  however,  a  distinct,  though  shallow 
crater,  breaclied  on  the  N.E. 

IV.    Chronology  of  Nimrud. 

No  sedimentary  strata  occur  nearer  to  Nimrud  thab 
tl^e  metamorphic  schists  and  marbles  of  Tidvan  (p. 
100),  and  the  Lower  Miocene  conglomerates  and  sand- 
stones of  Akhlat  (pp.  98,  14s).  The  latter  have 
been  folded  imd  much  denuded  previously  to  the 
later,  basaltic  eruptions  ci  Nimrud.  Probably  the 
old,  trachytic  mass  of  Kerkur  (the  predecessor  of 
Nimrud,  p«xi5)  was  erupted  in  the  Lower  Pliooeney  as 
a  result  of  the  fradures  to  which,  in  my  opinion^ 

•  /onm,  Ray.  Gtagr.  Soc.^  rt95,  *7J* 
t  Vtflcamc  Studies^  Iviii.  LmMbm,  19^. 


The  Nimrud  Volcano  151 

the  Armenian  area  was  then  subjeded  (p.  10).  I  have 
already  (Z.c.)  remarked  that  Nimrud  lies  on  the  great 
line  of  fradnire,  which  cuts  across  the  Armenian  tolds 
at  the  apex  of  their  bending  round  from  the  Persian 
ftil.W.)  to  the  Antitauric  (W,S.W.)  direcftion  (p.5). 
Furthermore,  Nimrud  lies  exaAly  at  the  point  of 
intersedlion  of  this  fradhire  with  the  great  fradtuxe 
(running  S.E.-N.WA  which  de-limits  I^e  Van  and 
the  Mush  plain  on  the  south  (pp.  lo,  1 10). 

Subsequently  the  vent  of  the  Nimrud  Dagh  arose 
Just  north  of  Kerkur.  The  succession  of  events  in 
its  life-history,  as  shown  by  my  observations,  may 
be  summarized  as  follows : — 

1.  Building  up  of  a  vast  cone,  at  least  12,000  feet 
high,  composed  of  augite-rh>[olite  (pp.  i^i,  139,  140), 
often  with  a  selvage  of  pumice  or  obsidian  (pp.  130, 
132,  139).  I  met  with  distal  portions  of  these  older 
acid  lavas  west  of  Kerkur  (p.  117),  on  the  shore  of 
Lake  Van,  near  Tadvan  (p.  143),  and  at  Kamirash  (p. 
148). 

2.  The  increasing  pressure  of  the  lofty  cone 
on  the  central  column  of  lava  became  relieved  by 
the  formation  of  parasitic  cones  on  its  flanks  (p.  131}, 
or  by  extrusions  of  augite-trachyte  along  nssures, 
at  a  greater  distance  from  the  central  mass,  e.^.,  at 
Kizvag  (pp.  141 -143),  Zighag  (p.  144),  Karmuch  (p. 
145),  and  south  of  Kerkur  (pp.  115,  116). 

3.  A  long  period  of  inactivity  may  be  presumed, 
so  as  to  account  for  the  complete  change  in  the 
character  of  the  lava. 

4.  A  series  of  violent  explosions,  destroying  the 
summit  of  the  cone,  and  distributing  the  material 
over  the  surrounding  coimtry,  especially  to  the  east 
and  north-east  (pp.  140,  12^3*145). 

5.  Extensive  outpounngs  of  andesitic  olivine- 
basalt,  sometimes  alternating  with  the  explosions  (pp. 


\ 


154 


k 


Geology  of  Armtnia  1 


1 

•        ■ 


inscription^  ^^molto  altevato  dal  tempos  recording 
the  eruption  of  Nimrud  in  1441,  occurred  "siiUa  chiesa 
del  villagio  di  Gamurtci  alV  0.  di  Kelatk  [Akhlat] 
e  pochissime  distante"  The  quotation  runs  as  fol- 
lows : — 

"In  Z441  a  great  sign  took  place,  for  the  mountain  called 
Nemrud  [su]^  which  lies  between  Kelath  and  Bitlis,  suddenly  began  to 
rumble  hke  heavy  thunder.  This  set  the  whole  land  into  terror 
and  constemationi  for  one  saw  that  the  mountain  was  rent  asunder  to 
the  breadth  of  a  city;  and  from  out  of  this  deft  flames  arose, 
shrouded  in  dense,  i^m^ling  smoke,  of  so  evil  a  stench  that  men  fell  ill 
by  reason  of  the  deadly  smell  Red-hot  stones  glowed  in  the  terrible 
flames,  and  bouldeis  of  enormous  sise  were  hurled  aloft  with  peals 
of  thunder.    Even  in  odier  provinces  men  saw  all  this  distinctly." 

12.  Earthquake-cracks  of  the  N.W.  rim  of  the  \ 
crater  (p.  136) ;  continued  slipping-down  of  the  plain 
of  Mush,  with  the  formation  of  fault-scarps  at 
Norshen  and  Kamirash  (p.  148),  and  at  the  head 
of  the  plain,  along  the  western  base  of  Kerkur  and 
Nimrud.  The  formation  of  these  fault-scarps  may, 
however,  have  commenced  at  any  period  subsequent 
to  the  emission  of  the  Nimrud  basalts  (p.  152). 

13.  Violent  earthquake  on  May  30th,  188 1,  about 
6.  30  a.in.^  preceded  by  rumbling  from  Nimrud,  and 
resulting  in  the  destrudtion  by  mud  of  the  village 
Teghurt  ♦  (p.  146),  at  the  N.E.  base  of  the  crater- wall 
of  Nimrua  The  mud  was  probably  due  to  the 
disturbance  of  underground  water,  rising  in  an 
earthquake-fissure.  Major  Clayton,!  who  brought 
relief  from  Van  to  the  survivors,  states  that  a  loss 
of  93  lives  in  Teghurt  (Segurt)  resulted  from  this 
earthquake,  which  caused  damage  in  other  villages  in 
the  line  connedtin^  Sipan  and  Nimrud,  but  nowhere 
else.    The  long,  N.-S.  cracks  (up  to  3  feet  across), 

*  Abich,  4!^.  «Sr.,  ii«  445. 

t  Alfitujmum^^  Aug.  1887,  293. 


KtzvAd  nis&r 


^'*'*»nn»^ 


S. 


M 


9f*9r'*\t>AH 


<    k     NiyH 


i 


I 


The  Nimrud  Volcano  f  55 

which  I  noticed  in  the  basalt-plain  at  the  eastern  base 
d  the  Nimrud  crater-wall,  south  of  Teghurt,  were 
probably  caused  by  the  same  earthquake. 

14.  Rise  in  level  of  the  crater-lake  within  the 
last  ten  years  (pp.  iy8,  126). 


CHAPTER  X 


AKHXAT  TO  SIPAN 


After  our  wanderings  round  the  Nimrud  volcano, 
over  sandy  plains  and  arid  lava-flows,  we  encamped 
again  in  our  orchard  at  Akhlat,  on  the  coast  of  Lake 
Van.  We  resumed  our  journey  the  next  day  (August 
2nd),  riding  along  the  coast  to  Adeljivas,  where  we 
proposed  to  make  arrangements  for  ascending  Sipan. 

'ftie  dark  red  sandstones  and  conglomerates  (Lower 
Miocene,  see  p.  98)  continue  for  some  distance  along 
the  shore,  dipping  about  20^  S.S.E.  They  are  capped 
at  about  50  yards  from  the  water's  ed^e  by  the 
pumice-sands  of  Nimrud,  which  extend  mland  for 
several  miles,  up  to  and  beyond  the  troglodyte  village 
of  Madavantz.  This  settlement  lies  on  the  Akhlat 
river,  a  few  miles  above  Takht-i-Suliman ;  the 
ancient  church  and  dwellings  have  been  excavated 
out  of  the  cindery  basalt,  which  flowed  down  the 
valley,  and  subsequently  became  covered  by  about 
20  feet  of  volcanic  ash  (cp.  p.  145).  At  Madavantz, 
however,  the  river  has  not  yet  beea  able  to  cut  its 
way  down  to  its  former  level. 

On  proceeding  along  the  low  coast  from  Akhlat, 
I  noticed  that  near  Tun  us  the  conglomerates  recede 
inland  (their  dip  having  changed  to  20°  N.N.W.),  and 
the  coast-line  oecomes  more  rocky,  owing  to  the 


AkUat  to  Sipan  157 


presence  of  a  c^uartz-mica-augite-diorite  (No.  140),*  in 
which  the  augite  has  for  the  most  part  become  altered 
into  hornblende. 

Along  this  part  of  the  coast  a  coarse,  grey  con- 
glomerate overlies  the  diorite,  from  which  it  has  been 
mainly  derived.  It  differs  essentially  from  the  red 
conglomerates  of  Akhlat,  and  is,  in  fadl,  part  of  the 
old  beach,  formed  when  the  waters  of  Lake  Van  stood 
at  about  100  feet  above  the  present  level  (p.  102).  Its 
inclination  is  that  of  the  shore,  vt>.,  12°  S.E.,  and 
it  does  not  exceed  10  feet  in  thickness. 

The  diorite  soon  gave  place  to  a  pjnroxene-por- 

*  The  rock  is  of  medium  spin ;  lustrous,  greenish  black  horn- 
blendes (up  to  5  mm.),  and  minute  scales  of  biotite  lie  in  a  matrix 
of  white  felspar  (up  to  3  mm.)  and  quartz.  Under  the  microscope, 
PLAGiocLASB  (an  add  labradorits  about  AbiAni)  is  the  most 
abundant  constituent,  with  a  strong  tendency  to  idiomorphism,  the 
prismatic  habit  predominating ;  the  periphery  is  often  dear  and  zoned, 
while  the  nudeus  shows  no  zoning,  and  contains  a  network  of  cracks, 
often  dusty  with  caldte  and  white  mica.  Orthoolase  is  present 
in  allotriomorphic  grains,  turbid  with  brownish  decomposition- 
products;  quite  subordinate,  and  less  in  amount  than  the  quartz, 
which  comes  next  in  abundance  to  the  plagiodase.  A  pale  green,  and 
malacolitic  auoite  is  present,  but  very  little  has  remained  unaltered ; 
a  rather  large  crystal  (approaching  idiomorphism)  is  surrounded  by 
a  zone  of  green  hornblende  and  magnetite  granules,  with  some  biotite 
iSakes  bordering  this  wreath;  the  deavages  of  the  augite  are 
continued  into  tiie  hornblende ;  calcite  has  in  places  formed  along  the 
d^vages.  The  augite  has  for  the  most  part  been  altered  into  reedy, 
URAunc  HORNBLENDE,  in  allotriomorphlc  grains,  but  a  core  of  augite 
is  frequently  present,  even  in  the  compact  hornblende  with  its 
characteristic  deavages  and  strong  pleodir9ism :  a — ^pale  straw, 
^ — green  with  a  brownish  tinge,  r--dark  sap  green.  In  some  cases 
the  hornblende  itself  has  been  altered  to  TiRiDms,  a  paler  green,  only 
slightly  pleochroic  micro-aggregate  of  brightly  polarizing  partides. 
BioTiTB  is  present  in  numerous,  small  flakes,  usually  assodated  with 
hornblende  and  magnetite ;  they  have  been  frequently  bent,  and  oc- 
casionally assume  a  greemish  colour.  Magnetite  is  frequently 
associated  with  the  ferromagnesian  minerals.  Apatite  is  accessory. 
Quartz  is  not  only  primary  and  interstitial,  but  is  also  secondary,  and 
then  contains  small  hornblende  needles,  having  formed  along  a  cradL 


158  Geology  of  Armenia 

phyrite  fNo.  141),*  rather  andesitic  in  character,  rising 
up  in  cliffs,  which  at  the  same  time  receded  inland 
as  we  approached  Sor.  This  village  is  situated 
among  fine  old  walnut-trees,  on  the  alluvial  fiats 
of  a  small  river.  Here  again,  a  spit  of  beach  has 
formed  (p.  105),  with  shallow  lagoons,  in  which  tall 
reeds  grow  luxuriantly. 

After  fording  the  shallow  stream  we  left  the  sandy 
beach  to  rise  over  another  zone  of  pyroxene-porphyrite 
(No.  142),!  much  decomposed,  with  occasional  reddish 

*  The  rock  contains  abundant,  white  felspars  (up  to  2  mm.),  and  a 
few,  greenish  blade  augites  (up  to  2  mm.)  in  a  dark  green,  compact 
matrix.  It  has  been  greaUy  cracked,  and  the  dracks  have  been 
filled  with  chalcedony,  a  porcellanous  white,  with  concentric  markings 
along  the  margins,  the  centre  of  the  vein  showing  a  slight  pinkisli 
tinge.  Under  the  microscope,  the  plaoioclasb  (akde^ine  about 
Ab4An3)  is  present  in  abundut  idiomorphs  of  prismatic  habit,  turbid 
with  paie  brown,  dusty  decomposition-products,  and  often  with 
chloritic  particles ;  traversed  by  a  network  of  cracks ;  a  peripheral 
zone  of  dust  is  frequent;  often  zoned;  sometimes  encloses  augite 
granules.  Augite  occurs  in  pale  green  idiomorphs,  quite  fresh, 
sometimes  in  groups;  frequently  twinned,  sometimes  repeatedly; 
encloses  magnetite.  Bronzitb  is  sparingly  present  in  pale  green 
idiomori)hs,  feebly  pleochroic  (to  psde  pink);  mostly  altered  into  a 
vesicular  network  of  chlorite  (in  a  fine  microaggr^;ate),  epidote, 
and  chaicedony.  Maonetits  is  accessory.  Small,  irregular  grains  of 
QUARTZ  (some  showing  a  narrow  margin  of  chalcedony)  not  seldom 
occur,  essentially  as  part  of  the  gboukdmass,  which  is  an  extremely 
fine-grained,  pale  brownish,  obscurely  micropcedlitic  felt  of  closely 
ranged  felspar  (oligoclase)  microlites,  showing  fluidal  arrangement, 
with  abundant  magnetite  granules,  a  few  tiny  augite  granules,  and 
much  disseminated  chlorite.  The  chalcbdont  veins  consist  of 
mutually  interfering  spherulites,  showing  centripetal  growth  from  the 
margins  of  the  veins ;  chloritic  mater  occurs  in  some  of  the  veins,  and 
also  in  the  interior  of  a  cavity  lined  by  chalcedony. 

t  The  rock  is  compact,  with  abundant,  striated  felspars  (up  to 
3  mm.),  often  with  a  green,  chloritic  core,  and  lustrous,  greenish 
black  augites  (up  to  5  mm.)  in  a  dark  green  matrix,  mudi  stained 
with  iron  oxide.  Under  the  microscope,  the  plaoioclase  (andesimx 
about  Ab4An3)  i^  present  in  idiomorphs  mostly  of  prismatic  habit ; 
sometimes  soned,  with  a  narrow  mar^n  of  dear,  more  add  felspar; 


Akhlat  to  Sipan  159 

bands,  and  with  many  white  veins  pf  chalcedony. 
The  rock  is  not  quite  so  fine-grained  as  the  specimens 
(Nos.  141  &  143)  on  the  west  and  east  respecftiyely, 
which  were  in  nearer  contaA  with  the  intruded 
sedimentary  strata.  All  these  porphyrites  probably 
proceeded  from  the  same  magma.  The  porphyrite 
is  concealed  here  and  there  by  a  thin  layer  of  the 
old  beach-conglomerate ;  and  in  one  place  near  the 
shore  this  is  in  tum  hidden  by  a  patch  of  horizontal 
travertine.     Here  the  pyroxene-porphyrite  (No.  143)  * 

inclusions  of  dust  occur  centrally  or  along  a  narrow,  peripheral  zone  ; 
frequently  with  a  network  of  cracks,  with  particles  of  chlorite  or 
yellow  stains  of  iron  oxide  along  the  cracks ;  often  with  specks  of 
white  mica ;  sometimes  encloses  grains  of  augite.  AugiTk  is  present 
in  numerous,  very  pale  green,  almost  colourless  idiomorphs,  quite 
fresh,  often  in  aggregates  :  sometimes  twinned,  occasionally  repeated- 
ly; encloses  apatite  And  magnetite.  Pseudomorphs  after  khombic 
PVBOXENE  are  less  frequent ;  the  characteristic  outlines  and  cleavages 
are  still  distinct,  but  none  of  the  original  mineral  has  escaped  aK 
teration ;  they  consist  now  of  opal  and  chlorite ;  the  latter  is  a  bright 
green,  forming  along  the^  margins  and  cleavages,  sometimes  with 
separation  of  magnetite ;  it  is  strongly  pleochroic  {£ — ^peacock  green, 
O — greenish  yellow);  isotropic  01^  or  else  chalcedony  (in  radial 
aggregates)  sometimes  fills  the  centre  of  the  crystals.  Magnetitk, 
APATITE,  and  ZIRCON  are  accessory.  The  gboundmass  is  an  obscurely 
micropoecilicic  felt  of  minute  felspar  laths  (oligoclase-andesine  about 
Ab5An3),  augite  granules,  and  abundant  magnetite  granules.  Small, 
irregular  cavities  occur,  now  lined  by  chlorite,  and  filled  with  a  fine 
aggregate  of  chalcedony  and  Chlorite,  or  else  with  granules  of  quartz. 
The  rock  has  been  much  saturated  by  chlorite  and  chalcedony,  and 
stained  by  iron  oxide. 

*  The  rock  is  compact  with  irregtdar  fracture,  containing  numerous 
striated  felspars  (up  to  3  mm.),  lustrous,  greenish  black  p3rroxenes  (up 
to  5  mm.),  and  magnet  te  granules  in  a  dark  greenish  bku^k  matrix, 
stiined  brown  in  places  by  iron  oxide.  It  is  slightly  magnetic  (2^). 
Under  the  microscope,  tie  vr«AOiocLASE  (labradouite  about  Ab5An6) 
IB  present  in  abundant  idiomorphs,  mostly  of  prismatic  habit,  forming 
al>out  two^thirds  of  the  rock;  frequently  shows  zoning,  which  is 
marked  out  by  concentric  lines  of  small  inclusions  of  pale  broi^ 
dusty  glass  and  augite  granules;  sometimes  fresh,  but  often  with 
%.  network  of  '.racks,  stained  by  iron  oxide.    Augite  is  present  in 


i6o  Geology  of  Armenia 

was  rather  finer-grained  and  occurred  not  far  from 
the  jundlion  with  a  grey  limestone  (No.  143^),  which 
it  has  marmorized  and  rendered  sphntery.  The  lens 
reveals  a  fragment  of  a  Lower  Miocene  coral,  vix.i 

Cladocora  articulata^  Abich. 
This  fossil,  which  Abich*  discovered  in  the  Jafarlu 
Dagh  (Daralagoz  region,  south  of  Lake  Gokcha), 
occurred  in  unaltered  limestone  near  the  Jil  Gdl, 
about  8  miles  N.N.W.  of  this  point  of  our  route  (wcfe 
infYO). 

Here  the  limestone  dips  about  60®  N.N.W.,  and 
at  once  caused  a  striking  change  in  the  character 
of  the  landscape :  from  the  rugged  cliffs  and  barren^ 
sombre  hills  of  the  intrusive  porphyrites  we  passed 
to  a  pastoral  country  of  open,  grassy  downs.  Patches 
of  the  old  beach-conglomerate  still  occurred  near 
the  shore,  dipping  12*^  S.E.,  i.e.,  towards  the  lake. 

We  again  descended  to  the  water's  level  near 
Hassiindluk,  where  two  broad  valleys  unite,  just 
before  reaching  the  lake.    On  looking  to  the  N.W*, 

numerous,  pale  green,  rather  large  idiomorphs,  often  in  groups,  which 
sometimes  surround  bronzite  crystels;  quite  fresh;  often  twinned, 
sometimes  repeatedly ;  encloses  apatite  and  magnetite.  Bronzite 
is  less  abundant  than  the  augite,  and  in  rather  smaller  idiomorphs ; 
quite  frtsh ;  slightly  pleochroic  (pale  green  to  pale  pink) ;  endoses 
magnetite,  and  in  one  case  a  kind  of  wreath  of  bronzite  surrounds  an 
irregular  grain  of  magnetite.  Maqnetitb  is  fairly  abundant,  diiefly 
associated  with  the  pyroxenes.  The  oroukdmass  is  a  crypto- 
crystalline,  dusky  felt  of  idensdy  crowded  felspar  crystallites,  abun- 
dant magnetite  granules,  and  some  tiny  augite  granules,  with  sparing, 
colourless,  interstitial  glass.  It  b  much  stained  by  iron  oxide, 
and  traversed  by  numerous,  small  cracks,  now  fiUed  with  dedr 
CHALCEooNT,  showiug  bcautiful  spherulitic  structure ;  slight  shifting 
has  taken  place  along  the  cracks.  Green  globules  of  chlorite  occur 
along  the  margins  of  &e  chalcedony  veins. 

■ 

*  Das  Steinsahf  6*^.,  14.  Mhn.  And.  ScUntes^  s^.  6,  vii.  St' 
Petersburg,  ^[859.  See  also  Chapter  XXIII  on  the  Miocene  of 
Armenia. 


Ahhlat  to  Sipan  i6i 

up  this  opening  in  the  hills,  I  could  see  a  paxt  of 
tiie  lofty  escarpment — the  southern  limit  of  the  broad 
zone  of  Miocene  limestone,  which  we  had  already 
crossed  between  Melaskert  and  Akhlat  (p.  97).  In  the 
lower  part  of  the  valley  the  imderlying  red  sandstones 
could  be  distinguished,  dipping  about  10®  N.N.W- 

After  crossing  the  wide,  alluvial  embouchure  we  rose 
up  again  over  fissile,  dark  grey  marble,  much  tra^ 
versed  by  white  veins  of  calcite,  and  overlaid  here  and 
there  along  the  shore  by  patches  of  the  old  beach. 
The  marble  forms  cliffs,  increasing  in  height,  and  has 
a  general  dip  of  45^-60^  N.N.W.;  the  strata  are  often 
locally  contorted,  perhaps  indicating  thereby  a  still 
concealed  boss  of  intrusive  rock.  Reddish  brown 
patches  on  the  bare  hillside  are,  moreover,  suggestive 
of  the  proximity  of  igneous  rock. 

As  we  proceeded  along  the  coast  the  marble 
changed  gradually  into  its  normal  condition  of  a  hard 
cream-coloured  limestone.  The  change  of  character 
was  graphically  indicated  by  the  colour  of  the  lake 
passing  from  a  clear  azure  to  a  milky,  turbid  con- 
dition, in  a  zone  extending  about  a  quarter  of  a  mile 
from  the  shore. 

We  had  now  approached  the  bold  headland,  so 
conspicuous  from  Akhlat ;  the  narrow  track  left  the 
clQse  proximity  of  the  shore,  and  rose  above  the  lower 
clilF,  rotmd  wnich  I  had  traced  remnants  of  the  old 
high-level  beach  of  Lake  Van.  The  limestone  pre- 
sents a  rather  travertinous  appearance  at  first,  dipping 
so°  S.S.E.,  but  but  I  could  nnd  no  fossils  here  except  a 
fragment  of  Alectryonia  (Ostrea)  Virletij  Desh.  The 
loose  texture  of  the  rock  in  places  induced  Lof tus  *  to 
make  the  following  statement : — 

'<  The  range  to  the  north  of  Addjivas,  and  the  rock  on  which  the 
*  Quart.  Jimm.  GeoL  Soc,  xi.  322. 


1 62 


Geology  of  ArtMnia 


castle  stands  is  nothing  more  than  a  tnvertinous  deposit  on  a  hug^ 
scale ;  the  stone  sometimes  assumes  the  character  of  a  true,  compaot* 
yellow  limestone  (containing  indications  of  corals  and  marine  shells).^ 

About  50  yards  further  on,  however,  corals  wi 
present  in  great  numbers;  indeed  the  limestone] 
possesses  the  charader  of  a  coral  reef.  Here,  at  the 
entrance  to  a  small  cave  in  the  cliff,  I  obtained  the 
following  specimens,  which  indicate  the  Lower  Mio- 
cene (Upper  Helvetian)  age  of  the  limestone,*  Mr.  R. 
Bullen  Newton,  F.G.S.,  has  kindly  determined  the 
fossils  as  follows : — 

SoUnastroea  Turonensis^  Mich. 

Asiraa  T^mmnuts^  MicheUn,  Iconpgr.  Zoophjt,  1840- 
1847,  pL  75,  fig.  I,  a,  p.  3x2.  The  original  was 
described  hj  Michelin  from  Miocene  beds  (Helvetian 
or  Tortonian)  of  Manthelan,  France. 

SoUmasirma  Titronemis^  Gregoiy,  Journ.  Linn.  Soc., 
yd.  S7,  p.  4s8,  Persian  specimens  from  Lake  Urmi 

Orbicella  Defrancei^  Edwd.  &  Haime. 

Asiraa  argus^  Mich. 
Orbiuiia  Deframd^  Gregory. 

Styhphora  subreticulatay  Reuss. 

Natural  cast  of  calides,  with  their  six  septa. 

Tkamnarcea  polymorpha^  Abich. 

Forites  polymofpha^  Abich,  Das  Steinsalz,  &c»  pL  iz, 

figs,  la-e^  p.  loo. 
ThamnaroM  polymorpha^  Grq^ory,  /.r.,  pL  s8|  fig.  3» 

p.  4S9. 

Heliastrosa^  sp.  indet. 

The  sea,  in  which  these  corals  flourished,  was 
evidently  continuous  with  that  of  the  Urmi  region  on 
the  east,  and  that  of  Ashkala,  in  the  Upper  Frat 
basin  on  the  west,  where  these  fossils  also  occur.* 

After  passing  the  promontory  with  its  two  small 
islands,  our  path  rose  higher  and  higher,  up  the  face 
of  the  clifis  to  about  200  feet  above  the  lake.    The 

*  See  Chapter  XXIII  on  the  Miooene  of  Annenia. 


181. 


.w^??.-»t:x-:  :»:*s!" 


'■^Ittoiii ,  ' 


-J. 


iM^'^"'^    ''"'^    '^  ''"!» 


Akhlat  to  Sipan  163 

dip  still  remained  the  same,  viz.^  50°  S.S.E.,  for  our 
route  made  only  a  slight  angle  with  the  E.N.E.  strike. 

About  a  quarter  of  a  mile  further  on  we  descended 
again  to  the  old  shore-line,  at  about  100  feet  above 
the  lake ;  here  the  beach  consisted  entirely  of  lime- 
stone pebbles.  In  the  clijBF  I  noticed  a  fragment  of 
a  small  Peden  like  P.  urmtensis,  Abich,  and  numerous 
Bryozoa  {Retepora,  &c.),  similar  to  those  in  the 
Madrak  limestone  (p.  52),  but  the  corals  did  not 
not  re-appear,  for  they  belong  to  a  somewhat  higher 
horizon  \mde  infra).  At  the  second  conspicuous 
promontory  the  dip  was  still  50°  S.S.E.  The  lime- 
stone has  weathered  into  numerous  caves,  the  for- 
mation of  which  is  favoured  by  bold,  vertical  jointing, 
as  in  the  Miocene  limestone  of  Gozel  Baba  (p.  74), 
and  at  Al  Khes  (p.  82). 

Adeljivas  now  lay  open  to  our  view.  After  crossing 
a  broad,  well-wooded  combe,  we  descended  to  the 
shore.  An  old  Seljuk  burial-ground,  upon  which 
the  lake  is  encroaching,  lies  outside  the  walls  of 
the  old  city,  now  almost  deserted.  It  is  dominated 
on  the  west  by  a  bold  cliff  of  limestone,  crowned  by  a 
mediaeval  castle  with  extensive  ramparts.  A  crag 
to  the  north  of  the  ruined  city  showed  a  small, 
reversed  fault,  in  addition  to  the  still  constant  dip 
to  S.S.E.  The  present  town  of  Adeljivas  (15  miles 
from  Akhlat)  lies  on  the  banks  of  a  river,  about  half  a 
mile  north  of  the  old  city.  Lagoons  have  been 
formed  at  the  mouth  of  the  river,  as  elsewhere  on  the 
west  coast  (p.  105).  The  lake  at  this  spot  is  very 
shallow,  with  a  gently  shelving  shore  of  calcareous 
silt.     Here  I  picked  up  a  slightly  rolled  specimen  of 

Pe6ien  urmiensis^  Abich.^ 

*  Mr.  R.  Bullen  Newton,  F.G.S.,  in  kindly  confirming  this  deter- 
mination of  the  specimen,  states  that  it  is  "a  water-worn,  frag- 
mentary, lower  valve,  without  expansions,  of  probably  Feetm  wrmientU, 


164  Geology  of  Armenia 

A  fragment  of  the  same  species  occurred  in  a  large 
block  of  limestone  close  to  our  camp,  but  no  time  was 
available  to  examine  the  surroundings,  for  we  left 
Adeljivas  as  soon  as  sufficient  men  had  been  pro- 
duced by  the  kaimdkam  for  our  ascent  of  Sipan. 

After  leaving  the  last  orchards  behind  us,  at  a 
height  of  about  300  feet  above  the  lake,  we  traversed 
a  terrace  of  travertine,  and  then  ascended  into  the 
heart  of  the  high,  limestone  downs  (the  Suteh  Dagh), 
up  a  rocky  ravine  with  steep  screes.  Here  again 
the  upper  part  of  the  limestone  escarpment  (facing 
south)  weathers  to  a  pale  brown  colour,  with  the 
same  characteristic,  vertical  joints  as  previously 
noticed  (p.  163). 

Large,  angular  blocks  of  black,  scoriaceous  lava  lay 
scattered  here  and  there,  evidently  ejected  from  Sipan 
during  some  violent  eruption.  Here,  about  3^  miles 
inland,  I  noticed  an  instance  of  an  intrusion  of  a 
pyroxene-porphyrite  (No.    144)*  into  the  limestone, 

Abich,  (=ip.  henedic^  Abich,  Das  Steinsals,  &c.,  Mim.  Acad.  Imp. 
Sa.f  s^r.  6,  vii,  St.  Petersbuig,  1859,  pi.  z,  figs.  5  a,  ^,  &  8,  p.  61 ; 
also  =/>.  9ulmrmienai$9  see  Def^et  &  Roman,  Ifim.  Soc,  Ghl.  France^ 
1902,  vol.  10,  fasc  I,  p.  38). 

^  The  rock  is  compact  and  fine-grained,  with  irregular  fracture; 
it  contains  striated  felspars  (up  to  t  mm.),  and  lustrous,  greenish 
black  augites  (up  to  3  mm.)  in  a  greenish  black  matrix,  traversed 
by  white,  thread-like  veins  (chalcedony).  Under  the  microscope, 
the  PLAOiocLASE  (an  acid  labkadoritk  about  AbiAni)  is  present  in 
numerous  idiomorphs  of  prismatic  habit,  very  turbid  with  dusty  brown 
decomposition^products  and  chloritic  particles,  but  with  a  narrow 
margin  of  clear  felspar ;  sometimes  zoned ;  occasionally  encloses 
augite  granules.  Augite  is  frequent  in  pale  green,  almost  colourleK 
idiomorphs,  quite  fresh,  and  larger  than  the  fdlspars ;  often  in  large 
groups;  frequently  twinned,  sometimes  repeatedly.  Bkonzite  is 
sparingly  present,  in  smaller  idiomorphs  than  tiie  augite,  with  which  it 
is  associated ;  quite  fresh ;  only  slightly  pleochroic  (pale  green  to  pale 
pink).  The  oboundbiass  is  coarser  in  grain  than,  in  the  preceding 
poiphyrites;  it  is  a  felt  of  small  plagiodase  laths  (andesine  about 


Akhlat  to  Sipan  765 

altering  it  into  crystalline  marble  (weathering  a 
rusty  brown)  for  about  20  feet  from  the  jundlion. 

We  continued  to  ascend  to  the  N.W.  by  a  narrow, 
winding  track,  high  above  the  mountain-torrent. 
On  reaching  the  summit  of  the  col,  the  volcanic  plain 
of  the  Jil  Gcil  opened  out  suddenly  to  our  view.  A 
gentle  descent  of  40  minutes  brought  us  to  Uran  Gazi 
on  the  north  side  of  the  plain  (7736  feet;  2  hours 
from  Adeljivas),  This  plain  is  bounded  on  the  south 
by  the  high  downs  of  the  Suteh  Dagh,  composed 
of  Miocene  limestone,  dipping  20°  N.N.W.  In  the 
lower  slopes  of  the  limestone,  S.E.  of  the  lake,  I  found 
the  following  Miocene  corals  *: — 

Cladocora  articulata^  Abich  (p.  160). 
Trochocyathus  sp.  (height  =  110  mm.,  diameter 
of  cup = 58  X  45  mm.) 

The  former  forms  masses  2-4  feet  thick,  and  some 
foraminifera  {Amphistegina)  can  be  distinguished  in 
the  limestone  matrix  between  the  corallites. 

The  result  of  my  observations  on  the  Miocene  of 
Lake  Van  may  be  summarized  in  the  following 
succession: — 

i«  LowsR  ToRTONiAN.  White  limestone,  rather  travertinous 
in  appearance,  with  Atechyoma  {Ostna)  Virleti^  Desh,  (p. 
i6x). 

AbsAnj)  with  abundant  granules  of  angite  and  magnetite,  and  a  little 
interstitial,  chloritic  matter.  A  few,  small  cavities  have  been  filled 
with  CHLORITE  and  chalceix)NY,  the  former  acting  as  the  lining. 
Chalcedony  also  fills  a  crack,  along  which  a  little  shifting  has  occurred. 

*  Kindly  confirmed  by  Mr.  R.  BuUen  Newton,  F.G.S.,  who  states, 
with  reference  to  the  Trochocyathus^  ''compare  with  7*.  Narienm^ 
Duncan,  Pd.  Indka^  1880,  ser.  14,  pi.  ix,  figs.  1-13,  pp.  71-73-" 
Indeed  the  resemblance  of  the  Armenian  specimen  to  the  figures 
pf  this  variable  Indian  species  is  so  dose  that,  to  my  mind,  it  might 
be  safely  r^arded  as  identicaL 


{66  Geology  of  Armenia 

2.  Ufpib  Helvstian  : — 

(i).  Compact  Coral-limestone,  with  SoUnastrita  TYmm" 
ensiSf  Mich.,  OrbUdla  Defranm^  Edwd.  &  Haime, 
Styicphora  subrdiculaia^  Reuss,  l^amnaraa  poly- 
morpha^  Abich,  and  PteUn  urmiensis^  Abich,  (pp. 
162,  163). 

(ii).  Z/'/A^MaivMAw-limestone  with  Bxyozoa  {Ceriopora 
anomala^  Abichi  &c.)  and  Foraminifera  (pp.  97, 

163). 

3.  Lower  Hilvetian  : — 

(i).  Coral-limestone  with  Qadoeora  artiadata^  Abich, 
and  Trachocyaihus  cf.  Nariensis^  Duncan  (pp. 
160,  165). 

(ii).  Basal,  red  sandstones  and  conglomerates  of  Akh- 
lat  on  the  south  of  the  limestone-zone  (pp.  98, 
145,  156),  and  greenish  grey  sandstones  and 
conglomerates  of  Akhviran  (p.  95),  and  of  Shebu 
(Chapter  XI)  on  the  north. 

The  limestone  has  been  highly  marmorized  in  the 
proximity  of  the  numerous,  elliptical  bosses  of  black, 
scoriaceous  lava,  a  glassy  olivine-basalt  with  bronzite 
(No.  145),*  lying  between  the  limestone  and  the  east 
end  of  the  Jil  (5ol.  The  lava  of  the  Melaskert  plain 
(No.  64,  p.  92)  is  a  closely  allied  olivine-bronzite- 

*  The  rock  is  extremely  fine-grained,  with  rough,  irregular  fracture 
and  with  occasional,  small,  irregular  steam-cavities.  Tiny  felspars  (up 
to  2  mm.),  and  smaller  olivines  (less  than  i  mm.)  are  visible  with 
the  lens.  The  rock  is  slightly  magnetic  (4**).  Under  the  microscope, 
the  PLAGiocLASE  (andesine  about  Ab4An3)  is  present  in  small  pheno- 
crysts  of  prismatic  habit;  pericline  twinning  sometimes  occurs  in 
addition  to  the  usual  albite  and  Carlsbad  systems ;  occasionally  with 
a  broad,  peripheral  nct\^ork  of  black  or  dark  brown  glass.  Olivine  is 
present  in  abundant  small  idiomorphs  and  fragments,  often  in  groups ; 
quite  fresh.  Bronzite  idiomorphs  and  fragments  are  fairly  frequent, 
but  altogether  subordinate  to  the  olivine,  although  on  the  whole 
larger  in  size ;  only  very  slightly  pleochroic  (pale  green  to  pale  pink). 
AuGiTB  is  present  in  a  less  degree,  in  smaller,  crystalline  grains.  The 
OROUKOBCASS  consists  of  dark  brown  glass,  densely  crowded  with 
margarites  and  with  magnetite  in  skeletal  needles  and  dust,  together 
with  abundant,  small  felspar  laths  (oligodase-andesine  about  AbaAni) 
showing  fluidal  arrangement,  and  small  bronzites  and  augites. 


AhhUa  to  Sipan  167 

basalt,  but  with  a  well^developed  porphyritic  charac* 
ter.  These  lava-bosses  extend  also  along  the  south 
side  of  the  Jil  Gdl  plain,  bordering  the  limestone  hills 
as  on  the  east  side.  The  bosses  are  not  more  than 
30  feet  high,  and  all  of  them  seem  to  show  the 
same  W.S.W -E.N.E.  orientation ;  not  infrequently 
the  axis  is  emphasized  by  a  grassy  furrow,  1-2  feet 
deep,  evidently  a  gja^  similar  to  that  of  the  basalt 
hill  (S)  in  the  Mush  plain  (p.  150).  The  barren 
surface  on  either  side  of  the  gja  shows  the  hexagonal 
ends  of  radiating  columns  formed  on  cooling.  The 
lava  has  clearly  issued  from  fissures  in  the  plain. 

The  Jil  Gol  possesses  an  outflow  by  means  of  two 
small  whirlpools  at  its  eastern  end,  each  about  a  yard 
in  diameter.  The  lake  must  recently  have  stood  at 
a  slightly  higher  level,  for  a  funnel-shaped  depression 
close  at  hand  in  the  muddy  shore  marked  a  former 
similar  outlet.  Indeed,  Shakir  Effendi — the  headman 
of  Uran  Gazi  and  the  neighbouring  Circassian  vil- 
lages and  former  kaimakam  of  Adeljivas — told  us  that 
the  lake  drains  away  by  five  holes,  the  water  finding 
its  way  down  to  Adeljivas.  A  belt  of  bleached 
Conferva  felt — containing  many  shells  of  Limnoece — 
surrounds  tha  shallow  lake  at  a  distance  of  about 
50  yards  and  clearly  marks  a  higher  level  at  an 
earlier  period  of  the  year,  probably  at  the  time  of  the 
melting  of  the  snows.  Abundant  rushes  (whence  the 
name  of  the  lake)  flourished  in  the  black  mud. 

Shakir  EflFendi  was  also  our  authority  for  stating 
that  there  is  more  water  now  in  the  Jil  Gol  since 
the  earthquake  at  Adeljivas  in  1893 ;  many  houses 
were  destroyed  there,  while  only  harmless  shocks  were 
felt  at  Uran  Gazi.  It  was  an  earthquake  extending 
over  a  considerable  area,  for  it  was  felt  at  Bitlis 
and  lasted  on  and  off  for  40  days  at  Van. 

On  the  west  of  the  Jil  Gol  low  spurs  of  limestone 


i68  Geology  of  Armenia 

and  the  underlying  sandstone  rise  out  of  the  lava- 
plain  ;  and  on  tiie  north  a  low,  grassy  lava-flow  from 
Sipan  bounds  the  view. 

Uran  Gazi  (7736  feet)  stands  about  a  mile  north 
of  the  lake,  near  the  base  of  a  huge  lava  stream, 
which  has  flowed  down  from  Sipan  in  a  W.S.W. 
diredtion.  Here  the  adhial  ascent  commenced.  A 
rather  stiff  climb  up  the  rocky  flank  brought  us  to 
its  upper  surface,  smooth  and  grassy.  The  lava — a 
basaltic  hypersthene-augite-andesite  (No.  146)* — has 
a  twisted  and  ropy  appearance  on  the  outer  edge 
of  the  flow ;  the  large  boulders,  which  chiefly  encum- 
ber both  flanks,  show  a  conchoidal  fradlure,  with 
sometimes  a  dull  brown  lustrous  surface,  and  with 
a  general  absence  of  steam-holes. 

We  pitched  our  camp  halfway  up  this  lava-flow, 
at  a  height  of  10,300  feet ;  a  wide  panoramic  view 
lay  at  our  feet,  extending  over  100  miles  westward  in 
this  clear  atmosphere. 

^  The  rock  is  dark  grey  and  compact,  with  abundant  striated 
felspars  (up  to  2  mm.),  and  lustrous,  greenish  black  p3nroxenes  (up  to  x 
mm.)«  It  is  slightly  magnetic  (a"*),  and  effervescence  with  acid  takes 
place  along  irregular,  thread-like  veins.  Under  the  microscope,  the 
PLAOiocLASE  (ahdesikb  about  Ab4An3)  is  present  in  vety  numerous 
idiomorphs,  mostly  of  prismatic  habit;  sometimes  zoned;  often 
encloses  a  network  of  brown  glass  as  well  as  both  pyroxenes ;  calcite 
occurs  along  cracks.  Hypersthene  idiomorphs  are  numerous,  down 
to  very  small  dimensions ;  markedly  pleochroic :  a — ^pinkish  brown, 
^ — pale  straw-brown,  c — pale  bluish  green.  Augite  in  very  pale 
green,  almost  coloiu-less  idiomorphs  is  much  less  abimdant.  Both 
pyroxenes  have  sometimes  been  intruded  by  groundmass,  and  enclose 
magnetite  grains.  Magnetite  is  fairly  abundant.  The  groundbcass 
is  markedly  streaky — pale  grey,  fresher  streaks  alternating  with 
purplish  bands,  somewhat  weathered,  and  stained  with  iron  oxide,  but 
both  consist  of  a  very  fine  microfelt  of  densely  crowded  felspar 
microlites  (oligodase),  showing  fluidal  arrangement,  with  numerous 
magnetite  granules,  and  sparing  augite  granules,  in  a  colourless  or 
pale  brown  glass.  A  good  deal  of  CAuaTE  is  present  along  irregular 
cracks.    The  laiger  phenoaysts  occur  in  the  paler  streaks. 


AkfUat  to  Sfpan  169 

On  the  following  day  (August  4th)  we  climbed 
the  remaining  3000  feet  on  foot  up  the  great  lava 
flow,  which  became  glassy  and  streaky  in  places — a 
dull  black  glass  with  imperfedl  spherulites,  sometimes 
weathering  a  brick- red.  The  general  slope  did  not 
exceed  33^^  even  near  the  summit. 

After  skirting  the  south  side  of  the  western  peak 
(13,754  feet^  the  whole  of  the  summit-region  lay  open 
to  view.  In  spite  of  the  extensive  mantle  of  snow 
it  was  at  once  evident  that  we  were  standing  on 
the  edge  of  an  old,  elliptical  crater,  still  comparative- 
ly perfedt  on  the  north  and  south.  The  slightly 
shorter  N.-S.  axis  is  nearly  a  mile  long,  according 
to  my  measurements  with  Stewart's  Telemeter  (see 
plan). 

The  western  peak  of  Sipan  consists  of  a  glassy 
pyroxene-andesite,  and  is  a  later  cone,  which  has 
reared  itself  on  the  western  edge  of  the  old  crater,  and 
gave  rise  to  the  great,  W.S.W.  lava-flow,  up  which 
we  had  ascended.  At  a  still  later  date  a  pumiceous 
mica-andesite  with  subordinate  hornblende  (No.  147)  ^' 
has  formed  a  colossal  mass  on  the  N.E.  edge  of  the 
crater.  This  lava  was  evidently  highly  viscous,  and 
incapable  of  flowing  any  distance.  Its  eruption  must 
have  been  the  last,  dying  effort  of  the  great  volcano. 

*  The  rock  is  pale  grey,  streaky,  and  pumiceous,  with  abundant 
tiny  flakes  and  plates  of  biotite  (up  to  i  mnu),  and  glassy  felspars  (up 
to  5  mm.)y  which  are  conspicuous  on  the  brown,  weathered  surface. 
Specks  of  SULPHUR  occur  in  some  of  the  little  cavities.  Under  the 
microscope,  the  fslspab  is  mostly  plagioclasv  (andksinb  about  Abj 
Ana)  in  numerous  small  idiomorphs  of  prismatic  habit ;  clear  and 
fresh,  sometimes  zoned ;  occasionally  with  inclusions  of  pale  brown  or 
colourless  glass.  A  Uttle  subordinate  sanidine  is  also  present. 
BiOTiTS  occurs  in  abundant  flakelets  and  hexagonal,  dark  brown 
tablets,  slightly  resorbed.  Hornblende  is  very  sparingly  present, 
rKorbed  to  opacity  except^when  enclosed  in  felspar.  The  ground- 
ttASs  is  a  water-dear  glass,  highly  pumiceous,  with  felspar  microlites 
(oligoclase),  and  dust. 


170  Geology  of  Armenia 

The  numexous  small,  conical  peaks  on  the  summit 
of  this  eastern  mass,  and  its  long  screes  render  it 
strangely  like  Kerkur  in  appearance  (p.  11 6). 

A  small  secondary  crater  (without  any  emission 
of  lava)  has  been  formed  in  the  S.W.  quadrant  of 
the  crater,  and  is  now  partly  filled  by  the  crystal 
waters  of  the  Kirklar  Gol  (Lake  of  the  forty  holy 
men).  Another  sea-green  lake  at  the  western  base 
of  the  eastern  biass  is  clearly  due  to  the  melting  of 
the  snow. 

After  crossing  the  large  snowfield  in  the  western 
half  of  the  crater  we  finally  pitched  our  tent  in  the 
shelter  of  a  rock  on  the  narrow,  northern  rim  (13,000 
feet) — no  easy  task,  owing  to  the  hurricane  of  wind, 
which  rushed  up  at  sunset  from  the  vast  plain  of 
Melaskert  and  Patnotz,  8000  feet  beneath  lis.  A 
rounded  bastion,  a  little  to  N.E.  of  our  position, 
seemed  to  consist  of  grey,  volcanic  ash.  Before  it 
became  dark  we  climbed  up  to  the  eastern  summit, 
up  steep  slopes  of  large,  angular  boulders  of  the 
pumiceous  mica-andesite ;  the  rock  weathers  to  a 
pale  brown,  and  is  absolutely  barren — not  even  a 
lichen  can  gain  a  foot-hold  on  the  frothy,  glassy 
surface.  We  determined  the  altitude  of  the  highest 
peak  to  be  13,790  Feet.  Close  at  hand  is  a  xiaret — ^a 
cairn  surmounted  by  a  pair  of  ibex*  horns,  which 
have  been  covered  with  rags  by  the  women  who  visit 
this  arduous  place  of  pilgrimage.  Between  the  many 
conical  bosses  lilliputian  glaciers  occur,  with  crevasses 
up  to  4  reet  in  width  gaping  in  the  blue  ice. 

We  experienced  nothing  of  the  mountain-sickness 
which  Brant  ^  and  his  companions  suffered  in  their 
ascent  of  Sipan  from  Norshunjik,  but  we  foimd  it 
impossible  to  sleep.    Daylight  came  at  last  at  5  a.m.^ 

*  /mni.  Rcy^  Gtogr.  S0C*^  1840^  x  409. 


Akhlat  to  Sipan  I7I 

and  great  whirling  columns  of  mist  rose  up  swiftly 
from  the  plain  of  Melaskert,  and  spread  out  suddenly 
into  clouds  at  about  ip,poo  feet.  The  rest  of  my  time 
on  the  summit  prior  to  our  descent  was  occupied  in 
measuring  the  crater  by  means  of  Stewart's  Telemeter. 

Several  minor  cones  occur  on  the  lower  slopes  of 
Sipan,  but  our  time  did  not  admit  of  investigating 
them.  The  Eighri  or  Aiger  Gol  in  the  plain  at  the 
southern  base  of  Sipan  appears  to  occupy  a  shallow, 
secondary  crater,  with  grey  diSs. 

The  derivation  of  the  name  Sipan  or  Subhan  has 
[iven  rise  to  several  conjedtures,  mostly  unsatis- 
>ry,  e.g.^  Subhan  Allah =PtB,\se  to  God.  A  not 
improbable  explanation  seems  to  me  to  lie  in  the  fadt 
that  one  of  the  peaks  of  the  great  volcano  Kuh-i- 
Taftan,  in  Baluchistan,  at  the  bend  of  the  Southern 
Iranian  Arc  (p.  2),  is  named  Subh  Kuh^  which,  ac- 
cording to  mjor  Sykes,*  signifies  "the  peak  of  the 
morning.'*  To  the  inhabitant  of  Van  the  snowy  peak 
of  Sipan  (Subhan),  second  only  to  Ararat,  is  the  first 
object  to  receive  ^e  sun's  rays  at  dawn,  and  hence 
would  be  most  aptly  named  "the  mountain  of  the 
morning  sim.*' 

*  Urn  TkoMioni  MUa  im  Pimd,  133.    London,  1909. 


CHAPTER  XI 


SIPAN    TO    KHAMUR 


From  Uran  Gazi  our  course  lay  at  first  westward 
across  the  Jil  Gdl  plain  for  about  3  miles  to  t^e 
low  ridges,  which  rise  up  from  beneath  the  flood  of 
lava.  We  struck  the  third  ridge  north  of  the  Suteh 
Dagh  at  the  Kurdish  village  Karaghor ;  here  it  con- 
sists of  a  greenish  grey,  calcareous  sandstone  (No. 
J48),  dipping  15®  N.N.W.  It  is  friable,  and  very  fine 
in  grain — the  minute,  angular  particles  of  quartz, 
green  serpentine,  and  hematitic  and  limonitic  sand 
lying  in  a  sparing,  calcareous  cement.  The  outcrop 
forms  the  eastern  part  of  the  Khanik  ridge,  which  we 
had  crossed  on  July  6^^  whilst  riding  from  Melaskert 
to  Akhlat  (p.  96).  Probably  an  anticline  exists 
between  this  ridge  and  the  Suteh  Dagh. 

We  soon  descended  into  the  broad  Akhviran  valley 
(p.  95),  passing  through  Shebu,  and  leaving  Akhviran 
about  half  a  mile  to  the  westward.  Before  reaching 
Shebu,  however,  i.e.,  about  halfway  down  the  slope, 
the  sandstone  changed  its  dip  to  25®  S.S.E.,  and 
became  coarser,  containing  small  fragments  of  ser- 
pentine and  dark  red  porphyrite ;  it  is^  in  facSl,  very 
similar  in  composition  (though  finer  in  grain  and  free 
from  the  dark  red  colouration)  to  the  sandstones  and 
conglomerates  of  Akhlat  (pp.  98,  166).     It  is  probably 


Sipan  to  Khamur  173 

continiious  with  these,  but  must  have  been  deposited 
at  a  greater  distance  from  the  land. 

Shebu  lies  in  the  hollow  of  the  Akhviran  valley, 
which  finally  opens  out  into  the  plain  at  the  N.W. 
foot  of  Sipan.  The  walls  in  the  village  are  built 
of  black,  scoriaceous  lava,  and  of  a  hard,  white, 
oompadt  limestone,  crowded  with  shells  of  Dreyssensia 
Mymorpha^  Pallas.  The  stream  (flowing  to  E.N.E.J 
nas  cut  through  about  10  feet  of  a  grey,  alluvial 
deposit  (cp.  p.  95).  Large  blocks  of  the  lava  (doubt- 
less ejected  by  Sipan,  q>.  p.  164)  lie  scattered  on 
the  groimd. 

At  the  foot  of  the  northern  slope  of  the  valley 
we  crossed  a  larger  stream,  flowing  from  Akhviran. 
Here  I  found  the  Pleistocene  lake*deposits  in  situ; 
their  jimdlion  with  the  Miocene  strata  is  concealed  by 
the  alluvial  deposit  of  the  Akhviran  valley.  The 
lowest  visible  bed  on  the  left  bank  of  the  stream 
is  a  white,  travertinous,  luifossiliierous  limestone, 
dipping  35®  N.N.W. — a  rock,  in  facJk,  presenting 
precisely  similar  charadleristics  to  those  of  the  lime- 
stone of  Demian,  on  the  north  side  of  this  ridge  (p. 
95).  Higher  up  I  found  the  limestone  to  become 
hard  and  compadl,  full  of  Dreyssensia  polymorpha ;  it  is 
interbedded  with  grey,  calcareous  marls,  and  cor- 
responds to  No.  I  of  the  Demian  sedlion  ^p.  94^,  Near 
the  summit  of  the  ridge  I  foimd  small  individuals  of 
Bithinia  and  Melanopsis^'^  associated  with  numbers 
of  Dreyssensia  polymorpha  in  these  soft  marls. 
'*We  Irft  Demian  on  our  right  during  our  oblique 
descent  to  N.W.  into  the  Upper  Bulanik  plain — the 
basin  of  the  Murad.  The  broad  hollow  between 
Demian  and  a  lower,  more  westerly  ridge  (which  soon 

*  Kindly  identified  by  Mr.  R.  Sullen  Newton,  F.G.S.,  who  has 
also  ocmfinned  my  determination  of  the  Dr^ssauia  bom  this  plac^, 
and  from  other  localities  along  our  lottte. 


174  Geology  of  Armenia 

dies  out  to  E.N.E.)  is  composed  of  a  fine-grained,  grey 
calcareous  sandstone,  dipping  2d^  N.N.W.,  (No.  lo 
of  the  Demian  sedtion).  Here  I  noticed  an  inter- 
bedded,  contemporaneous  sheet,  about  ao  feet  thicks 
of  ophitic  olivine-basalt  (No.  149),*  similar  to  the 
wide-spread  basalt  between  Gopal  and  Tutakh  (pp. 
72,  76).  The  sandstone  is  overlaid  by  unaltered  lime- 
stone and  grey,  calcareous  marls,  full  of  Dreyssensia 
^lymorphaj  Pallas.  The  dip  changed  to  35  S.S.E. 
oefore  reaching  Leter,  at  the  edge  of  the  plauu 

These  observations,  in  conjundtion  with  those  made 
at  Demian  on  July  6^\  show  that  this  Pleistocene 
zone,  bounding  the  Melaskert  plain  on  the  souths 
consists  (from  N.  to  S.)  of  a  gently  folded  syncline, 
anticline,  and  second  syncline  ^ee  sedtion).  This  was 
the  only  instance  in  which  I  found  evidence  of  any 
folding  in  these  recent  lake-deposits ;  and  in  all  the 
occurrences  in  Armenia  recorded  by  Abich  and  others 
(see  Chapter  XXIV)  the  strata  are  either  horizontal  or 
only  slightly  uptilted.  The  shallowness  of  this  fold, 
however,  and  the  fadt  that  it  occurs  on  the  steep 
flank  of  the  high,  Miocene  block  of  land  (2000  feet 
above  the  Melaskert  plain)  induce  me  to  regard  it 
as  due  to  a  local,  lateral  slipping  of  plastic  strata 

*  The  rock  is  black,  compact,  -and  ytry  fine-gnined,  no  pheno- 
crysts  being  visible  with  the  lens.  An  occasional,  small  steam-cavity 
is  filled  witib  a  seolite,  and  there  is  local  effervescence  with  add.  The 
rock  is  slightly  magnetic  (a^.  Under  the  microscope,  ouvmB  is 
fairljr  abundant  in  small  idiomorphs,  usually  fresh,  but  sometimes 
partially  seri)entinized.  Auoitb  is  present  in  hum  pale  purplish 
brown,  ophitic  plates,  formbg  the  bidk  of  the  dide ;  usually  quite 
flesh,  but  along  certain  sones  (piobaUy  of  easy  percolation)  it  has 
underline  some  alteration  into  a  Politic,  finely  fibrous  minenJ, 
sometimes  stained  with  iron  oxide,  and  still  enclosing  isolated  patches 
of  augite,  to  wbiA  the  fibres  stand  aotmaL  Pla(hoclasb  (labra- 
DORTTS  about  AbxAni)  occurs  in  abmdanti  small  laths  in  fluidal 
arrangement,  enclosed  in  the  angite.  ICAOMBTm  grains  are  fairty 
frequent 


Sipan  to  Khamur  175 

upon  an  inclined  base.  The  plain  of  Melaskert  seems 
to  me  to  be  an  area  of  relative  depression,  like  the 
plains  of  Erzerum  (p.  46),  and  of  Khinis  (p.  62),  and 
the  southern  bounding-iault  from  Bilejan  to  Sipan 
cuts  obliquely  across  the  rounded  E.N.E.  ridges  of  the 
lake-deposits  (see  map).  Owing  to  this  circumstance 
we  traversed  higher  beds  again  on  nearing  Leter — a 
travertinous  limestone  f=No.  9  of  the  Demian  sec- 
tion). It  has  been  broicen  through  by  an  elliptical 
boss  of  lava — sl  comx>adt|  iron-grey  augite-andesite 

?io.  150),  with  numerous  plagioclases  up  to  2  mm. 
he  long  axis  of  the  boss  runs  from  W.N.W.  to  E.S.E. 
Leter  itself  is  built  on  a  similar  mound  of  lava ;  and, 
as  we  proceeded  westwards,  the  horizontality  of  the 
vast  plain  was  interrupted  only  by  these  lava-bosses, 
similar  to  those  of  the  Jil  Gdl  plain  (p.  166),  but  on  a 
larger  scale.  They  probably  indicate  the  position 
of  the  fissures  whence  the  lava  welled  up  to  flood  this 
plain  of  Upper  Bulanik  and  Melaskert. 

The  Pleistocene  zone  (ever  decreasing  in  height), 
to  which  our  route  now  lay  nearly  parallel  at  a 
distance  of  3  or  4  miles,  seemed  to  have  been  covered 
up  and  overwhelmed  by  lava.  On  looking  south- 
wards, up  a  transverse  valley,  I  could  see  a  high 
escarpment  of  pale  buff  strata  (about  7  or  8  miles 
south),  which  in  all  probability  is  the  westerly  con- 
tinuation of  the  similar  Khanik  escarpment  of  Mio- 
cene limestone  (p.  96). 

After  passing  through  Kekeli  we  halted  on  an 
isolated,  meridional  hill  rising  out  of  the  plain.  Here 
we  obtained  a  fine  view  of  Lake  Bulama,  due  west  of 
our  position,  with  the  Bilejan  Dagh  on  the  further 
side.  Its  steep,  rounded  spurs,  radiating  from  a 
serrated,  N.-S.  ridge,  form  a  conspicuous  feature  in 
the  landscape.  Bilejan  is  evidently  an  old  volcano, 
which  has  undergone  considerable  denudation,  only 


176  Geology  of  Armenia 

leaving  the  harder  ribs  and  core — ^the  ^'castSy**  so  to 
speak,  of  the  channels  up  which  the  viscous  lava 
welled.  In  short,  it  illustrates  admirably  the  '*  half- 
closed  umbrella  "  appearance  so  characteristic  of  those 
volcanoes  which  have  been  largely  formed  of  tuff  {pp. 
TeneriffeJL 

Lake  Bulama,  Kesan,  or  Shailu — as  it  is  variously 
termed — is  nearly  circular,  2-3  miles  in  diameter,  and 
by  no  means  attractive  to  the  senses,  for  its  colour  is  a 
muddy  brown  (J3tflafna= muddy),  and  it  possesses  the 
penetrating  odour  of  rank  manure.  Nevertheless  its 
waters  are  capable  of  supporting  animal  life.  Shells 
of  a  Unio  lie  scattered  along  the  gently  shelvin] 
sandy  shore ;  pelicans  abound,  and  two  small  islan< 
are  white  with  their  droppings.  The  presence  of 
these  birds  must  imply  an  abundance  of  fish. 

A  small  sample  of  the  water,  which  I  coUedted,  was 
submitted  for  analysis  to  the  late  Mr.  William  Thorp, 
B.Sc.,  with  the  following  result : — 

'*  Quantities  estimated  in  parts  per  100,000. 

...         .••         ...  *35 


Chlorine  .. 
Sulphates.. 
Nitrates  .. 
Sodium  and 
Magnesia.. 
lime 

Iron  oxide 
Silica 
Alumina  .. 


•  ••             ...            •••  O.  OS 

potassium  carbonates  8.80 

•  ••                      •••                     •••  laSQ 

•  ••            •••             •••  2.TI 

•  ••                     •••                     •••  OaOO 

•••          ...          •••  3*5 

I.7I 


•  !• 


Total  solids  in  solution S5.86 

Suspended  matter         3x*33 

The  water  is  slightly  ferruginous  and  yet  slightly  alkalina  The 
unpleasant  odour  from  the  lake  doubtless  arose  from  the  fermentatioii 
of  much  vegetable  matter  in  suspension  and  solution ;  it  could  not 
be  due  to  sulphur  compounds,  since  there  is  an  absence  of  sulphates, 
and  the  low  proportion  of  ddoiine  indicates  freedom  from  animal 
contamination." 


Sipan  to  Kkamup  177 

The  shore  to  the  north  and  east  of  the  lake  is  low 
and  rocky;  to  the  south  and  west  there  are  broad 
stretches  of  sand.  The  level  of  the  water  is  5550  feet, 
but  the  appearance  of  the  shore,  especially  of  the 
m^tfshy  traA  on  the  south  side,  renders  it  probable 
that  the  lake  was  once  about  50  feet  higher.  It  is 
drained  at  its  N.E.  end  by  a  small  stream,  flowing 
to  Gop  by  a  rather  rapid  and  rocky  descent. 

On  looking  northwards,  over  the  Murad  plain,  I 
could  distinguish  a  very  low  line  of  flat-topped  hills, 
rising  out  or  the  plain  on  the  south  side  of  the  river ; 
they  are  grassy,  with  bare,  white  patches  here  and 
there.  In  all  probability  they  consist  of  the  lacustrine 
marls  and  limestones  which  I  found  a  few  miles 
further  west  {vide  infra).  Koch  indeed  traversed  this 
very  district  between  Yimgali  (Yondjalii)  and  Melas- 
kert,  and  states  *  that  "  here  and  there  rocks  appeared 
of  a  dazzling  white,  Tertiary  limestone;  the  soil 
besides  was  mixed  with  shell-remains."  Furthermore 
on  the  north  bank  of  the  Murad,  "white  Tertiary 
limestone  again  composed  the  moderate  heights."!  I 
have  already  (p.  92)  given  reasons  for  regarding  the 
latter  to  be  laJce-marls  with  Dreyssensia  polymorpha, 
and  the  two  occurrences  are  doubtless  continuous. 

To  the  south  of  our  position,  the  linking  heights 
between  Sipan  and  Bilejan  are  terminated,  south  of 
the  lake,  by  a  curved  ridge,  which  has  somewhat  the 
appearance  of  a  crater,  but  it  may  be  merely  a  large 
cirque  open  to  the  S.E.  Nimrud  was  still  visible 
across  the  low,  lava  heights  surrounding  Lake  Nazik, 
which  forms  the  subjedt  of  the  following  digression. 

Lake  Nazik,  which  we  visited  on  July  30tb,  during 
our  ride  round  Nimrud,  stands  at  a  higher  level  (6380 

^  Rdsi  imponHsdim  GMrge^  tL  495. 
t  Op.  di.^  ii.  497. 


lyS  Geology  of  Armada 

feet)  than  Lake  Bulama,  and  agrees  with  it  in  being 
closed  in  by  lava-flows.  It  is  roughly  triangular  in 
shape,  the  apex  lying  to  the  north,  and  probably 
marking  the  in-flow  of  a  river,  although  this  was  not 
clearly  visible  from  the  southern  shore.  The  N.W. 
shore  is  formed  by  low  spurs  of  lava  from  Bilejan,  the 
southern  shore  partly  by  lava-flows  from  the  Kekeli 
Dagh  (a  symmetriod,  volcanic  cone,  south  of  Bilejan), 
and  by  a  black,  scoriaceous  lava,  wliich  has  welled  up 
from  a  meridional  fissure.  This  is  now  defined  by 
a  long,  low  ridge  of  tiie  lava — a  pyroxene-andesite, 
with  accessory  olivine  (No.  151)* — jutting  out  into 
the  lake,  forming  a  bold  headland  west  of  Jezirok.  A 
small  island — ^the  resort  of  iimumerable  pelicans-^ 
continues  this  volcanic  line  into  the  lake,  while  it 
is  itself  the  northerly  continuation  of  the  meridional 

*  The  rock  is  ooandy  scoriaceous  (steam-hoks  iq>  to  7  mm.),  with 
rough,  irregular  fiacture.  It  *^ft"»^«M  nimieroas,  ^aisj,  striated 
felspars  (up  to  r3  mm.),  and  a  few,  dark  green  bnmates  (up  to  a 
mm.).  Under  the  microscope,  the  plagioclasb  {labradobitb  about 
AbiAni)  occurs  in  laige  idiomoiphs  of  prismatic  habit,  sometimes 
in  groups ;  often  finely  soned ;  ytxf  fresh  and  free  from  induaions, 
but  rarely  with  a  network  of  £^m,  and  augite  granules ;  often  cracked. 
BRONzrra  idiomorphs  are  sparing^  present;  very  pale  yellow-green, 
slightly  pleochroic  (to  pale  pinkish  brown) ;  with  somewhat  corroded 
terminations,  but  otherwise  imaltered;  the  larger  crystals  contaj^n 
small  inclusions  of  brown  glass  (with  bubble),  and  granules  of  augite 
and  magnetite.  OLnriN s  is  fairly  frequent  in  colourless  crystals,  down 
to  quite  small  dimensions ;  sometimes  with  orange  stains  of  iron  oxide 
along  cracks ;  one  crystal  is  surrounded  by  a  wreath  of  smaQ 
bronzites.  Aitgitb,  very  pale  green  in  colour,  rarely  occurs  as  a 
phenocryst,  but  is  occasionally  inteigiown  with  bronzite.  The 
OROuiTDifASS  is  a  dark  brown  olass,  densely  crowded  with  skeletal 
needles  of  magnetite,  and  dust,  and  thereby  often  rendered  opaque; 
it  contains  abundant  plagiodase  ladu  (andesine  about  Ab4An3)  m 
flutdal  arrangement,  and  numennu  augite  granules.  The  steam-holes 
are  sometimes  lined  by  a  pale  brown,  mammillated  noLiTB;  under 
the  microscope  it  appears  water-dear,  nearly  isotropic  with  stains 
of  iron  oxide,  and  showing  concentric  Ihies  of  growth. 


i 

x 


Sipan  to  Khamur  179 

seme  of  weakness,  whic^  biaedts  the  Nimrud  crater 

(pPr  I3h  153)- 
The  eastern  shore  seemed  to  be  likewise  composed 

of  lava,  but  it  was  not  evident  whence  it  had  pro- 
ceeded. A  large  stream  enters  the  lake  at  its  S.W^ 
angle,  flowing  gently  through  an  alluvial  plain  about 
han  a  mile  wide.  The  stretches  of  sandy  beach 
between  the  low,  rocjky  headlands  show  that  Lake 
Nazik  stood  recently  at  a  slightly  higher  level, 
amoimting  to  about  10  feet  A  small  samjs^le  of 
the  water,  which  I  coUedted,  was  submitted  to  the 
late  Mr.  William  Thorp,  B.Sc.,  with  the  follo\<ring 
result: — 


it 


Qnaatities  esdouted  in  parts  per  xoO|Ooo. 

Chlonnc  •••       •••        ...        ...  1.50 

SnlpiUitei...        ...        ...        ...  ••• 

riitiAtct  •••        •••        .••        ...  0*05 

Sodium  and  potasdam  cvbonatet 

aWapiCBMt ...         .••         .«.         ...  ••* 

i.iUn9       ...        •••        ...        ••*  3*3' 

Iron  ondc         •••        ...        ...  0.01 

oiuca       ...         .*•         ...        ...  .«• 

Alimina  ...        ...        ...        ...  0.94 

Total  tolidt  in  loltttion 18.74 

Suspended  matter         0.36 

A  soft  water,  with  very  litde  contamination.*' 


to  Layard*  the  lake  is  drained  at  its 
S£.  angle  by  a  river,  which  must  eventually  join 
the  Akhlat  river  flowing  into  Lake  Van.  We  were 
unable  to  verify  this  statement  personally,  but  it  was 
confirmed  hv  the  kaimakam  of  Akhlat,  who  knew 
his  district  thoroughly. 

After  leaving  Lake  Bulama  we  rode  over  lava 

*  J)is£aomtt  h  ^  Rmm  rf NbiiOik  mii  B^^flm^  si. 


i8o  Geology  of  Armmtia 

slopes,  down  the  valley  of  the  river  which  drains 
it,  passing  through  the  village  of  Sheikh  Yakub. 
Thence — so  far  as  the  increasing  darkness  would 
permit  of  observation — the  lower  slopes  of  the  grassy 
spurs  on  the  way  to  Gop  consist  of  beds  of  brown 
sand  and  gravel,  which  may  perhaps  be  old  river- 
gravels  of  uie  Murad  plain. 

Gop  (5150  feet)  lies  in  the  plain,  at  the  foot  of 
a  northern  outspur  of  Bilejan.  The  well-cultivated 
fields  sink  gently  into  the  Murad  plain,  and  display 
no  sections.  On  the  steeper  slopes  south  of  the 
village   blocks   of   white    limestone,   crowded   with 

Dreyssmsia  polymorpha^  Pallas, 
frequently  occur;  the  rock  must  occur  in  situ  at 
no  great  distance.  A  more  definite  observation, 
taken  a  little  further  south,  was  the  discovery,  just 
north  of  the  Surp  Daniel  monastery,  of  a  grey,  marly 
clay  with  the  following  fossils,  in  which  even  the 
brown  epidermis  is  welt  preserved : — 

DreysMensia  polymorpha^  Pallas. 
Viviparus  wviparus^  Miiller.* 

These  beds  dip  ao^  S.E.,  and  are  overlain  by  lava 
from  Bilejan.  They  would  appear  to  belong  to  a 
slightly  higher  horizon  than  the  Dreyssensia-limestone 
of  Demian  (pp.  94,  173). 

From  Gop  our  course  was  diredted  through  Yimg*^ 
ali,  where  we  forded  the  Kor  Su,  to  the  confluence 
of  the  Bingol  Su  (from  Khinis)  with  the  Murad.  The 
wide,  alluvial  plain  is  much  cultivated:  the  com- 
bination of  calcareous  mari  with  decomposed,  basic 
lava  forms  a  most  fertile  soil,  excellent  for  com; 
Bulanik  is  in  fad  richer  in  grain  than  Pasin. 

The  Murad  at  the  confluence  is  over  |pp  yards 

*  Kindly  detennined  by  Mr.  R.  BuUen  Newton,  F.G.S. 


Sipan  to  Khamur  |8| 

in  width  and  quite  shallow.  Like  all  the  larger 
rivers  of  the  Armenian  plateau  {cp.  pp.  34,  41),  the 
Murad   meanders   sluggishly    through    the    alluvial 

Slain,  and  backwaters  are  frequent.  The  Bingol  Su 
ows  even  more  lazily,  and  is  hardly  more  than  40 
yards  ii\  width,  but  it  brings  down  a  considerable 
volume  of  water,  even  in  .August.  The  low  cliffs 
on  its  western  (right)  bank,  to  which  we  crossed, 
consist  of  lake-deposits.  The  section  showed  white 
sands,  current-bedded,  overlying  grey,  sandy  clay, 
which  dips  25*^  S.E.,  and  contains  the  following :  — 

Dreyssensia  polytnarpha^  Pallas. 
Viviparus  viviparus,  Miiller.* 

This  bed  shows  a  close  agreement  with  that  of  Surp 
Daniel  on  the  south  side  of  the  Murad  plain  (p.  180). 

At  Kara  Ogli  we  met  with  a  black,  scoriaceous 
lava,  forming  part  of  a  low,  semi -volcanic  zone, 
which  links  Bilejan  to  Khamur,  and  divides  the 
Bulanik  plain  into  an  upper  and  a  lower  portion,  just 
as  the  Deveh  Bojrun  (likewise  consisting  of  lacustrine 
deposits  with  interbedded  sheets  of  basalt)  separates 
the  plains  of  Erzerum  and  Pasin. 

The  Murad  enters  this  zone  by  a  deep  ravine,  about 
a  couple  of  miles  below  iCara  Ogli.  At  first  sight 
it  seemed  remarkable  why  the  river  should  not  have 
diosen  the  lower  part  0!  the  zone,  a  little  to  the 
south.  An  exadlly  similar  waywardness  on  the  part 
of  the  river  occurred  at  Sheep,  further  down  the 
Murad  valley  (Chapter  XII).  These  two  instances 
appear  to  me  to  indicate  that  the  region  south  of 
KhsLmuT  has  undergone  a  slight  amount  of  tilting 
up  from  the  north,  and  that  the  Murad  has  been 
able  to  retain  and  deepen  its  original  bed  by  keeping 

•  Kindly  determmed  by  Mr.  R.  Bollen  Newton,  F.G.& 


1 


i8a  Geology  of  Armenia 

pace  with  the  recent  uplift    The  nearly  vertical 

of  its  ravine  in  both  cases  may  perhaps  indicate 

a  comparatively  rapid  elevation  of  the  gromid. 

We  rode  alons  the  north  side  of  the  ravine  at  some 
distance  above  the  river,  virhich  now  forgot  its  former 
sluggishness  from  Tutakh  downwards,  and  rushed 
along  with  the  impetuosity  of  a  mountain-torr^itt 

Esnt  up  between  narrow  walls  of  black,  coliunnar 
va.  it  soon  became  evident  that  this  lava  is  inter- 
bedded  with  the  lake-marls,  which  dip  15^  N.W.,  and 
contain  the  following  fossils : — 

Dreyssensia  polymorphaf  PallasJ 
Unto  tumidus^  Ketzius.^ 
Viviparus  cf.  vhiparus^  MuUer. 

The   successive   lava-flows   were  clearly  contem- 

E)raneous  with  the  deposition  of  the  marls,  for  the 
tter  have  been  disturbed  and  baked  when  beneath  a 
lava-sheet,  but  are  unaltered  above  it.  A  small  fault, 
hading  to  the  S.E.,  was  also  visible,  and  the  right 
bank  showed  the  following  succession : — 

I.  Olivine-buak  (Na  153,  tridi  mfm)   ...       c.  40  feet. 
9.  Laciutrine  marls         €.  $0   „ 

3.  Qlivine-bMalt  fNo.  xs9)t      a  80   „ 

4.  Lacuttrine  marls  with  shells,  as  above.  c  100  „ 

5.  Olivine-basalt c,  xoo   „ 

Level  of  the  Marad. 

*  KindlijT  determined  by  Mr.  R.  BuUen  Newton,  F.G.S.  The 
shells  in  tms  bed  were  much  cracked. 

t  The  rock  is  black  and  compact,  with  numerous,  striated  felspars 
(np  to  3  mm.),  and  frequent,  greenish  olivines  (up  to  t  mm.),  some 
of  which  are  stained  a  reddish  brown  with  iron  oxide.  Under  the 
niicroscope,  the  plagioclasi  (labradoritb  about  AbsAnd)  occurs 
in  hrge,  idiomorphs  of  prismatic  habit,  occupying  about  half  the 
slide ;  rerf  fresh  and  free  from  inclusions ;  offam  in  groups ;  some- 
times faindy  soned.  Outinb  is  present  in  abundant  idiomoiphs, 
down  to  smaU  dimensions ;   frequently  in  groups,  and  sometimes 


i/il 


Sipan  to  Khamuy  183 

Our  track  ascended  gradually  over  these  beds  in 
upward  succession.  We  left  the  river  just  where 
it  takes  a  sharp  bend  to  the  S.W.,  and  is  spanned 
by  a  modem  wooden  bridge.  Our  short  cut  across 
the  river-loop  traversed  a  nne-^rained  olivine-basalt 
(No.  I53)»* — No.  I  of  the  preceding  sedlion  (p.  182). 

The  small  village  of  Bitlu  lay  a  little  to  the  north 
of  our  route^  at  the  foot  of  a  high,  E.-W.  ridge  of 
lava,  which  runs  nearly  parallel  with  the  Murad  from 
Kara  Ogli,  gradually  increasing  in  height.    We  soon 

inyadod  by  groandmass ;  often  partly  rq>laced  by  caldte.  maTginally 
aad  along  cncks,  whidi  aie  stained  dark  brown  by  iron  o5udc ;  some- 
times wholly  altered  into  calcite^  with  subordinate  hyalite  anl 
delessite.  The  gboundmass  is  a  radier  coarse-grained  felt  of  plagio- 
clase  laths  (labradorite  about  Ab jAn6)^  with  interstitial  grains  of  pale 
purplish  brown  augite,  and  spanng  needles  of  magnetite.  Irregular 
cavities  frequently  occur,  now  filled  with  calqitIi  in  fibrous,  radiating 
masses,  often  stained  dark  brown,  especially  maiginally,  by  ir^n. 
oxide.  Some  cavities  have  been  filled  with  hyaute,  either  quite 
isotropic,  or  only  veiy  feebly  anisotropic,  sometimes  with  concentric 
lines  of  growth.  In  both  cases  there  is  always  a  narrow  lining  of  pal  i 
delessite 

*  The  rock  is  compact,  and  greenish  black,  with  abundant  felsp::rs 
(up  to  2  mm.),  and  numerous,  slightly  smaller  reddish  brown  crystals 
(altered  olivines).  A  little  effervescence  with  add  occurs  here  and 
there.  Under  the  microscope,  ouvinb  is  present  in  abundant,  small 
idiomorphs,  sometimes  fresh  and  colourless,  but  for  the  most  part 
altered,  wholly  or  partially,  into  iddingsitk  (a  rich  orange-brown, 
sightly  pleochroic  to  a  darker  tint,  and  occarsionally  showing  fine, 
parallel  cleavages);  the  alteration  usually  b^ins  centrally,  leaving 
a  narrow  margin  of  unaltered  olivine.  The  oroundkass  is  a  rather 
coarse  felt  (but  finer  than  in  the  last  rock)  of  plagioclase  laths  (labra- 
dorite about  AbiAni)  in  fluidal  arrangement,  abundant  granular,  sub- 
opbitic  augite,  and  small  crystals  and  needles  of  magnetite,  with 
a  little,  pale  brown,  interstitial  glass,  which  seems  to  be  undergoing 
alteration  into  delessite;  this  mineial  also  often  forms  a  narrow 
lining  to  irregular  steam-cavities  filled  with  caldte.  The  augite  was 
probably  originally  in  ophitic  plates,  which  became  broken  up  b^ 
differential  movements  prior  to  final  consolidation,  so  that  the  largest 
remaining  fra^  nents  are  only  able  to  endose  a  few  felspar  laths. 


184  Geology  of  Armenia 

descended  into  the  Murad  valley  again,  now  dbout 
2  miles  in  width,  with  gentle  contours  owing  to 
the  exclusive  presence  of  lacustrine  marls  and  days, 
which  are,  however,  encircled  by  hills  and  ridges 
of  lava.  A  specimen  of  lava,  taken  just  previously 
to  our  descent  into  the  valley,  is  a  glassy  h3rpersthene* 
andesite  (No.  154),*  later  in  date  than  the  olivine- 
basalts. 

At  this  spot  we  had  turned  the  position  of  Bilejan, 
and  the  great  bulk  of  Nirarud  was  again  clearly 
visible  in  the  S.E.  The  north  bank  of  the  Murad 
in  this  part  of  the  valley — just  south  of  Arenjik— 
consists  of  mounds  of  calcareous  lake- marls,  overlain 
further  north  by  a  thick  sheet  of  the  hypersthene- 
andesite.  The  greyish  white  marls  dip  ^o^  S.E.,  and 
contain  a  thin  layer  of  fragments  of  Dreyssensia 
polymorpha  and  Viviparus.  The  river  flows  along  the 
strike  of  these  beds:    in  fadt,  it  occupies  a  longitu- 

*  The  rock  is  compact,  and  iron-grey,  with  irregular  fracture.  It 
contains  abundant,  striated  felspars  (up  to  6  mm.),  and  numerous, 
grtenuh  yellow  pyroxenes  (up  to  3  mm.),  with  a  very  few,  tiny  steam- 
pores.  Under  the  microscope,  the  plagiocuisk  (labradobite  about 
AbiAni)  occurs  vi  large  idiomorphs,  mostly  of  prismatic  habit; 
fairly  clear,  but  sometimes  with  inclusions  of  brown  glass  (usually  with 
bubble),  and  ai^gite  granules;  peridine  twinning  is  occasionally 
present,  in  add  tion  to  the  usual  albite  and  Carlsbad  plans;  rarely 
showing  faint  zoning.  A  second,  smaller  generation,  of  prismatic 
habit,  consists  of  anoesxne  (about  Ab4An3).  Hypersthenk  ia 
present  in  abundant  idiomorphs,  down  to  quite  small  dimensions; 
colourless  or  very  pale  brown,  showing  marked  pleochroism :  a-^ 
pinkish  brown,  d — brownish  yeUow,  <'— green,  all  in  pale  tints. 
Aug  ITS  in  pale  green  idiomorphs  is  less  frequent,  but  on  the  whole 
larger  than  the  hypersthene;  sometimes  twinned;  occasionally  in 
groups.  The  larger  grains  of  icagkbtttb  are  associated  with  the 
augite.  The  groundmass  is  a  compact,  andesitic  microfelt  of  felspar- 
Afiicrolites  (oligoclase),  in  crowded,  fluidal  arrangement,  with  a  lesl 
number  of  pyroxene  and  magnetite  granules  in  a  pale  brown  glass. 
The  glass  in  the  interstices  of  groupa  of  felspars  or  pyroxenes  is  purer, 
and  nearly  free  from  oystallites  and  microlites. 


Sipan  to  Khamur  185 

dinal  valley  for  30  miles  from  its  confluence  with 
the  Khinis-Bingol  Su  as  far  down  as  Derik.  I  noticed 
two  small  faults  in  the  marls,  with  a  downthrow 
of  a  foot  to  S.E.  (cp,  p.  182),  just  before  reaching 
Shakhberat.  The  level  of  the  Murad  at  this  place 
is  4900  feet — the  difference  of  150  feet  from  its  level  in 
the  Upper  Bulanik  plain  being  accounted  for  during 
the  passage  of  its  short,  narrow  gorge. 

The  plain  of  Shakhberat  is  bounded  on  the  west  by 
the  volcanic  height  of  Koli  Baba — on  the  N.N.W, 
by  the  great  mass  of  Khamur.  Here  I  could  see,  more 
clearly  than  before,  that  Khamur  is  not  a  volcano,  but 
a  mountain  of  denudation,  consisting  largely  of  white 
lake-deposits,  interbedded  with,  and  capped  by  lava. 

In  making  oxir  ascent  of  Khamur  from  Shakhberat, 
the  alternation  of  lake-deposits  with  sheets  of  basalt 
was  again  very  noticeable,  and  affedled  the  contour  of 
the  ground  in  a  marked  degree.  Our  course  at  first 
lay  nearly  due  north,  passing  over  calcareous  marls, 
dipping  40°  N.W.,  and  forming  gentle  undulations. 
Thence  we  rode  up  a  steep,  rocky  slope  of  basalt,  to 
reach  a  wide,  upper  terrace  of  marls.  A  second  sheet 
of  lava,  about  40  feet  thick,  overgrown  with  oak- 
scrub,  formed  another  giant's  step  to  a  narrower 
terrace  of  marls.  Its  continuity  is  interrupted  by 
a  sheet  of  lava,  about  20  feet  thick.  Yet  another 
outcrop  of  lava  occurred  before  reaching  Ganibuk, 
where  the  village  pond,  shaded  by  willows,  testifies 
to  the  clayey,  retentive  nature  of  the  lake-deposits. 
From  Ganibuk  upwards  our  path  lay  exclusively  over 
long  rocky  slopes  of  basaltic  lava,  which  became  more 
scoriaceous  as  we  ascended.  On  nearing  the  summit 
of  the  ridge  the  numerous  steam -cavities  in  the  lava 
become  long  and  drawn-out  along  the  lines  of  flow. 

We  were  now  on  a  broad,  grassy  platform  of  lava, 
sloping  very  gently  up  to  the  base  of  the  lava-cap 


i86  Geology  of  Armema 

of  Khamur.  Just  before  commencing  the  final  ascent 
we  stood  at  the  head  of  a  wide,  W.-E.  valley, 
opening  out  eastwards,  between  the  Zimek  Dagh  and 
the  Kara  Ogli  ridge.  Here  grey  marls,  dipping  30^ 
S.E.,  underlie  the  comparatively  thin  sheet  of  lava, 
which  has  baked  them  into  porcellanous,  yellow 
shales. 

The  an^le  of  the  lower  slope  of  the  lava-cap  on 
its  south  side  is  ii%  rapidly  increasing  to  i8^  and  the 
final  gradient  amounts  to  30^  The  rock  is  a  dark 
gi^ey*  gl^Lssy,  andesitic  oliyiae-ba^lt  (No.  155),^  with 
occasional,  large  hornblende  crystals.  An  allied  rock 
has  been  described  by  Tho6t;t  his  No.  74,  from  a 
point  S.W.  of  Gerjiisy  on  the  Central  iCarabagh 
plateau,  is  **  pyroxene-andesite  with  olivine,  con- 
taining completely  resorbed  hornblende.'* 

Perhaps  the  most  remarkable  feature  of  the  summit* 

*  The  rod(  if  fiady  porous  and  streaky,  with  scattered,  striated 
febpars  (up  to  6  mm.),  a  few  olivines  (up  to  s  mm.),  and  oocasionaDf 
a  conspicuous  hornblende  (sons  mm.)»  The  rock  is  markedly 
magnetic  (sq"*),  and  diere  is  a  little  effervescence  with  add  in  places. 
Under  the  microscope,  the  vlaoioclask  (AVMsma  about  AbjAns) 
idiomorfihs  are  mosUy  of  fvismatic  habit,  but  some  large,  tabular 
individuals  are  present;  pericUne  twinning  occasionally  oocmrs,  in 
addition  to  the  usual  albite  and  Carbbad  plans ;  faint  soning  is 
not  infrequent;  usually  veqr  free  from  indusions,  but  sometimes 
endoses  hornlJende  and  olivme.  Ouvma  occurs  in  large,  very  pale 
yellow  idiomorphs,  down  to  quite  small  dimensions  ;  endoses  apatite 
and  magnetite ;  sometimes  with  stains  of  iron  oxide.  Hornblende  is 
onl^  of  rare^  occurrencct.  in  brown  idiomorphs,  sometimes  showing 
a  slight  reSoibtion-border ;  pleodiroism :  0— pale  brown,  ^  and  ^very 
dark  brown,  c  being  practically  opa<^ue^  endoses  apatite.  Mao* 
NBTiTE  in  rather  large  grains  is  assoaated  with  die  olivines.  The 
GEouNDBf  ASS  IS  Streaky : — pale  brown,  neariy  colourless  glass  alter* 
nating  with  darker  brown  j^ass ;  it  b  crowded  with  febpar  laths  {ohgQ^ 
clase  about  AbjAni),  showing  fluidal  arrangement,  numerous  smaB 
otivines,  sparing  augite  in  very  pale  green  granules  or  little  crystab, 
i^undant  magnedte  granules,  and  occasional  bdoniles* 

t  Abhamdl,  StnOent.  Naturf.  Gu.  Ftankfurt  1^1  XVIII,  ii.  947^ 


i 

f 
\ 

{ 


c 

s.  m 
s  - 
.  «> 

1^ 


1 

A        % 

n 


i 


Sipan  to  Khamur  iSy 

r^on  of  Khamur  is  the  presence  of  the  two  Galen 
laxeSy  lying  on  a  broad  outspur  to  the  S.W.  The 
mass  of  Khamur  is  oontinued  eastwards  into  the 
lower  Zirnek  (Zemak)  Dagh,  the  summit  of  which 
is  not  a  ridge,  but  a  broad,  grassy,  synclinal  valley. 
The  high  edges  of  this  trough  fall  away  to  the  north 
and  south  respedtively  by  bare  cliffs,  composed  of  buff, 
evenly  bedded  strata  (probably  Miocene  limestone,  see 
p.  74,  and  Chapter  aXIII).  So  far  as  I  could  jud^e 
nrom  a  distance,  in  this  clear  atmosphere,  the  dip 
of  these  beds  (passing  from  south  to  north)  changes 
from  15^  N.N.W.  to  35^  S.S.E.  The  line  of  the  Zirnek 
Dagh  is  continued  by  lower  heights  further  to  the 
E.N.E.,  across  the  Bingol  Su  near  Bayaz  Tuzla  (p.  74), 
to  disappear  beneath  the  wide  basalt-plateau  of  the 
Lala  Dagh,  and  the  Torlu  steppe  (p.  81).  The  recent 
lake-deposits,  conspicuous  for  their  chalky  whiteness, 
flank  the  southern  slope  of  the  Zirnek  Dagh,  and  form 
the  low  cliSs  on  the  north  bank  of  the  Murad  up 
to  its  great  bend  near  Melaskert  (pp.  93,  177^. 

From  this  lofty  standpoint  01  Khamur  (9850  feet) 
I  could  also  connrm  my  previous  observation  (p.  69), 
that  the  Ak  Dagh  (north  of  Khinis  plain)  consists  of  a 
series  of  parallel,  W.S.W.-E.N.E.  ridges,  closely 
ranged  together  in  ^helon.  A  small,  upland  plain 
lies  a  little  south  of  the  centre  of  the  mass  of  barren, 
marble  peaks.  The  selvage  of  travertine  terraces 
along  the  southern  base  of  the  Ak  Dagh  (p.  69)  was 
also  conspicuous.  Darkness  precluded  any  obser- 
vations on  our  return  from  Khamur  to  Shakhberat. 


CHAPTER  XII 

KHAMUR  TO  THE  BINGOL   CLIFFS 

On  August  1 2th  we  resumed  our  journey  down  fhe 
Murad  valley.  The  greater  part  of  the  plain  between 
Shakhberat  and  KoU  Baba  is  composed  of  a  sheet 
of  basalt,  extending  to  the  south  of  a  terrace  of  lake- 
deposits.  After  crossing  the  opening  of  the  valley 
down  which  we  had  descended  the  previous  night 
from  Khamur,  we  left  the  basalt  sheet  (weathering  to 
a  brown  sand),  and  traversed  a  slight  depression  in 
the  ground,  composed  of  dark  grey  marl,  over  which 
the  lava  had  flowed. 

Here  we  left  the  Shakhberat  plain,  and  ascended  a 
southern  spur  of  Koli  Baba.  A  slanting  path  up 
the  cliff  brought  us  to  the  summit  of  the  ridge,  just 
south  of  a  small,  symmetrical  cone.  Our  route  now 
lay  westward,  across  an  extensive  lava-plateau.  The 
rock  is  a  grey  augite-andesite,  with  small,  prismatic 
felspars,  and  well-developed  augite  crystals  (up  to 
5  mm.).  This  plateau  narrows  to  the  north,  where 
it  separates  Koli  Baba  from  the  Galen  Gol  buttress  of 
Khamur.  As  we  rode  across  it,  a  black  mud,  cracked 
by  the  heat,  showed  that  this  shallow  basin  must 
have  been  a  lake  at  no  distant  period.  It  is  still 
marshy  in  the  centre,  and  full  of  rushes ;  a  small 
gorge,    notching   the  low,  southern   rim   (a  black, 


Kkamur  to  the  Bingol  Cliffs  189 

ooliuxmar  basalt),  shows  where  the  waters  were 
drained  away  to  the  Murad.  Maay  rounded  blocks  of 
lava  lie  scattered  over  the  surface  of  the  plain ;  and  a 
horizontal  linfe,  extending  round  the  S.  W.  base  of 
Koli  Baba,  about  50  feet  above  the  plain,  may  per- 
haps mark  a  former  beach-level  of  the  lake. 

The  western  rim  of  this  basin  is  formed  by  a  S.E. 
spur  of  oli vine-basalt  fNo.  156)  *  from  the  Galen  Gdl 
buttress,  and  here  we  looked  down  upon  the  Murad,' 
a  few  himdred  feet  below.  About  50  feet  beneath  our 
standpoint  a  wide  terrace  extended  parallel  to  the 
Murad,  but  still  high  above  its  right  bank,  and 
bounded  on  the  north  by  the  steep  lava-ciiSs  of  the 
Galen  Gdl  massif.  In  the  northern  part  of  this 
terrace  a  small,  evaporating  lake  lies  at  the  foot  of  the 
cliff. 

Whilst  descending  to  this  terrace  I  observed  that 
the  lower  part  of  the  spur,  beneath  the  olivine-basalt, 
consists  of  a  basaltic  augite-andesite,t  which  presents 

*  The  rock  is  daak  gny  with  a  parpKah  tinge,  shoiniig  panllel 
layers  1-2  mm.  thick ;  very  homogeneous.  An  occasional  steam-hole 
{wp  to  a  nmt)  is  lined  with  Uack  g^ass.  Upder  the  microscope, 
oi^y  a  very  few,  small,  tabular  phenocrysts  occur  of  plagioclask 
(LABRADORm  about  AbxAnx),  honeycombed  with  a  network  of 
groundmass,  whilst  showing  a  clear,  narrow  margin.  Olivinb  is 
sparingly  present  in  small  idiomorphs,  stained  brown  along  the 
margins  with  iron  oxide.  Adgits  occurs  in  small,  pale  pinkish  brown, 
ophitic  plates,  quite  fresh,  containing  felspar  laths  and  magnetite 
grains.  The  groundicass  is  a  dosdy  packed  felt  of  plagioclase  laths 
(andesine  about  Ab^Anj)  in  fluidal  arrangement,  with  abundant 
granules  of  augite  and  ma^etite,  and  much  pale  brown  or  colourless 
glass.  The  glass  becomes  more  dourly  visible  round  the  irregular, 
long-drawn  steam-cavities.  The  ophitic  plates  of  augite  (which 
resist  weathering  better  than  the  granules)  have  i»robably  been  broken 
t^>  to  a  great  extent  intQ  granides  by  differential  movements  prior 
to  final  consolidation. 

t  The  rock  is  a  dark  purplish  grey,  very  compact  and  fine-grained, 
with  broadly  conchoidal  fracture.    The  lens  only  reveals  a  very  few 


iQO  Geology  of  Armenia 

some  points  of  resemblance  imder  the  microscope  to 
the  lavas  of  the  eastern  plateau  of  the  Bingdl  Dagh 
^ee  Chapter  XIII),  and  to  Becke's  No.  22  from  the 
Dumly  Dagh,  and  his  No.  23  from  Chobankopri  in 
Pasin.* 

On  reaching  the  terrace  I  found  it  to  consist  of  grey 
marls  and  clays,  interbedded  with  sheets  of  basalt — 
thus  presenting  precisely  similar  characfteristics  (al- 
though I  could  see  no  fossils)  to  the  lake-deposits 
higher  up  the  valley  (p.  182).  Wherever  they  are 
overlain  by  lava  they  have  oeen  baked  into  buff, 
porcellanous  shales. 

As  we  rode  along  this  wide  terrace  I  noticed  that 
the  surface  is  strewn  with  frequent,  rounded  pebbles 
of  white  marble  and  brick-red,  scoriaceous  lava, 
exadlly  similar  to  the  pebbles  of  the  Murad  near  Kara 
Ogli,  and  at  Shakhberat.  It  seems  probable  therefore 
that  this  terrace  marks  a  former  level  of  the  Murad 
before  the  river  had  cut  its  way  down  to  its  present 
bed,  200-300  feet  below. 

At  Derik  springs  arise  at  the  edge  of  the  terrace,  at 
the  junction  of  a  lava-sheet  with  the  imderlying, 
impervious  clay. 

The  soft,  lacustrine  beds,  which  (judging  from  the 
strudlure    of    the    right  bank)   must   have   overlain 

felspais  (up  to  2  mm.)  and  fewer  augites  (up  to  z  mm.).  The 
rock  is  slightly  magnetic  (2"*).  Under  the  microscope,  the  sparing 
idiomorphs  of  plagioclase  (andesine  about  Ab3An2)  are  of  prismatic 
habit,  with  scanty  albite  twinning.  An  occasional,  irregular  grain 
of  Auarrs,  stained  by  iron  oxide,  is  also  present.  The  grouvdmabs  is 
a  very  fine  microfelt  of  tiny  plagiodase  laths  (oligodase)  in  fluidal 
arrangement,  with  abundant  granules  of  augite  (sometimes  stained 
b^  iron  oxide),  and  equally  numerous  magnetite  granules  with  a  httlei 
pale  brown  glass.  Under  a  high  power  some  of  the  characteristic, 
mibricated  aggregates  of  tridtmitb  are  visible.  Darker  streaks  oociir 
where  the  magnetite  granules  are  more  densely  crowded. 

*  Abidi,  Geal.  Fbrsch.  in  dm  kauk.  Landem^  iL  329. 


Khamur  to  the  Bingol  Cliffs  191 

fhe  broad  lava-sheet  sloping  down  to  the  opposite 
(left)  bank  o(  the  Murad,  can  only  have  been  denuded 
away  at  a  comparatively '  recent  period,  for  as  yet 
the  merest  runnels  have  been  carved  in  the  slope  of 
the  lava.  The  high  cliff  just  north  of  our  terrace 
furnished  a  very  dear  instance  of  the  interbedding 
of  a  basalt-sheet,  40-50  feet  thick,  in  grey  marls, 
which  dip  15°  N.W.  by  N.  This  is  also  approximate- 
ly the  inclination  of  the  lower- lying  lava-sheet  of  the 
opposite  (left)  bank  of  the  Murad.  A  little  further  on, 
the  low  clifiE  of  the  left  bank  no  longer  consists  of 
lava,  but  of  greenish  white  strata  (probably  lake- 
marls),  dipping  20°  N.W.  by  W.,  and  overlain  con- 
formably by  a  sheet  of  black,  columnar  lava,  30-40 
feet  thick. 

However,  the  lava  soon  predominates  on  the  right 
bank;  and,  shortly  before  reaching  Sheep  (Shiip) 
on  the  Murad,  we  passed  over  a  sheet  of  olivine- 
basalt  (No.  158).*  Here  the  terrace  of  the  overlying 
lake-marls  extends  for  about  f  of  a  mile  to  the  north. 

A  little  further  down  the  valley  I  noticed  a  second 

*  The  rock  is  daik  grey  with  a  slight  greenish  tinge;  it  is  fine- 
grained and  very  homogeneous,  with  broadly  conchoidal  fracture ; 
even  the  lens  only  reveals  an  occasional  tiny  felspar  or  olivine. 
Under  the  microscope,  only  sparing  phenocrysts  occur  of  plagioclasb 
(labraoorttb  about  Ab5An6),  of  tabular  habit;  sometimes  in 
groups  ;  frequently  faintly  zoned ;  very  fresh  and  free  from  inclusions, 
only  seldom  enclosing  olivine.  Ouvinb  occurs  in  a  few,  large  idio- 
morphs,  but  mostly  m  a  generation  of  smaller  individuals ;  much 
replaced  by  caldte,  sometimes  entirely ;  much  stained  by  iron  oxide 
along  cracks.  The  oroundmass  is  a  felt  of  plagioclase  laths  (andesine 
about  AbjAna)  in  fluidal  arrangement,  with  abundant  granules  of 
au^te  and  magnetite,  and  much,  dark  brown,  interstitial  glass,  witii 
belonites.  A  considerable  amount  of  CALcrrB  is  present,  especially 
along  certain  zones,  as  if  it  had  infiltrated  from  the  calcareous  marls, 
in  which  this  lava  is  interbedded ;  it  occurs  in  dusty,  radially  fibrous 
aggregates,  either  pale  yellow-brown,  owing  to  staining;  by  iron  oxide, 
or  occasionally  with  a  greenish  tinge,  due  to  chlorite;  apparently 
it  has  filled  up  small,  irregular  steam-pores. 


192  Geology  of  Armenia 

instance  (see  p.  181)  of  the  Murad  having  cut  a  deep 
ravine  through  hard  basalt  instead  of  keeping  to 
the  lower  and  softer  ground,  a  little  to  the  south. 
The  probable  reason  of  this  apparent  anomaly  has 
already  been  suggested  (pp.  181,  182). 

So  far  as  the  increasing  darkness  would  permit 
of  observation,  the  remainder  of  our  day's  march  to 
Chsurbahur  appeared  to  lie  over  the  same  series  of 
lake-deposits,  forming  undulating  ground,  with 
gentle  contours. 

Charbahur  (4760  feet ;  30  miles  from  Shakhberat)  is 
situated  about  a  couple  of  miles  north  of  the  great 
bend  of  the  Murad,  where  it  suddenly  changes  its 
course  from  W.N.W.  to  due  south.  This  spot  marks 
also  its  confluence  with  the  Southern  Bingol  Su, 
which  flows  from  the  W.N.W.,  draining  the  southern 
half  of  the  Bingol  plateau. 

On  the  right  bank  of  the  Murad,  just  above  this 
confluence,  I  noticed  (from  a  distance)  that  the  high 
cliff,  about  i^  mile  north  of  the  river,  displayed 
five  successive  sheets  of  lava  (30-100  feet  thick), 
alternating  with  beds  of  lake-marls  (50-150  feet 
thick) — the  whole  'series  showing  a  dip  of  about  20® 
N.W.  by  N. 

After  crossing  the  alluvial  plain  between  Charbahur 
and  the  confluence,  we  forded  the  Bingol  Su  (here 
about  30  yards  wide),  and  traversed  a  black  olivine- 
basalt,    (No.    159),*^    forming   the   gentle,    northern, 

*  The  rock  is  fine-grained  and  compact,  with  rough,  uregular 
fracture.  Even  with  the  lens  only  a  few  tiny  felspars  (less  than  i 
mm.),  and  smaller  specks  of  olivine  and  iron  pyrites  can  be  discerned. 
The  rock  is  highly  magnetic  {(20°).  Under  the  microscope,  there 
are  only  a  very  few  phenocrysts  of  plaoioclasb  (labradorite  about 
Ab5An6),  of  prismatic  habit ;  honeycombed  by  a  rather  fia&jietwork 
(chiefly  peripheral)  of  dusty  glass.  Olivinb  is  extremely  abundant,  in 
small,  very  pale  yellow  or  colourless  idiomorphs ;  sometimes  encloses 
magnetite.     Magnetite  occurs  in  rather  numerous  grains.     A  few 


Khamur  to  the  BingSl  Cliffs  793 

slope  of  the  block  of  land,  through  which  the  Murad 
has  cut  a  deep,  transverse  gorge.  Underlying  this 
sheet  of  basalt  the  white,  lacustrine  marls  occur, 
apparently  dipping  N.W.,  but  no  clear  sedlion  was 
visible. 

About  i^  mile  south  of  the  confluence  we  began 
^  enter  the  adtual  gorge.  Here  a  buff,  conglomeratic 
limestone  appeared,  dipping  25^  N.N.W.,  very  similar 
in  charadler  to,  and  perhaps  continuous  with  the  bed 
noticed  south  of  Khanik  (p.  97)  of  Miocene  age.  The 
river  rushes  in  rapids  over  the  long,  hard  ridges  with 
E.N.E.  strike.  The  jundtion  with  iJie  soft,  overlying 
lake-marls  was  unfortimately  overgrown,  and  not 
visible,  but  the  unconformity  cannot  be  great. 

We  forded  the  Murad  near  a  guardUiouse,  about 
3  miles  south  of  the  confluence;  the  current  was 
strong,  and  the  water  rose  above  our  stirrups.  Here 
the  river  is  120  yards  wide,  and  the  level  is  only  4570 
feet — the  lowest  level  reached  during  our  journey 
through  Turkish  Armenia. 

The  dip  of  the  limestones  was  still  to  N.W.,  but 
at  a  short  distance  further  south  they  bend  over  to 
about  30^  S.S.E.  We  ascended  the  high,  receding 
cliSs  of  the  buff  limestones  on  the  east  side  of  the 
Murad  gorge,  which  is  about  2  miles  in  width  at 
the  summit  of  the  cliffs.  The  slopes  are  much  obs- 
cured by  talus  and  vegetation,  and  unfortunately 
circumstances  and  failing  light  prevented  me  from 
stopping  to  make  a  detailed  examination  of  this 
interesting  locality.  Near  the  summit  of  the  cliff, 
however,  I  was  able  to  pick  up  a  few  fossils,  which 

tmaU  oyBtali  of  ibon  mms  are  also  present  The  gsoundiiass 
is  a  dose  felt  of  plagioclase  U^hs  (andesine  abont  Ab4An3)  in  fluidal 
anangement,  widi  fairly  numerous  augite  granulesi  and  abundant 
skeletal  needles  of  magnetite,  in  a  pale  brown  or  colourless  glass, 
densely  crowded  with  belonites  and  dust 


194  Geology  of  Armenia 

Mr.  R.  Bullen  Newton,  F.G.S.y  has  kindly  examine 
and  identified  as  follows : — 

"  MeUmgena  cf.  Lainei^  Basterot. 

PtobaUy  rebted  to  If.  Laind^  Basterot  (Deacr.  g6oL 
du  bassm  tert  da  sad-ouest  de  la  Fraiux,  Mtm* 
Sac,  tPkisi.  mat.f  Paris,  1895,  vol.  iL,  part  ly  pi.  vii, 
fig,  8,  a,  b),  torn  the  Aquitanian  of  St.  Avit  and 
Dax,  showing  spiral  sculpture,  but  without  evidence 
of  the  maiginal  spines. 

Lampusia^  sp.  indet. 

Two  whorls  in  matrix  of  Law^^usia  (=2W!flMi),  show- 
ing noduhitions  and  spiral  Unes;  associated  with 
remains  of  DtrrUdla  and  DmtaUtim^ 

Turritellaf  sp.  indet.,  (fragments). 
Dentalium^  sp.  indet.,  (fragments). 
Divaricardium  pe£tinatum^  Linnaeus. 

==Csn6iMi  aqmtamcum^  Mayer,  Joum.  ConchyL, 
Paris,  1858,  vol.  vii,  p.  89,  pL  4,  fig.  9,  from  the 
Aquitanian  of  St  Avit.  The  species  also  occurs  in 
the  Tortonian  of  the  ^enna  basin,  according  to 
Homes,  .AbhandL  k.-L  geol.  Reichsanst.,  1861, 
(F08S.  Moll.  Tert.-Beck.  Wien).  The  spedmen 
agrees  with  the  Vienna  form  in  sise  and  soilpture, 
but  differs  in  being  rather  more  oblique. 

Acesta  cf.  Miocenica^  Sismonda. 

A  laige,  Inoceramoid-looking  cast,  with  both  valves 
closed,  showing  radial  and  concentric  sculpture. 
An  oblique  furrow  extends  iJx)ut  35  mm.  across  the 
valves  from  the  umbonal  area;  umbones  not 
preserved.  The  byssal  character  is  situated  in  a 
depressed  and  elongated  area  of  the  anterior  r^oo 
of  the  shelL  Heightssi as  mm.,  lengths:87  mmo 
convexity=3S  mm. 

Sfxmdylus^  sp.  indet. 

Teredina^  sp.  indet. 

Portion  of  a  large  tube  in  matrix  belonging  to  dib 
genus. 

Ofbitoides  (Lepidocydina). 
AmphhUgina. 


-Sr 


lu-tntfTu/*^ , 


Khamur  to  the  Bingdl  Cliffs  195 


Operculina. 
MilioUnaJ* 

These  fossils,  with  the  exception  of  the  fragmentary 
TuYvitella^  Dentaliumy  and  Teredina,  occurred  only  as 
casts.  My  slide  of  this  arenaceous  limestone  revealed, 
under  the  microscope,  the  presence  of  a  CellepoYa^  in 
addition  to  the  above-mentioned  Foraminifera,  but 
there  is  no  trace  of  Lithothamnion  or  Ceriopara^  as  in 
the  Miocene  limestone  of  the  same  zone,  57  miles 
W.S.W.  of  this  locality  (p.  97). 

Although  some  of  these  fossils  are  Aquitanian 
forms,  yet  the  Divaricardium  at  any  rate  is  known 
to  persist  into  the  Tortonian,  and  it  is  probable  that 
these  beds  are  of  that  age.  Indeed  they  clearly  mark 
a  slightly  higher  horizon  than  the  Thamnaroea  poly- 
morpha  beds  (Upper  Helvetian  and  Lower  Tortonian, 
see  Chapter  XXIII),  for  at  the  summit  of  the  cliff 
I  foimd  lying  in  the  limestone  (from  which  it  had 
weathered  out)  a  slightly  rolled,  derived  specimen  of 

Thamnaroea  polymorpha^  Abich, 

with  burrows  of  Lithodomus  lithophagus^  Linnaeus.  In 
this  bed  the  foraminiferal  limestone  (No.  160)  also 
contained  fairly  numerous  pebbles  of  diabase,  a  green 
compadl  grit,  etc.,  indicating  the  close  proximity 
of  land  at  the  period  of  deposition.  The  littoral 
charadter  of  these  Miocene  strata  on  the  south  side 
of  the  Armenian  plateau  forms  an  interesting  parallel 
to  the  beach-deposits  of  similar  age,  which  have  been 
described  by  Aoich*  on  the  northern  border  of  the 
plateau-region,  e.g.^  at  Sarydash  and  Ashkala. 

Although  land  probably  existed  then  on  the  site 
of  the  present  Taurus  (about  20  miles  to  the  south), 

^  Gecfi.  Forsck.  in  den  kattk.  ZSndem^  ii.  xax,  axx-saj. 


196  Geology  of  Armenia 

yet  it  is  evident  that  the  mica-schists  and  marbles, 
which  now  form  the  axis  of  that  range  (p,  109),  were 
not  exposed  in  the  Miocene  period,  for  the  pebbles 
in  the  limestone  show  no  trace  of  such  rocks,  and 
even  mica-flakes  do  not  occur. 

This  discovery  of  Miocene  fossils  corroborates  and 
amplifies  observations  made  by  Koch  at  points  about 
16  miles  to  S.W.  and  S.E.  respectively  of  this  place. 
Firstly  he  states*  that  immediately  beneath  Surb 
Karapet   the    zigzag  path   towards   Mush   traverses 

'^a  limestone,  almost  dazzling  in  whiteness,  containing  PeniacrinHes 
and  Encrinites^  besides  Foraminifera  and  Nummulites^  while  in  other 
places  Ostracites  and  other  shells  were  intermingled,  and  in  a  third 
locality  actual  coral-reefs  occurred  of  a  kind  of  Tuhifora,  A  fine- 
grained molasse-sandstone  was  interbedded.  Trachyte  and  basalt  had 
broken  through  these  strata  in  places." 

Secondly,  to  the  S.E.  of  our  standpoint,  between 
Zronk  and  the  Cha  Dagh  (on  the  way  from  Mush 
to  Liz),  Koch  t  has  also  recorded  the  occurrence  of 

*'  shaly  marls,  alternating  with  a  very  fine-grained  sandstone  or  with 
a  whitish  limestone,  which  contained  fossil  shells  and  Encrinites  with 
Tubipora,'' 

It  is  not  improbable  that  Koch,  in  his  rather  vague 
determinations,  mistook  the  Orbitoides  of  this  lime- 
stone for  Nummulites.  Since  I  noticed  that  the 
thick,  uniform  Miocene  strata  extend  for  miles  both  to 
east  and  west  of  the  Murad  gorge,  it  can  hardly  be 
doubted  that  Koch's  observations  refer  to  beck  of 
the  same  series,  although  the  Crinoidal  remains  (if 
correAly  determined)  would  indicate  a  horizon  as  yet 
unknown  in  the  Miocene  of  Armenia  (see  Chapter 

The  lofty  Bingol  cliffs  (about  4000  feet  high  from 

*  Rme  impofUitchm  Gebirgi^  iL  393. 
t  INd.f  417. 


Khamur  to  the  Bingdl  Cliffs  197 

base  to  summit)  run  parallel  to  the  axis  of  the  plain 
of  Mush,  viz.,  W.N.W.-E.S.E.  The  line  of  cliffs 
could  be  well  seen  in  their  entirety  from  our  stand- 
point above  the  Murad  gorge,  and  may,  I  think,  be 
regarded  as  a  gigantic  fault-scarp.  The  lower  ground 
of  Giimgiim  and  Gundemir,  lying  between  the  Bingol 
cliffs  and  the  Surb  Karapet  plateau,  would  in  this 
case  be  an  area  of  depression  like  the  plain  of  Mush. 

On  August  15th  we  left  Charbahur  for  Giimgiim, 
crossing  a  broad  spur  from  the  Khamur  line  of 
heights.  On  looking  across  the  Bingol  Su  valley 
to  the  northern  slopes  of  the  Surb  Karapet  plateau 
I  could  see  that  its  buff,  Miocene  limestones  at  first 
dip  N.N.W.;  further  west  other  ridges  of  gentle 
contour  come  to  view  successively  in  Echelon,  showing 
a  dip  of  30°  S.S.E.  Later  on  in  the  afternoon,  whilst 
riding  from  Alagoz  to  Dodan,  a  still  more  westerly 
portion  of  the  limestone  heights  on  the  south  side 
of  the  Bingdl  Su  was  seen  to  dip  25®  N.N.W.,  sloping 
gently  down  to  the  river. 

So  far  as  the  distance  would  allow  me  to  judge, 
there  appeared  to  be  some  white  strata  (perhaps  a 
continuation  of  the  recent  lake-deposits)  flanking 
the  Miocene  beds.  Our  course  to  Giimgiim  lay  over 
a  block  of  grey  clays  and  marls,  with  interbedded 
sheets  of  basalt — sl  continuation  of  the  lacustrine 
series,  which  I  had  been  observing  all  the  way 
down  the  Murad  valley  from  the  Bulanik  plain.  The 
whole  series  dips  here  about  25®  N.N.W.,  so  far  as 
I  could  estimate,  but  the  clays  are  very  liable  to  slip, 
owing  to  the  springs  they  throw  out,  and  no  clear 
se(5lion  presented  itself.  No  fossils  were  anywhere 
visible. 

Before  reaching  the  summit  of  the  broad  downs 
we  passed  near  the  Kurdish  village  Kara  Seid, 
situated  on  a  terrace  of  marly  clay,  which  has  been 


198  Geology  of  Armenia 

baked  into  hard,  porcellanous  shale  by  an  overlying 
sheet  of  basalt.  Still  higher  up,  a  flow  of  brown, 
dndery  lava,  about  50  feet  thick,  is  also  intercalated 
in  the  series  of  clays.  A  similar  flow  occurred  in 
the  cliffe  north  of  Charbahur  (p.  192) ;  it  has  a  con- 
siderable resemblance  to  the  streaky,  cindery  lava 
of  Khinis  (p.  65). 

The  adlual  summit  of  this  broad  spur  from  the  low 
line  of  heights  linking  Khamur  to  the  Bingol  Dagh  is 
composed  of  a  black  olivine- basalt  (No.  161),*  some- 
times showing  a  ropy  appearance,  and  inclined  slight- 
ly to  N.W.,  like  the  rest  of  the  series.  From  this 
standpoint  our  view  extended  on  the  one  hand  to 
the  high  Bingol  cliffs,  and  up  the  deep  Gumgiim  or 
Goshkar  gorge  into  the  very  heart  of  the  Bingol 
massifj  the  western  summit  rising  up  behind  the 
sharp  Goshkar  Baba  cone  (see  Chapter  XIII).  On  the 
other  hand,  we  could  see  right  down  the  Murad  gorge 
into  the  plain  of  Mush.  The  dip  of  the  strata 
immediately  on  the  north  side  of  this  plain  appeared 
(in  this  clear  air)  to  dip  about  35^  N.N.W.  A 
summary  of  my  observations  on  the  dip  of  these 
Miocene  beds  (see  pp.  193,  197)  would  indicate  two 

*  The  rock  is  very  compact,  homogeneous,  and  minutely  porous ; 
small,  pale  yellow  olivines  (up  to  i  mm.)  are  the  only  constituents 
visible  with  the  lens.  The  rock  is  slightly  magnetic  (a*").  Under 
the  microscope,  olivine  is  the  only  phenocryst,  down  to  quite  small 
dimensions ;  quite  fresh,  but  some  crystals  show  a  marked  penetration 
by  the  groundmass.  There  are  no  actual  phenocrysts  of  felspar, 
but  the  larger  laths  of  the  groundmass  show  extinction-angles  pointing 
to  LABRADORiTE  (about  Ab5An6).  MAGNsrrrE  is  accessory.  The 
OKOUNDMASS  is  a  felt  of  pla^odase  laths  (andesine  about  Ab4An3), 
with  small  olivines,  and  smaller  augite  granules  in  an  umber-brown 
GLASS,  showing  cumulites  here  and  there,  and  local  opacity  round 
the  numerous  steam-pores,  or  bordering  the  felspar  laths.  Thost 
(Abhandl.  Senckenb.  Naturf,  Ga.  Frankfurt  a/M.,  XVIII,  ii.  254)  has 
described  an  almost  precisely  similar  basalt  from  the  Shipker  Gdl, 
near  Gerjiisy,  on  the  Central  Karabagh  plateau. 


vi'^i^Aii-;-^^ 


I 


Khamur  to  the  Bingtil  Cliffs  199 

gentle  folds  in  the  Surb  Karapet  block,  from  north 
to  south,  as  follows : — 

25®  N.N.W.-30**  S.S.E.-25^  N.N.W.-30*'  S.S.E.-35°  N.N.W. 

That  merely  two  symmetrical  folds  should  have 
been  formed  in  a  distance  of  10  miles  is  an  indication 
that  the  tangential  pressure  at  the  period  of  folding 
(mid-Miocene)  was  far  from  being  acute.  All  my 
previous  observations  on  the  Miocene  beds  of  the 
Armenian  plateau  show  that  this  folding  preserves 
its  gentle  charadler  for  a  considerable  distance  further 
north. 

From  the  same  position  it  was  evident  that  whilst 
the  Giimgiim  plain  is  sharply  bounded  on  the  north 
by  the  great  Bingol  cliffs,  there  is  on  the  east  only 
an  irregular  mass.^  of  low  heights,  linking  Khamur 
to  the  Bingdl  Ddgh.  On  descending  a  broad  valley 
leading  to  Giimgiim,  one  of  these  spurs  on  the 
north  side  of  the  stream  showed,  at  a  distance  of 
about  a  mile,  an  interesting  cliff-secftion  of  grey  clays 
(probably  lacustrine)  showing  remarkable  contortions 
caused  by  an  overlying  and  contemporary  sheet  of 
basaltic  lava.  Here  again  the  series  dips  gently  to 
the  N.N.W. 

After  a  short  halt  at  Giimgiim  (4800  feet ;  6|  miles 
from  Charbahur)  we  continued  our  march,  fording  the 
Giimgiim  (or  Gc^hkar)  river  at  Alagoz.  On  the  south 
side  of  the  river,  the  ridge,  which  we  had  crossed 
in  riding  from  Charbahur,  breaks  off  in  a  low  cliff. 
The  river  flows  at  the  base  of  this  cliff,  preventing 
access  to  it,  but  it  appeared  to  consist  of  limestone, 
dipping  40°  N.W.  by  W.,  weathering  a  rusty  brown 
in  places,  and  capped  by  basaltic  lava.  Judging  from 
the  blocks  of  building-stone  in  the  village  of  Alagoz, 
the  rock  is  a  travertinous  limestone,  very  similar 
to  the  limestone  which  I  had  found  at  Demian  (p.  95) 


200  Geology  of  Armenia 

to  form  part  of  the  lacustrine  series  with  Dreissensia 
polymorpha.  Proceeding  southwards  from  Alagoz,  vire 
traversed  the  outcrop  St  this  travertinous,  unfossili- 
f erous  limestone  on  the  west  side  of  the  river,  where  it 
is  also  overlain  by  a  scoriaceous,  basaltic  augite- 
andesite,  weathering  to  a  brick-red. 

We  soon  changed  o^r  course  to  WJ^.W.,  leavings 
this  limestone,  and  rising  over  overlying  dark  grey 
clay,    dipping  N.N.W.,   with  interbeddc^  sheets    ^ 
lava.    In  these  softer  beds  a  broad,  upland  valley 
has   been  formed  between  the  parallel,  N.W.-S*£., 
lava  ridges  of  the  Kamiishlii  Dagh  (Reed  Mountain). 
This  valley  has  evidently  been  the  site  of  a  sheet 
of  water;    indeed,  before  we  left  its  N.W.  end  we 
passed  two  marshy  remnants  of  the  lake,  full  of  reeds. 
The  rounded  lava  ridges  are  flanked  by  steep  screes, 
and  present  a  very  similar  appearance  to  the  lava 
heights  lying  between  Khamur  and  the  Murad.    They 
are  probably  the  remnants — isolated  by  denudation — 
of  a  thick  lava  sheet,  capping  the  lacustrine  deposits. 

This  valley  eventually  opens  at  Dodan  into  the 
ova  of  Gundemir  (lilanlu  according  to  Koch).  Here 
we  found  ourselves  m  a  level  plain,  in  which  Gun- 
demir (5025  feet ;  9^  miles  from  Giimgum)  occupies  a 
nearly  central  position.  The  view  to  the  west  is 
limited  by  a  broad  lava  slope  (part  of  the  Serokh 
Dagh),  into  which  the  Sherefeddin  river  has  cut  its 
deep  ravine,  with  Baskan  situated  at  its  entry  into 
the  plain  of  Gundemir,  A  similar,  gentle  lava  slope 
bounds  the  plain  on  the  souths  rising  at  an  angle 
of  lo'^  from  Diadin  up  to  the  Surb  Karapet  plateau 
(Kyel  Dagh  according  to  Koch).  On  the  east  side 
a  low  terrace  of  grey  clay  (lake-deposits)  forms  a 
^selvage  to  the  lava  ndges  of  the  Kamiishlii  Dagh. 
On  the  north  the  colossal  Bingdl  clifiEs  dwarf  all 
the  neighbouring  features. 


Khamur  to  the  Bingbl  Cliffs  201 

The  village  of  Gundemir  is  built  an  a  small  hill 
of  lake-depositSb  The  western  $lope  of  the  hill  shows 
a  white  or  brownish  grey  pipeclay  (Nos.  162,  163X 
dipping  20^  N.N.W.,  stained  yellow  here  and  there 
along  the  bedding-planes  by  iron  oxide.  It  contains 
carbonized  remains  of  broadly  lanceolate  leaves  of 
a  monocotyledonous  plant  like  3,  reed,  1-2  inches 
m  breadth.  This  pipeclay  passes  downwards  into 
a  grey  shale,  with  narrow  seams  of  impure  lignite 
from  2-6  inches  thick,  composed  entirely  of  these 
leaves.    No  other  fossils  were  visible. 

From  Gundemir  we  proceeded  in  a  N.W.  direction 
to  the  Bingol  clifiEs.  The  Kamiishlii  ridges  soon  die 
out  to  the  N.W.,  and  a  lava  ridge  with  similar 
orientation  rises  up  on  the  west  side  of  the  Bingol  Si;. 
We  skirted  the  eastern  slopes  of  this  high  ridge  of 
lava — a  fine-grained,  basaltic  augite-andesite.  Two 
low  hills  of  grey  marls  lay  on  our  right.  At  the 
plane  of  contacft  with  the  overljdng  lava  the  lake- 
deposits  show  signs  of  baking. 

Near  ChagheUk  we  descended  into  a  small  ova^ 
whence  we  obtained  an  uninterrupted  view  of  the 
Bingol  cliffs.  From  this  oua  up  to  the  gorge  of  the 
Bingol  Su,  which  notches  the  cliff,  there  is  first  of 
all  a  gentle  rise  over  a  broad,  alluvial  fan ;  then 
comes  broken  ground  of  grassy  hillocks,  lying  at 
the  foot  of  the  clife,  which  rise  over  3000  feet  above 
our  standpoint*  They  are  composed  of  horizontal 
layers  of  lava  and  tuff— the  lower  half  being  grassy 
and  sandy  (perhaps  including  lake-deposits]h-the 
upper  half  rocky  and  precipitous,  with  waterfalls 
emerging  at  about  200  feet  below  the  edge. 

After  crossing  the  level  (jva  of  Chaghelik  we  rose  up 
gradually — still  in  a  N.W.  direction — over  low  hills  of 
grey  marls;  soon,  however,  all  sign  of  sedimentary 
strata  disappeared,  and  the  remainder  of  our  ascent 


202  Geology  of  Armenia 

lay  entirely  over  volcanic  rocks.  The  lava  of  the 
lower  slopes  is  a  rhyolitic  homblende-andesite  (No. 
164),*  and  is  very  prone  to  decompose  into  brightly 
coloured  sands,  crimson,  reddish  brown,  yellow,  and 
white.  It  was  not  possible  to  obtain  a  specimen 
of  imaltered  rock;  it  evidently  forms  part  of  an 
earlier  series  of  rather  acid  andesites  in  the  volcanic 
succession  of  the  Bingol  Dagh. 

About  i^  mile  from  the  adhial  cliffs  we  reached 
a  broad,  sloping  terrace  of  a  dark  brown,  cindery 
lava,  with  scoriaceous  streaks,  in  flows  inclined  20*^  S. 
The  terrace  is  apparently  a  great  block  which  has 
slipped  down  a  few  hundred  feet,  for  I  could  see  a 
very  similar  sheet  of  lava  in  situ  in  the  cliffs  above. 
Moreover,  a  little  further  on,  as  we  continued  our 
slanting  ascent,  we  again  passed  over  a  dark  crimson 
sand,  due  to  the  decomposition  of  the  same  rhyolitic 
andesite  which  underlay  the  lava  of  the  terrace. 

*  The  rock  shows  a  arimson,  homogeneous  matrix,  easily  scratched, 
with  yellow  stains  of  iron  oxide,  and  contains  abundant,  white, 
striated  felspars  (up  to  7  mm.),  and  fairly  numerous,  lustrous,  black 
specks  (hornblende).  Under  the  microscope,  the  plagioclasb  (ougo- 
CLASE-AND£siNB  about  AbsAnj)  is  present  in  abundant  idiomorphs, 
occupying  more  than  half  the  slide;  mostly  of  tabular  habit;  the 
albite  twinning  is  either  fine  and  close,  or  faint  and  with  a  washed-out 
appearance;  much  cracked,  the  cracks  sometimes  marked  out  by 
limonite ;  occasionally  enclosing  hornblende  and  granules  of  epidote. 
Hornblende  occurs  chiefly  in  sparing,  pale  brown  fragments,  showing 
no  resorbtion ;  pleochroic  to  a  dull  brownish  green.  A  few  pseudo- 
morphs  after  auoite  are  present,  consisting  of  a  pale  yellowish  green, 
serpentinous  mineral,  stained  with  iron  oxide,  and  with  limonite 
and  opal  in  its  meshes.  The  groundmass  is  cryptocrystalline  and 
devitrified ;  it  is  extremely  streaky  and  rhyolitic — ^the  stresses  bending 
round  the  felspars  in  wavy  contortions.  The  opacity  of  the  darker 
streaks,  and  the  general  colour  of  the  rock  is  due  to  hematitic  dust. 
Where  the  streaks  sweep  apart  at  a  bend  the  intervening  space  is 
clear  and  more  crystalline,  polarizing  in  irr^ular,  ill-defined  blotches, 
and  occasionally  showing  spherulites.  Fiequent,  tiny,  orange  specb 
may  possibly  represent  weathered  partides  of  augite. 


Khamur  to  the  Bingdl  Cliffs  203 

Close  to  the  summit  of  the  cliffs  the  rock  is  a  dark 
greenish  grey  olivine-basalt  (No.  165).*  This  rock 
forms  the  surface  of  the  lofty,  volcanic  plateau,  on 
which  we  pitched  our  camp,  at  a  height  of  8500 
feet  above  the  sea. 

'^  The  rock  is  compact,  homogeneous,  and  somewhat  fissile. 
Scattered  striated  felspars  (up  to  5  mm.),  and  still  fewer  olivines 
^up  to  I  mm.)  are  visible.  Under  the  microscope,  the  plaoioclase 
(lABRADORiTE  about  Ab5An6)  is  present  in  idiomorphs  mostly  of 
prismatic  habit;  sometimes  faintly  zoned;  often  with  numerous, 
elongated  glass-inclusions  (brown  to  opaque),  as  well  as  abundant 
granules  of  augite  and  magnetite,  ranged  in  parallel  Unes.  Only 
a  few,  large  phenocrysts  of  psJe  greenish  olivine  are  present ;  smaU 
aystals,  however,  are  fairly  abundant;  sometimes  with  an  irregular 
wreath  of  augite  and  magnetite.  Auqite  is  hardly  present  as  a 
phenoayst,  for  it  occurs  in  rather  small  dimensions;  pale  brown, 
almost  colourless;  one  of  the  little  prisms  shows  an  approach  to 
the  hourglass  form.  The  groukdmass  is  a  felt  of  plagiodase  laths 
(labradorite  about  AbiAni),  showing  fluidal  arrangement,  with 
abundant  augite  in  granules  and  tiny  prisms,  and  numerous  magnetite 
granules. 


CHAPTER  Xni 

THB  BINGdL  VOLCANO 

I.    Ths  Southern  Region. 

The  preceding  chapter  has  dealt  with  our  approadi 
to  the  S.W.  portion  of  the  Bingol  plateau,  which 
we  reached  by  an  oblique  ascent  up  the  great  cli£Es. 
Here  they  approximate  to  the  ShereSeddin  Dagh,  and 
finally  meet  it,  forming  a  great  bastion  called  the 
Kantar  Dagh. 

Our  first  camp  on  the  plateau  lay  on  the  grassy 
banks  of  a  small  stream,  flowing  westward  between 
low,  rounded  slopes  of  a  grey,  basaltic  lava,  which 
readily  decomposes  to  a  brown,  clayey  sand.  From 
this  position  (8500  feet ;  3500  feet  above  Gimdemir)  I 
could  see  that  the  country  west  of  the  Bingdl  plateau 
is  a  tumbled  mass  of  peaks,  ridges,  and  broken 
heights,  beloaging  to  the  Javresh  Dagh  and  Shaitan 
Dagh.  These  mountains  form  the  high  watershed 
between  the  Lichig-Khindris  valley  on  the  north,  and 
the  valleys  of  the  Gunek  Su  and  Kighi  Su  on  the 
south ;  they  lie  on  the  same  line  of  strike  as  the 
Ak  Dagh  and  Kazbel  Gedik  on  the  east  side  of  the 
Bingol  Dagh,  and  it  is  not  improbable  that  they 
consist  of  the  same  ancient  marbles  and  schists,  with 
overl}ang  Miocene  limestone  and  intrusive  serpentines 


The  Bingdl  Volcano  205 

(pp.  61,  62,  73).  The  field-glass,  however,  only 
revealed  thick  beds  of  white  and  buff  strata,  with 
bright  red  patches  on  the  lower  slopes,  perhaps 
indicating  decomposed  igneous  rocks.  Turning  to 
the  S.W.,  I  could  see  tlmt  the  Gunek  Su  has  cut 
through  the  Surb  Karapet  plateau,  which  breaks 
off  in  high  river-cliffs.  This  zone  of  Miocene  lime- 
Stone  is  continued  on  the  west  side  of  the  river,  where 
the  dip  appeared  to  be  30°  S.S.E. 

The  S.W.  part  of  the  Bingol  plateau  is  composed  of 
broad  lava-streams  of  low  curvature,  and  with  a 
general  direction  to  the  S.W.  The  lava  was  clearly 
able  to  travel  long  distances  before  solidifying.  The 
older  lava  is  the  black  olivine-basalt  already  described 
(p.  203) ;  it  has  only  been  partially  concealed  by  a 
grey,  basaltic  augite-andesite,  with  large,  zoned 
telspars  and  distindl  augites.  The  low  ridges  of 
porous  lava  are  dry  and  barren,  but  the  intervening 
hollows  of  sandy  clay  (the  decomposition-produdt  of 
the  lava)  form  little,  ^een  pastures,  studded  with 
pools  and  traversed  by  nvulets. 

After  a  ride  of  ij  hour  from  our  camp  over  the 
plateau  to  the  N.E.  we  came  to  a  N.-S.  ridge  of  lava, 
capped  by  a  low  cone.  On  reaching  this  standpoint, 
we  looked  down  on  a  nearly  level  depression,  open 
both  to  the  north  and  south,  and  about  a  mile  in 
width.  It  is  probably  of  the  nature  of  a  gja  (cp. 
p.  150);  and  here  the  headwaters  of  the  southern 
Bingol  Su  oollaft  (see  plan).  Indeed,  on  looking 
southward,  I  could  see  the  commencement  of  its  deep 
^orge,  which  was  so  conspicuous  a  feature  in  the 
Jingol  cliffs,  as  viewed  from  Chaghelik  (p.  201). 
Lava  has  clearly  flowed  down  into  this  depression 
from  several  points  alQng  the  high  rampart  which 
bounds  it  on  the  east.  On  riding  up  the  rather  ste^, 
grassy  slope  of  this  lampart  we  found  the  grey, 


2o6  Geology  of  Armenia 

andesite  to  become  fissile  and  slaty. 

From  the  summit  of  the  ridge  the  whole  of  the 
southern  region  of  the  Bin^oi  plateau  lies  open  to 
view.  It  has  been  deeply  dissedled  by  the  Goshkar 
Su,  and  its  tributaries,  into  winding  ravines,  which 
imite  to  form  the  great  canyon  north  of  Giimgiim. 
This  region  is  bounded  on  the  north  by  the  transverse 
rampart,  4^  miles  long,  between  the  eastern  and 
western  summits  of  Bingol — on  the  east  by  the 
continuation^  of  this  rampart,  which  bends  roimd  to 
the  S.E. — on  the  west  by  the  rampart  (on  which 
we  were  standing),  which  runs  to  the  S.W.  from 
the  western  summit — and  the  view  is  framed  in  on 
the  south  by  some  rounded  peaks  near  the  edge  of  the 
great  cliffs  (see  plan).  These  ramparts,  which  (with 
their  northerly  prolongations,  vide  infra)  form  an  H, 
evidently  mark  the  original  fissures  of  eruption.  A 
long  time  would  have  to  be  devoted  to  thoroughly 
explore  the  splendid,  natural  sections  of  the  numerous 
ravines,  and  I  was  only  able  to  examine  a  small 
portion  of  this  great  cauldron  of  erosion. 

We  proceeded  down  the  eastern  slope  of  the  S.W. 
rampart  to  the  marshy  platform  lying  at  the  southern 
base  of  the  transverse  rampart,  which  rises  about 
1000  feet  above  it.  Here  numerous  peatv  tarns  lie 
scattered,  and  form  the  headwaters  of  the  Cxoshkar  Su. 

Our  second  camp  (91 61  feet)  was  situated  on  a  lava 
terrace,  to  the  south  of  this  marshy  tradl,  and  at  a 
slightly  higher  level.  About  half  a  mile  S.S.E.  of  our 
camp,  the  rivulet  draining  the  tarns  suddenly  plimges 
down  into  a  deep,  precipitous  chasm,  revealing 
basaltic  lava-streams  in  section.  The  rock  near  the 
surface  is  pale  grey  and  fissile ;  lower  down  it  shows 
spheroidal  jointing,  with  concentric  coats,  which  peel 
off  in  weathering ;  still  deeper  down  it  is  black  and 
columnar — a  doleritic  olivine-basalt,  similar  to  the 


The  BingSi  Volcano  207 

lava  (No.  168,  p.  aog)  of  the  spur  separating  the  low 
ground  near  our  camp  from  the  eastern,  more  fertile 
vatloj  in  which  the  Kurdish  chieftain  Mahmud  Bey 
had  pitched  his  tents.  A  similar  rock  (No.  169)  also 
occurs  just  south  of  the  western  summit,  and  is 
probably  part  of  the  same  lava-sheet. 

About  500  yards  S.W.  of  our  camp  the  lava-terrace 
again  breaks  off  suddenly.  The  precipitous  slope 
01  the  ravine  displays  a  yellowish  grey  tuff  (No.  i66),* 
underlying  this  lava  (cp.  p.  209).  The  tuff  is  well 
bedded,  dipping  20*^  E.S.E.,  and  is  full  of  angular 
fra^jments  of  black,  scoriaceous  lava,  up  to  2  feet 
in  diameter;  the  layers  are  frequently  bent  down 
beneath  the  larger  fragments.  This  tuff  is  very  basic 
in  characfter.  The  dip  is  inclined  away  from  the 
conspicuous  cone  of  Goshkar  Baba  (p.  iq8),  where 
it  is  capped  by  a  glassj*  mica-andesite  (No.  167)! 

*  The  rock  contains  abundant,  angular,  black,  porous  lapilli  (up  to 
6  mm.)  in  a  friable,  earthy,  pale  yellowish  grey,  non-calcareous 
matrix.  A  few  olivines  (less  than  a  mm.)  and  tiny  fragments  of 
felspar  are  the  only  minerals  discernible  with  the  lens.  Under  the 
microscope,  the  groundmass  consists  of  glassy  dust  and  abundant, 
angular  fragments  and  splinters  of  ytilow-brown  or  colourless  glass. 
This  GLAB:i  is  often  highly  vesicular:  the  small  bubbles  are  usually 
ranged  in  parallel  lines,  giving  a  streaky  appearance ;  it  is  frequently 
crowded  with  felspar  microlites  (oligodase),  with  jagged  ends,  in 
fluidal  arrangement ;  it  encloses  phenocrysts  of  olivine  and  plagioclase, 
but  no  crystallites  or  spherulites  are  present.  The  olivine  idiomorphs 
are  frequent  both  in  the  groundmass  and  in  the  glass ;  sometimes 
in  aggregates ;  quite  fresh,  but  in  one  case  deeplv  penetrated  by  glass. 
The  PLAGIOCLASE  (labradoute  about  Ab5An6)  is  only  idiomorphic 
when  enclosed  in  the  glass ;  otherwise  in  fnpneDts ;  quite  dear 
and  fresh;  sometimes  delicately  zoned;  twinning  occasionally  very 
fine  and  close ;  when  inclusions  occur,  they  are  central  and  consist  of 
yellow-brown  glass.  An  occasional,  tiny  flake  of  biotitb,  and  a  few, 
small  fragments  of  pale  green  AUorrB  occur  in  the  groundmass,  but 
not  in  the  glass. 

t  The  rock  is  compact ;  pale  bluish  grey,  somewhat  stained  by  iron 
oxide,  with  white,  sinuous  streaks;  rather  fissile  along  the  lines  of 


I 


208  Geology  of  Armmna 

— ^the  mere  remnant  of  a  former  lava-flow,  which  does 
not  seem  to  have  extended  further  east. 

A  somewhat  similar  cone  of  erosion  lies  on  the 
west  side  of  the  stream,  which  separates  it  from  the 
Goshkar  Baba  cone.  I  found  it  to  be  composed  of 
the  same  grey,  sandy  tuff,  capped  by  mica-andesite. 
A  small  waterfall  occurs  just  between  the  two  cones, 
where  the  stream  is  engaged  in  cutting  back  a  hard 
layer  in  the  tuff.  Proceeding  southwards,  we  crossed 
the  western  tributary  of  the  Goshkar  Su,  and  climbed 
up  its  steep  bank  of  grey  tuff  to  a  rounded  hill^ 
composed  of  the  same  substance;  here  its  layers 
dip  away  on  both  sides  of  the  hiU  at  about  20  to 
N.N.E.  and  S.S.W.  Still  higher  up  (to  S.W.),  the  tuff 
is  again  capped  by  the  mica-andesite,  which  doubtless 
proceeded  from  a  point  a  little  further  to  the  S.W., 

flow;  it  contains  abundant,  striated  fdspacs  (up  to  3  min.X  and 
numerous  flakes  and  tablets  of  biotite  (up  to  a  nun.).  The  rock  ii 
distinctly  magnetic  (10").  Under  the  microscope,  the  plaoioclasi 
(ANDBsms  al^ut  Ab4An3)  is  present  in  large,  tabular  idiomorpbs, 
sometimes  in  groups;  often  cracked,  and  occasionally  honeycombed 
with  a  network  of  groundmass,  or  endoong  granules  of  biotite,  augite, 
hornblende  and  magnetite.  BiOTiis  idiomorohs  are  fairly  abundiant ; 
the  larger  individuals  show  only  slight  resorbdon,  while  the  smaller 
flakes  are  nearly  opaque,  with  a  very  narrow  border  of  secondary 
augite  granules ;  some  flakes  are  bent  and  fractured ;  encloses  small 
apatites,  and  in  the  basal  sections  sagenite  w^  of  fine  rutile-needles 
(intersecting  at  60^)  are  visible.  HoRNBLtNDS  is  very  spaxin^y 
present  in  pale  brown  idiomorphs,  showing  only  slight  resorbtioo; 
plepchroism :  a — pale  straw-brown,  d  &  ^— deep  brown,  c=d>^ 
Aucm  in  pale  green  idiomorphs  is  less  frequent  than  the  biotite; 
quite  fresh ;  sometimes  showing  twinning,  both  simple  and  lamellar ; 
encloses  magnetite,  and^more  rudy  biotite.  ICAOimm  and  ai-axits 
are  accessory.  The  caouNDMAsa  is  markedly  strealnr  with  dusky,  pale 
brown  streaks ;  an  axiolitic  structure  is  visible  under  crossed  nicoh, 
the  wavy  axioUtes  being  often  separated  from  each  other  by  a  micro> 
felt  of  densely  crowded  felspar  ladis  (oligodase),  with  abimdant  augite 
granules,  sometimes  stained  brown  by  iron  oxide,  and  even  more 
numerous  granules  of  ma^etite.  Some  colourless  g^s  is  visible  in 
tile  ditnner  parts  of  the  slide. 


The  Bingdl  Vdcano  209 

perhaps  from  some  of  the  peaks  which  are  ranged 
along  the  edge  of  the  great  Btngol  clilk. 

The  grey  tuff  extends  also  to  the  N.£.  of  the 
Goshkar  Baba  cone,  for  I  met  with  it  again  cm  the 
eastern  slopes  of  the  low  spur  separating  our  camp 
from  Mahmud  Beys  ycUla.  Here  it  underlies  a 
doleritic  olivine- basalt  (No.  168),*  similar  to  the  rock 
in  the  gorge  S.E.  of  our  camp  (pp.  206,  2oy).  The 
gently  flowing  stream  through  this  yaila  has  carved 
its  shallow  course  out  of  the  tuff,  which  further  west 
dips  40®  W.S.W.  The  meadows  are  strewn  with^ 
boulders  of  black,  scoriaceous  lava^  which  have 
evidently  weathered  out  of  the  gre3rish  white,  sandy 
tuff. 

On  proceeding  southwards  to  a  small  peak  (x  on 
plan)  near  the  edge  of  the  Bingdl  cliffs,  we  crossed 
another  valley,  almost  dry  at  this  time  of  year 
(August),  and  then  rose  up  gradually  to  the  lava- 
plateau  on  the  east  (left)  side  of  the  Goshkar  gorge. 
A  gentle  slope  led  us  to  the  peak  itself,  v/hich  is 
composed  of  huge  boulders.  It  clearly  marks  the 
point  of  emission  of  the  lava — a.  black, .  basaltic 
pyroxene-homblende-andesite  (No.  170).!    No  steam- 

*  The  rock  is  very  compact,  and  greenish  black,  much  stained 
brown  by  ircm  osdde ;  with  irr^fular  fracture.  A  good  maiiy  olivines 
(up  to  s  mm.),  stained  brown,  numerous  striated  felspar^  (up  to  5 
mm.),  and  smaller  black  augites  are  discernible  with  the  lens.  Under 
the  microscope,  the  plagioolase  (labradobitb  about  Ab^An6)  is 
present  in  large  idiomoiphs,  mostly  of  prismatic  habit ;  qmte  fresh, 
and  often  beautifully  zoned;  occasionally  with  a  central  network 
of  opaque  inclusions.  Ouvikb  occurs  in  abundant  idiomorphs, 
sometimes  in  ^ups  of  6  or  &;  wholly  or  partly  altered  into  dul| 
green  serpentine,  which  is  often  stamed  orange  by  iron  oxide. 
AuGnK  is  present  in  large,  pale  brownish  purple  ophitic  plateSi 
enclosing  plagiodase  laths  (andesine  about  Ab4An3),  and  divines.  A 
good  many  scattered  grams  of  maohbtits  are  present  A  little 
serpentinous  matter  occurs  interstitially  between  felqpar  laths. 

t  The  rock  is  compacti  with  irregular  fracture,  aad  is  markedly 


2IO  Geology  of  Armenia 

cavities  were  visible  in  the  blocks,  and  the  lava- flow 
was  doubtless  of  great  fluidity.  When  seen  in  sedlion 
from  the  opposite  (west)  side  of  the  gorge,  this  super- 
ficial lava  appeared  to  be  horizontal,  and  about  60 
feet  thick,  forming  long,  vertical  columns.  It  is  not 
represented  on  the  west  side  of  the  Goshkar  gorge, 
but  it  has  flowed  for  considerable  distances  to  the 
N.E.,  £.,  and  S.E.,  meeting  the  great  S.E.  rampart  (p. 
206),  and  extending  beyond  it  (to  S.E.)  to  meet  the 
western  spurs  of  Khamur.  The  underlymg  lava-flows 
are  thinner,  with  an  inclination  of  about  7^  S.S.W.,  as 
seen  in  sedlion  in  the  gorge. 

II.    The  Summit-Region. 

The  great  transverse  and  central  rampart  of  the 
Bingol  Dagh  (p.  206)  rises  about  1000  feet  above 
the  marshy  terrace  at  its  southern  base.  On  this  side 
it  shows  a  smooth,  grassy  slope  of  not  more  than  26% 
gradually  curving  over  to  form  a  nearly  level  terrace, 

magnetic  (ao*"]).  Sparing,  striated  felspars  (up^  to  3  mm.),  and  less 
numerous  augites  (up  to  2  mm.)  can  be  distinguished.  Under  the 
microscope,  the  plaoioclasb  (labradoutb  about  AbiAnx)  idio- 
morphs  are  mosdy  of  prismatic  habit,  down  to  small  dimensions; 
occasionally  with  a  central  network  of  brown  glass ;  sometimes  faintly 
zoned.  Augitb  is  frequent  in  very  pale  green,  almost  colourless 
idiomorphsy  often  in  groups;  frequently  twinned;  quite  fresh,  and 
usually  free  from  inclusions,  excepting  magnetite  grains,  and  an 
occasional  fragment  of  hornblende.  Bronzite  is  less  frequent, 
oocuning  in  small,  pale  green,  almost  colourless  idiomorphs ;  slighUy 
pleochroic  (to  pinkish  brown).  Ho^lnblends  is  sparingly  present 
m  small,  brown  idiomorphs,  always  greatly  resorb^,  except  when 
protected  by  an  adjacent  felsparp  or  when  endcwed  in  aogite ; 
occasionally  invaded  by  groundniass ;  markedly  pleochroic :  a — ^pale 
brownish  yelloif,  d — dark  brown,  ^--dull  ydlow-brown  with  a  shght 
greenish  tinge.  Maqnetitx  occurs  in  scattered  grains.  The  ground- 
MASS  is  a  microfelt  of  plagiodase  microlites  (oligodase)  with  jagged 
ends,  in  a  base  of  pale  brown  or  colourless  glass,  densely  crowded 
with  belonites  and  tmy  magnetite  granules. 


The  Bingil  Volcano  211 

X50-200  yards  broad,  on  the  top  of  the  rampart.  A 
lava-flow  extending  from  the  eastern  summit  forms 
the  easiest  approach  from  the  south;  this  lava  is 
light  grey,  and  very  fissile  on  the  surface ;  scoriaceous 
portions  weather  to  a  brick-red ;  the  rock  is  of  the 
same  character  as  the  pyroxene-homblende-andesite 
of  the  eastern  summit  (vide  infra).  Scattered  blocks 
of  a  banded  pitchstone,  with  dull  lustre,  have  rolled 
down  from  the  ridge  a  little  west  of  the  eastern 
summit.  When  we  reached  this  part  of  the  rampart, 
an  obsidian  (No.  171)*  appeared  as  the  surface-rock, 
fra(5lured  by  the  frosts  into  small  pieces  not  exceeding 
an  inch  in  diameter. 

On  the  adhial  eastern  summit  (10,770  feet)  the  lava 
is  a  pyroxene-homblende-andesite  (No.  172),!  with  a 

*  The  rock  is  a  homogeneous,  bkck  glass,  with  die  usual  con- 
choidal  fracture,  and  with  occasional  iridescence.  Under  the  micro- 
scope, it  is  seen  to  be  a  very  pure,  sepia-brown  glass,  neither  banded 
nor  cracked.  One  fair-sized  crystal  of  sanidiks  is  present  in  the 
dide,  but  no  plagiodase,  although  the  rock  is  clearly  die  maiginal 
glass  of  the  andesite  Na  172.  Small  laths  also  of  sanidine,  often 
very  narrow,  lie  scattered  in  the  glass,  and  are  usually  gr^^arious ;  in 
some  casej  forked ;  occasionally  bent.  A  very  few,  small  idiomorphs 
of  pale  green  augitBi  and  still  fewer  grains  of  MAGNvrrrK  are  the  only 
otiier  crystalline  oonstitutnts.  The  felspar  needles  form  the  nuclei 
of  incipient  q;Aerulites,  consisting  of  very  dark  brown,  irr^;ular, 
cloudy  masses,  often  congregated  at  each  end  of  a  needle,  so  as  to 
present  a  dumb-bell  appearance ;  occasionally  showing  an  indistinct 
radiate  structure  under  crossed  nioolSi  with  an  ill-delink,  bhck  cross. 
Sometimes  faint»  parallel  striae,  normal  to  a  felspar  lathi  are  visible  in 
the  glass.  Numerous,  small  bubbles,  sometimes  drawn  out,  are 
present* 

t  The  rock  is  dark  grey  and  compact,  with  broadly  conchoidal 
fracture.  It  is  slightly  magnetic  (i^).  Sparing  striated  felspars  (up  to 
to  3  mm.),  and  still  fewer  smaller  augites  are  visible  with  the  lens  m  a 
slightly  streaky  matrix.  Under  the  microscope,  the  flagioclass 
(andesinb  about  Ab4An3)  is  present  in  tabular  idiomorphs ;  frequent- 
ly with  a  peripheral  line  of  dust;  occasionally  with  a  network  of 
groundinass,  and  sometimes  enclosing  augite  and  bronzite ;  sometimes 
soned.    Auorrs  occuts  in  pale  green,  almost  odourless  idiomorphs. 


212  Geology  of  Armenia 

smooth,  rather  varnish-like  surface;  it  breaks  up 
readily  into  thin  slabs  like  broken  tiles,  and  weathers 
to  a  pale  brown.  A  specimen  (No.  173),*  taken  froi? 
the  precipitous,  northern  side  of  the  rampart,  just  to 
N.E.  of  the  eastern  summit,  is  substantially  of  the 
same  nature,  and,  in  addition,  contains  tridymite. 

On  its  south  side  the  eastern  summit  is  a  mass 
of  enormous  boulders  of  grey  andesite;    from  some 

often  in  gioups ;  frequently  twinned;  sinki  to  imther  smmll  dimensiont| 
encloses  magnetite.  Bronzitb  is  fairly  abundant  in  almost  colourless 
idomorphs ;  smaller  than  the  augite  with  which  it  is  often  associated ; 
slightly  pleochroic  (pale  green  to  pale  brown).  Hornblkmdi  is 
sparingly  present  and  almost  completely  resorbed,  with  a  narrow 
selvage  of  secondary  augite  and  magnetite  granules.  MAONBrrrB  is 
accessory.  The  oboukdmass  is  dusty,  cryptocrystalline  and  streaky ; 
it  contains  abundant  laths  and  microUtes  of  plagiodase  (oligodase), 
often  with  jagped  ends,  small  augites,  and  magnetite  granules ;  there 
is  only  very  little,  interstitial  glass,  densely  crowded  with  belonites, 
crystallites  and  dust. 

*  The  rock  is  pale  grey,  with  parallel  flow-structure  indicated  by 
minute,  drawn-out  steam-pores,  each  with  a  white  lining ;  on  weather- 
ing,  a  reddish  brown  tint  develops  round  each  pore.  It  shows  a  fissile 
firacture.  Sparing  striated  febpars  (up  to  3  mm.),  and  fewer  augites 
(up  to  I  mm.)  are  visible  with  the  lens.  Under  the  microscope, 
the  PLAOiocLASB  shows  two  sizes  of  phenocrysts :— 0)  oligoglasb-am* 
DBSINB  about  AbsAnj  in  iairly  large  crystals  of  prismatic  habit; 
occasionally  faintly  zoned;  with  a  rounded  core,  honeycombed  by 
a  network  of  groundmass,  surrounded  by  a  border  of  clear  fdspar  with 
sharp  crystal-outline,  both  with  con:inuous  optical  structure;  often 
encloses  tiny  augite  granules  in  a  peripheral  zone ; — (ii)  oliooolasb  in 
smaller  idiomorphs  is  clear  and  fresh,  and  often  zoned.  Augitb  ia 
present  in  pale  green  idioiaorphs;  quite  fresh,  and  free  from  in- 
clusions ;  one  case  approaches  the  hourglass  form ;  the  larger  crystals 
aore  usually  in  groups,  but  the  majority  are  quite  smalL  Bronzttb, 
resorbed  hornb^bmob,  and  maonbtxtb  occur  just  as  in  the  last  rckrk 
(No.  173);  an  instance  occurred  of  a  skeletal  crystal  of  hornblende. 
The  groundmass  is  a  dusky  microfelt  (sometimes  stained  by  iroft 
oxide)  of  plagiodase  microUtes  (oligodue),  with  jagged  ends  and 
in  fluidal  arrangement,  together  with  abundant  augite  and  magnetite 
granules,  and  some  interstitial,  colourless  glass.  TRmncm  is 
present  in  little  nests  of  imbricated  aggregates. 


The  BingA  Volcano  213 

^.  ^,  distance  these  boulders  present  the  appearance  of  a 

Kds^  Cyclopean  wall,  doubtless  giving  rise  to  the  legend 

^  ,^t  that    Tamerlane   (Timur),    during   his    invasion    of 

T  ' -'  •  Armenia,  built  a  castle  here — Whence  the  name  Demir- 

r^  kala  (Timur's  castle)  applied  to  this  peak.    A  speci- 

-05232^  ^^^^  taken  from  the  base  of  this  natural  wall  proved 

to  be  a  pyroxene- hornblende-andesite  (No,  174),*  very 

.^   .  similar  to  the  last  two  rocks  (Nos.  172,  173),  and  they 

^^'  *  doubtless  all  form  part  of  the  same  effusive  mass. 
hsaik  The  great  transverse  rampart  continues  its  E.S.E. 

jiw»  direAion  for  about  2  miles  beyond  the  eastern  summit 

^^  before  it  curves  round,  first  to  S.E.,  and  then  south- 

i  mi:  ward.    Its  southern  side  is  rounded  and  grassy,  al- 

'.'^  *  The  rock  is  pale  bluish  grey,  showing  irregular  fracture;  with 

^/i^  many  small,  drawn-out  steam-vesicles,  which  have  a  reddish  border 

^.  due  to  weathering;   minute,  white  crystals  (tridymite)  often  project 

^^'  into  the  cavities.    A  few  felspars  (up  to  a  mm.)  and  scattered  augites 

(up  to  2  mm.)  can  be  distmguished.    Under  the  microscope,  the 
PULOiocLASE  (andesinb  about  Ab4Aji3)  is  present  in  sparing  idio- 
sjo?  morphs  of  tabular  habit;    usually  rather  free  from  inclusions,  but 

^^  occasionally  containing  a  netwprk  of  groundmass ;  a  peripheral  line  of 

ni/5  dust  is  not  uncommon;   sometimes  faintly  zoned;   encloses  augite. 

pie?  AuoiTE  is  rather  sparingly  present  in  pale  green  idiomorphs,  down 

^  to  rather  small  dimensions;   often  in  groups;   sometimes  twinned; 

jg^i  occasionally  endosv^s  magnetite.     Brohzitx  occurs  in  smaller  and  still 

^  less  frequent  crystals,  as  in  the  two  last  rocks.    A  few  hornblende 

^>  oystals,  completely  resorbed^  are  still  recognizable  by  their  outlines ; 

>fC  they  show  a  core  of  magnetite,  an  inner  border  of  secondary,  colour- 

^  less  or  very  pale  brown  augite,  and  an  outer,  mirrow  border  of 

12  magnetite   and   augite   graniUes.     Magnetite    is   accessory.     The 

j  oiu>ONDMASS  is  a  very  fine,  streaky  microfelt  of  plagioclase  laths 

^  f  oligedase),  with  fluidal  arrangement,  sinking  down  to  microlites  with 

j  jagg^  ends,  together  with  augite  and  magnetite  granules,  and  sparing, 

interstitial,  colourless  glass,  crowded  with  belonites  and  dust.  Tri- 
DTMi te  occurs  in  nests  of  the  characteristic  imbricated  aggregates ; 
when  projecting  into  cavities  they  develop  into  crystals  with  pyramidal 
terminations.  A  fragment  of  more  crystalline  lava  is  enclosed;  it 
b  composed  of  oligodase  laths,  with  augite  and  magnetite  granules, 
all  of  larger  dimensions  than  those  of  die  groundmass.  Some  pale 
yellow  chalcedony,  with  concentric  markings,  fiOs  irregular  cavities. 


214  Geology  of  Armenia 

though  steep  (p.  210),  but  it  breaks  away  to  the  north 
in  precipitous,  rocky  walls,  with  snow  still  lying 
(August)  wherever  it  could  retain  a  foothold.  The 
transverse  rampart,  throughout  its  length,  shows  the 
same  remarkable  contrast  between  its  northern  and 
southern  aspedls,  and  this  is  intensified  in  the  stretch 
between  the  eastern  and  western  summits.  From 
either  of  these  high  positions  it  was  easy  to  see  that 
the  northern  area  of  the  Bingol  Dagh  consists  of  two 
great,  rocky  cirques,  widely  open  to  the  north,  and 
partially  separated  from  each  other  by  the  long 
tongue  jutting  out  from  Karakala,  about  midway 
between  the  eastern  and  western  summits.  The  key 
to  the  contrast  is  to  be  sought  in  the  faA  that  the 
northern  and  north-eastern  slopes  of  the  Armenian 
mountains  receive  less  sunshine  than  the  southern; 
hence  the  accumulation  of  snow  on  the  colder,  and  its 
absence  on  the  warmer  aspedl.  The  adlion  of  frost  is 
greater  on  the  north  side,  the  rocks  in  consequence  are 
continually  being  shattered  and  riven,  whilst  the 
debris  is  rapidly  cleared  away  by  the  streams  fed  by 
the  melting  snows,  and  in  former  times  by  glaciers,  of 
which  I  fotmd  unmistakeable  traces  {vide  infra). 

On  another  occasion  we  ascended  the  transverse 
rampart  at  its  western  end,  where  it  joins  the  S.W. 
rampart  (p.  206).  Skirting  the  extensive  marshes  at 
its  southern  base,  we  rose  at  first  by  a  gentle  slope 
of  6°,  increasing  to  15^  The  rock  is  a  grey  andesite 
similar  to  that  of  Demirkala.  When  about  a  quarter 
of  a  mile  south  of  the  western  summit  or  Bingolkala 
(10,757  f^^t)t  I  discovered  a  stone  (3Xi^xi  feet),  in 
shape  like  a  milestone,  with  9  lines  of  a  cuneiform^ 
Vannic  inscription,^*  which  had  been  greatly  defaced 

*  Professor  A.  H.  Sayce  has  kindly  furnished  me  with  the  following 
statement  with  r^;ard  to  this  fragmentary  inscription  : — 

''.  .  You  are  right  in  thinking  that  it  is  imperfect  at  the  beginning, 


N.W. 


■ 


I 

I 


^Osw'ALT>    ,fe7 


ft 


V 


^''•'i 


•«*l*.J*t 


V 


**••»*«  **.. 


ffliSfS**^  A?. «  -imf,. 


m 


The  Bingdl  Volcano  215 

by  the  rude  carving  of  an  Armenian  sun-cross. 

HI.    The  Northern  Region. 

An  entry  into  the  great  cirque  from  its  southern 
wall,  the  transverse  rampart,  is  only  possible  at  a  spot 
a  little  west  of  Karakala,  owing  to  the  presence  of 
a  larger  sheet  of  snow  here  than  elsewhere.  The 
floor  of  the  western  cirque  is  extremely  uneven,  but 
shows  a  continuous  slope  northwards ;  it  is  a  tumbled 
m^s  of  lava-hummocks.  The  lava  is  an  olivine- 
basalt  (No.  175),*^  very  similar  to  that  of  the  N.W. 

as,  according  to  the  analogy  of  the  other  Vaxmic  texts,  it  ought  to 
oommence  with  an  invocation  to  the  god  Khaldis.  I  beEeve  that  the 
stone  must  have  been  cut  into  its  present  form  at  the  same  time  that 
the  cross  was  engraved  upon  it,  and  that  consequently  the  beginnings 
of  the  lines  are  lost  The  characters  have  unfortunately  been  so 
mutilated  by  the  cross  that  but  few  of  them  are  legible. 

The  first  line  is  evidently : — Sari-du-u-ri-s. 

The  second  line  might  have  contained  the  word  ArgisHkhinis^  "  the 
son  of  Argistis.^' 

Of  line  3  nothing  remains  that  is  decipherable. 

line  4  begins  with  character  pu^  and  line  5  begins  with  ka^  and 
ends  with  1. 

In  line  6  we  have  apparently  the  name  of  a  country. 

Line   7   I  should  read: — kha-a-ldhuyn^    ''the  country 

of I  cut  off." 

Line  8  '.-—pa-ri -^-/»-« [?]-[i]-a,  "  from  the  country  of  Alusia." 

In  line  9  we  have  the  suffix  -i  of  the  nominative,  although  it 
may  be,  as  you  note,  a  mutilated  si ;  then  come  ma  [?J,  /i,  gis^  a  lost 
letter,  and  the  suffix  ni^  which  must  be  the  termmation  of  an 
accusative. 

It  is  interesting  to  find  a  monument  of  Sarduris  near  Erzerum." 

It  may  be  noted  in  addition  that  Sarduris  II,  son  of  Argistis, 
was  at  the  height  of  his  power  in  750  b.c. 

*  The  rock  is  grey  and  compact,  with  irregular  fracture ;  it  contains 
abundant  striated  felspars  (up  to  3  mm.),  numerous  augites  (up  to 
5  mm.),  and  a  few  tiny  grains  of  olivine.  Under  the  microscope,  the 
PLAGiocLASE  (LABRADOftiTK  about  AbiAni)  occurs  in  large,  tabular 
idiomorphs,  honeycombed  by  a  network  of  groundmass;  faintly 
zouvjd ;   albite  twinning  very  fine  and  close ;  often  with  the  "  washed- 


ai6  Geology  of  Armenia 

Bingol  plateau  (No.  177,  vide  infra\  and  anterior 
to  the  pyroxene-andesite  of  the  transverse  rampart. 

The  former  presence  of  a  glacier  was  at  once 
evident  to  me,  for  the  lava-bosses  are  typical  roches 
moutonnees  or  *  dolphins*  backs ' :  they  are  elliptical, 
with  their  long  axis  from  south  to  north,  rounded 
from  side  to  side,  sloping  gently  to  the  south,  but 
breaking  ofiF  abruptly  to  the  north;  the  surface  is 
fluted  and  striated  in  a  northerly  diredlion  both  on 
tlie  top  and  on  the  sides,  especially  where  the 
vegetation  has  preserved  these  markings  from  the 
adtion  of  the  weather.  I  was  able  to  trace  these 
striations  nearly  up  to  the  head  of  the  cirque. 

Myriads  of  small  tarns  lie  in  the  peaty  and  marshy 
spaces  between  the  arid  lava-bosses,  and  have  given 
the  mountain  its  name  (Bingol  Dagh  =  the  mountain 
of  a  thousand  lakes).  Their  presence  is  probably  due 
to  the  accumulation  of  rock-flour  from  the  glacial 
adlion  furnishing  an  impervious  clay.  These  are 
the  spots  where  the  large  blue  gentian  (Gentiana 
septemjida,  Pall.,  var.  cordi folia)  loves  to  flourish. 
This  plant  may  indeed  be  said  to  mark  by  itself  the 
limits  of  former  ice-adlion  on  the  Bingol  Dagh,  for 
it  does  not  extend  beyond  the  morainic  region. 

The  western  cirque  is  bounded  on  the  west  by  the 

out"  twin-mottling  characteristic  of  the  Nimrud  basalts  (p.  133); 
encloses  olivine  and  augite.  There  is  also  a  smaller  generation 
of  prismatic  idiomorphs  of  andesinb  (about  Ab4An3),  with  well- 
defined  albite  twinning ;  clear  and  free  from  inclusions,  except  some 
pale  brown,  dusty  glass.  Augite  occurs  in  pale  green  idiomorphs, 
sometimes  a  deeper  green  centrally,  and  then  feebly  pleochroic  (to 
pale  pink);  occasionally  twinned.  Ouvinb  idiomorplu  are  smaller 
and  less  numerous  than  the  augite ;  very  pale  yellow,  almost  colour* 
less ;  encloses  magnetite.  MAGNETriB  is  present  in  rdatively  laige 
crystals  and  grains,  chiefly  associated  with  augite.  The  oroundmass 
is  a  close  felt  of  very  small  plagioclase  laths  (oligodase)  in  fluidal 
arrangement,  tiny  augite  granules^  and  still  more  abundant  magnetite 
granules. 


The  Bingbl  Volcam  217 

Aghrikala — a  high  ridge,  with  3  conspicuous  peaks, 
extending  northwards  n:om  a  point  a  little  east  of  the 
western  summit  (Bingolkala) ;  it  probably  marks  one 
of  the  lines  of  emission  of  the  later,  viscous  lava,  as  in 
ihe  case  of  the  transverse  rampart  (p.  206),  to  which 
4t  stands  at  right  angles. 

The  eastern  boundary  of  the  western  cirque  is  a 
long  tongue  of  terraced  lava,  extending  northwards 
of  Karakala,  which  rises  up  like  a  bastion  between 
the  two  great  cirques.  A  dike  of  harder  rock  could 
be  seen  to  traverse  the  face  of  Karakala.  This  ridge 
also  probably  marks  a  line  of  emission^  but  it  has 
been  muob  abraded  by  weathering  and  ice-adlion. 

After  crossing  the  Karakala  dividing-ridge,  I 
entered  the  eastern  cirque.  Ice-strise  occur  in  the 
rocky,  western  part  of  this  cirque,  but  the  eastern 
half  is  smoother,  grassy  and  marshy  in  places,  and 
with  larger  lakes.  The  eastern  boundary  of  the 
cirque  is  a  long  ridge  extending  northwards  from 
Demirkala  (see  plan),  and  very  gradually  decreasing 
in  height ;  it  clearly  marks  a  former  line  of  emission, 
for  the  lines  of  flow  of  the  lava  bend  over  on  both 
$ides. 

It  was  noticeable  that  snow  still  lingered  in 
August  on  the  east  side  of  these  three  S -N.  ramparts, 
and  this  aspedt  is  at  the  same  time  steeper  and  more 
precipitous  than  the  western  side,  which  receives 
more  sunshine,  and  was  quite  free  from  snow.  This 
is  doubtless  the  reason  why  the  area  embraced 
between  the  Aghrikala  ridge  and  the  N.W.  rampart 
(extending  from  Bingolkala)  has  not  been  scooped 
put  into  a  cirque,  but  merely  shows  a  gradual  slope 
of  about  20®  down  to  the  extensive,  western  plateau 
of  the  Bingol  Dagh.  This  little,  triangular  area  lies 
Just  outside  the  glacial  region,  and  probably  closely 
approximates  to   the  original  state  of  the  Bingol 


2i8  Geology  of  Armenia 

campaxts  after  the  last  effusion  of  lava  (which  must 
have  been  very  viscous),  and  before  the  formation 
of  the  great  cirques  had  commenced. 

On  the  other  hand,  the  ice,  which  during  the 
glacial  period  coUedled  on  the  transverse  rampart, 
and  flowed  in  glaciers^  grooving  and  scraping  the 
northern  cirques,  must  have  also  issued  through  a 
breach  in  the  eastern  wall,  just  N.E.  of  Demirkala 
(see  plan).  Thence  the  ice  flowed  for  a  few  miles 
to  the  N,E.  over  the  plateau  above  Khinis,  leaving 
moraine-heaps  behind  on  the  higher  slopes.  The 
lava  of  the  eastern  plateau,  just  below  and  east  of  the 
gap,  is  a  glassy  olivine-basalt  (No.  176).** 

The  evidence  of  former  glaciers  on  the  north  side  of 
the  Bingol  Dagh  became  amplified  by  my  observing 
three  distinct  moraines,  which  commence  at  about 
2  to  3  miles  from  the  cliffs  of  the  cirques.  The 
Bingol  glacier  must  have  been  partially  divided  near 
its  head  by  the  long,  terraced  ridge,  which  extends 
northwards  from  Karakala.  This  ridge  ends  in  a 
steep  cliff,  and  its  line  is  continued  by  the  median 
moraine  (see   plan).    Similarly  the    Demirkala  and 

*  The  rock  is  dark  grey  and  compact,  with  irr^;ular  fracture;  it 
contains  numerous  striated  felspars  (up  to  6  mm.),  and  a  fair  number 
of  augites  (up  to  3  mm.;  conspicuous  on  the  weathered  surface),  and 
smaller,  more  frequent  olivines.  Under  the  microscope,  the  plagio- 
CLASE  (LABRADORiTE  about  AbiAni)  is  in  large,  tabular  idiomorphs, 
sometimes  in  groups;  ddicately  zoned;  fairly  clear  and  free  from 
inclusions,  but  tiny  augite  granules  and  glass-inclusions  are  oc» 
casionallfr  ranged  along  a  peripheral  zone;  pericline  twinning  was 
observed,  in  addition  to  the  usual  albite  and  Carlsbad  plans.  Auorrt 
occurs  in  pale  green  idiomorphs,  sometimes  twinned.  Ouvinb  li 
present  in  smaller  and  more  abundant  idiomorphs  than  the  augite ;  it 
has  been  altered  in  part  to  iDDiNOsrrs ;  the  alteration  usually  begins 
centrally,  as  in  No.  153,  p.  183 ;  frequently  surrounded  by  a  wreath 
of  augite  and  magnetite  granules.  The  groundmass  is  a  rather  find 
felt  of  small  plagioclase  laths  (oligoclase),  with  abundant  granules  of 
augite  and  magnetite,  in  a  base  of  brown  glass,  dusty  with  belonites» 


The  BingSl  Volcano  219 

■\ 
Aghrikala  ramparts  of  the  cirque-region  are  produced 
northwards  into  broad,  lateral  moraines,  which  are 
characterized  by  innumerable  lakelets,  lyin^  between 
long  mounds  of  angular  boulders  and  debris  of  lava. 
The  level  of  the  median  moraine  at  a  distance  of 
about  4  miles  from  Karakala  is  9022  feet — sl  difference 
of  about  1000  feet  in  height ;  this  means  an  average 
gradient  of  i  in  20. 

On  approaching  the  western  moraine  from  our 
third  camp  on  the  N.W.  plateau  (about  3  miles  N.W. 
of  Bing5lkala)  it  presented  the  appearance  of  a  sharp- 
ly defined  wall,  20-30  feet  high,  crowned  by  elliptical 
mounds  ranged  in  lines  parallel  to  the  lateral  wall. 

The  two  V-shaped  angles  formed  by  the  jundlion  of 
the  median  and  lateral  moraines  are  respectively 
occupied  by  broad,  grassy,  level  areas,  in  marked 
contrast  to  the  rocky  hummocks  to  the  south.  These 
areas  are  probably  the  sites  of  lakes,  once  dammed  up 
by  the  raorainic  wall  during  the  thawing  period  of 
the  glacier,  and  now  drained  away  by  the  streams 
which  rise  in  the  two  cirques,  and  join  to  form  the 
Bingdl  Su  of  the  Khinis  plain.  But  this  river  has  had 
to  flow  8  or  9  miles  before  beginning  to  carve  out 
its  deep  and  narrow  canyon  (p.  63).  On  the  other 
hand,  the  Goshkar  Su  and  its  tributaries  on  the  south 
side  of  the  transverse  rampart  scarcely  flow  a  mile 
before  plunging  down  into  deep  ravines ;  this  sudden- 
ly developed  ^tosive  power  may  be  ascribed  to  the 
extremely  rapid  difference  in  level  between  the  Bingdl 
plateau  (9000  feet)  and  the  Giimgiim  depression  (4800 
feet),  at  the  base  of  the  great  Bingol  cliffs  (about  4000 
feet  high).  The  streams  are  far  more  numerous  on  the 
east  side  of  the  Bingdl  Dagh,  where  the  snow  lingers, 
than  on  the  sunnier  western  slopes,  whence  the 
Araxes  derives  its  waters. 

It  is  evident  that  the  interior  of  the  Bingdl  cirque 


220  Geology  of  Armenia 

was  not  visited  by  either  Tchihatcheff*  or  Radde,"f 
both  of  whom  ascended  the  mountain,  for  they  could 
not  otherwise  have  failed  to  notice  the  imequivocal 
evidence    of   ice-adlion.     Both   travellers   considered 
that  the  cirque-region  (Tchihatchefifs  "Central  Pla- 
teau ")  represents  a  volcanic  crater,  the  northern  wall 
of  which  had  been  totally  destroyed.    Tchihatcheff 
indeed  attributed  the  strong  northward  slope  of  this 
region  to  a  tilting-up  "induced  by  the  same  cata- 
strophe which  caused  the  complete  annihilation  of 
the  north  border  of  the  crater."     However,  there  is  no 
evidence  to  show  that  a  central  crater  ever  existed  on 
the  Bingol  Dagh.     My  observations  all  tend  to  show 
that  this  volcano  built  itself  up  by  means  of  fissure- 
eruptions,  and  that  the  later  lava  was  very  viscous, 
welling  up  to  form  the  rounded  ramparts   of   the 
central  region. 

IV.    The  North-West  Plateau. 

From  our  third  camp  we  proceeded  in  a  N.W. 
direction  over  the  western  plateau,  which  is  similar  in 
character  to  its  southern  portion  near  our  first  camp 
(p.  205).  Here  the  lava  is  an  olivine-basalt  (No. 
1 77), t — Tchihatcheff 's    "  doleritic    porphyry  " — ^fissile 

*  ^ste  Mtneure^  Ghlogie^  ii.  279-284. 

t  Der  Bin-gifl'dagh^  &»c.f  Petermann's  Mitth.  xxiii.  411-422. 
Gotha,  1877.  Mainly  botanical.  With  a  sketch-map  of  the  southern 
(Goshkar)  region,  drawn  by  his  companion  Sawrieff. 

t  The  rock  is  compact  and  light  grey  with  a  slight  purplish  tinge. 
Striated  felspars  and  frequent  augites  (both  up  to  3  mm.)  are  discern^ 
ible.  Under  the  microscope,  the  plagioclasb  (labradorite  about 
AbiAnx)  occurs  in  large,  tabular  idiomorphs,  much  corroded  and 
honeycombed  by  a  network  of  groundmass ;  with  the  faint,  ^'washed- 
out"  albite  twinning  already  noticed  in  the  basalt  of  the  western 
cirque  (No.  175,  p.  215),  and  in  the  Nimrud  basalts  (p.  133).  There 
is  a  smaller  generation  of  andssinb  idiomorphs  (about  Ab3An2)  of 


/ 


/ 


r 


kantah 


9#*S«'9C 


tATCAU      c^   r»t. 


lit 


f — ^ 


l^liM^   cUiiintf 


99*io'K.  «^^ 


^rr^X 


The  Bin£;dl  Volcano  221 

near  the  surface,  and  decomposing  into  a  brown, 
sandy  clay.  The  lava-jplatesiu  has  a  very  gentle  slope 
to  the  W.  and  N.W.  tor  a  few  miles,  until  it  meets 
a  S.W.'N.E.  line  of  rounded  serpentine^  hiJ^.  It  is 
not  improbable  that  the  isolated  hill  ^MSm  of  our 
third  camp  consists  also  of  sexpentine,  for  it  showed  a 
striking  resemblance  in  contour  to  these  hills,  which 
have  checked  the  basalt  in  its  westerly  flow,  and  have 
forced  it  to  turn  aside  to  the  S.W.  for  some  miles. 
The  lava-stream  finally  bends  round  and  sweeps 
northwards  along  the  western  foot  of  these  serpentine 
hills. 

A  narrow,  deep  valley  enabled  us  to  penetrate  this 
zone  of  hills,  soon  emerging  upon  a  lava-plateau  at  a 
much  lower  level.  We  now  followed  the  stream,  still 
in  a  N.W.  diredlion.  Its  right  bank  consists  of  a 
dark  green,  serpentinized  olivine-diallage  rock  (No. 
178),*  similar  to  that  occurring  in  the  plain  of  Khinis, 

prismatic  habit ;  fresh  and  free  from  inclusions.  Auoitb  is  present  in 
pale  green  idiomorphs,  slightly  pleochroic  (to  a  yellow-green) ;  some- 
times twinned ;  endoses  magnetite.  Olivins  occurs  in  less  frequent, 
rather  small  idiomorphs ;  very  pale  yellow,  almost  colourless ;  quite 
fresh,  but  with  dark  brown  stains  of  iron  oxide  along  cracks ;  some- 
what rounded  at  the  angles,  and  occasionally  intruded  by  ^undmass. 
The  OROUNDMASB  b  a  compact,  close  felt  of  small  plagiodafee  laths 
(oligodase  about  Ab^Ani\  with  abundant  granules  of  augite  and 
magnetite. 

^  The  rock  contains  abundant  crystalline  grains  of  diallagic  augite 
(np  to  3  mm.) ;  remains  of  olivine  grains  (up  to  5  mnu)^  coloured 
brown  by  iron  oxide,  are  also  visible  in  the  matrix  of  dark  green 
serpentine.  Under  die  microscope  it  is  evident  that  the  olivine 
originally  formed  about  2/3  of  the  rock ;  it  has  been  mainly  altered  to 
a  pale  greenish  yellow  sbrpsntine,  with  very  little  separation  of 
magnetite;  the  last-formed  serpentine  (in  the  meshes)  is  often  stained 
brown  by  iron  oxide.  The  augits  (chrohx-diopsidx)  is  unaltered 
and  allotriomorphic ;  occasionally  with  parallel,  narrow  rods  causing 
schillerization,  and  occurring  in  one  qase  at  30"^  to  the  cleavages; 
serpentine  has  insinuated  itself  along  cracks.  Bastitk  is  very 
sparingly  present.    Maghxtitb  is  also  accessory,  but  no  spinek  are 


aaa  Geology  of  Armenia 

near  Kilissadereh  (No.  40,  p.  63),  almost  due  east,  and 
probably  part  of  the  same  series  of  ultrabasic  in* 
trusions.  This  rock  does  not  appear  to  view  on  the 
left  bank,  which  is  composed  entirely  of  olivine- 
basalt  (No.  179),*  in  successive  lava-flows,  each  not 
exceeding  20  feet  in  thickness.  As  we  proceeded,  the 
valley  cut  more  deeply  into  the  lava-plateau  to  about 
100  feet. 

Shortly  before  reaching  Bastok  a  black,  cindery 
lava,  about  20  feet  thick,  occurs  in  the  bed  of  the 
valley.  Both  kinds  of  lava  show  rudely  columnar 
stru<5lure,  but  the  columns  of  the  cindery  lava  are 
much  thicker  (about  10  feet),  and  their  angles  become 
rounded  in  weathering,  in  contrast  to  the  sharp  edges 
of  the  columns  (about  2  feet  in  diameter)  of  the  stony 
olivine-basalt.  It  is  in  facft  the  same  cindery,  tra- 
chytic  lava  (with  black,  scoriaceous  streaks),  which 
I  had  noticed  at  Khedonun  (No.  32,  p.  56),  at  Kalaji 
(p.  57),  at  Khinis  (No.  43,  p.  65),  and  at  Kara  Seid 

Tisible. 

*  The  rock  is  compact,  iron-grey  with  a  slight  greenish  tinge,  and 
with  irregular  fracture.  It  (contains  fairly  numerous,  striated  felspars 
(up  to  9  mm.),  tand  less  frequent  olivines  (up  to  i  mm.).  The 
rock  is  slightly  magnetic  (4'').  Under  the  microsoope,  the  plaoio- 
CLASR  (labradoritb  about  Ab5An6)  occurs  in  tabular  idiomorphs, 
usually  clear,  but  sometimes  with  a  central  network  of  brown, 
dusty  glass  and  augite  granules;  frequently  zoned;  encloses  small 
olivines.  A  second  generation  of  smaller,  more  numerous  crystals 
consists  of  ANDBSINB  (about  Ab4An3) ;  they  are  of  prismatic  habit^ 
and  do  not  show  zoning.  Olivikb  is  the  only  other  phenocryst, 
occuning  in  numerous,  mostly  small  oystals;  quite  fresh,  neariy 
colourless  (with  a  faint  yellow  tinge),  sometimes  in  groups,  and 
occasionally  invaded  by  grouadmass ;  orange  stains  of  iron  oxide 
often  occur  along  cracks.  The  oaouimMASS  is  a  ttit  of  plagioclase 
laths  (andesine  about  AbsAnj)  in  fluidal  iorrangement,  together 
with  granules  of  augite  and  magnetite,  both  i^out  equally  abundant, 
and  about  the  same  size.  A  veiy  little  brown,  dusty  glass  is  oni^ 
exceptionally  visible. 


The  BingOl  Volcano  333 

(p.  198),  all  on  the  outskirts  of  the  Bingol  plateau. 
It  does  not  extend  far  west  of  Bastok.  We  soon 
emerged  upon  a  lower  level  of  the  plateau,  riding 
over  pale  grey,  stony  basalt,  and  about  2  miles  N.W. 
of  Bastok  we  forded  the  Aras  (Araxes)  at  Shekan, 
-which  lies  on  the  left  bank — a,  low  cliff  of  basalt 
in  columns  10  feet  in  diam«ter.  The  river,  which 
rises  just  west  of  Aghrikala,  is  here  a  swiftly  flowing 
stream,  with  an  alluvial  plain  about  a  mile  in  width. 
Still  proceeding  to  the  N.W.,  we  rode  over  the 
plain,  with  low  hills  on  our  left  of  scoriaceous  lava, 
weathering  to  a  brick-red.  This  lava  caps  another 
zone  of  serpentine  hills,  which  stretch  away  to  the 
W.S.W.  Finally  we  reached  Gugoghlan,  situated 
at  the  foot  of  a  conspicuous  hill  of  serpentine  and 
limestone,  capped  by  lava  {vide  infra). 

From  the  summit  of  this  hill  we  obtained  a 
commanding  view  over  the  northern  and  western 
slopes  of  the  Bingol  Dagh.  In  this  view  perhaps  the 
most  striking  features  are  the  two  great  blocks  of 
Kherbesor  and  Kiregan,  which  rise  up  above  the 
general  level  of  the  Bingol  plateau,  to  which  they 
present  steep  cliSs.  In  both  cases  the  lofty  cliffs 
reveal  white  strata  (presumably  lacustrine  cp.  p.  56), 
either  imderlying  the  lava,  as  at  Kiregan,  or  infer* 
bedded  with  it,  as  already  observed  at  Kherbesor 
(p.  58).  These  blocks  must  be  portions  of  the  plateau 
which  have  either  remained  in  place  whilst  the 
surrounding  country  has  sunk  down,  or  else — and  this 
alternative  seems  to  me  far  more  probable — they  have 
been  squeezed  up  to  a  higher  altitude  than  they 
originally  possessed. 

The  entire  volcanic  adlivity  of  the  Bingdl  Dagh 
would  appear  to  have  taken  place  withtn  the  limits  of 
the  recent  lake-deposits  of  Armenia,  for  the  oldest 
observed  lavas  (on  the  south  side)  overlie  lacustrine 


214  Geology  of  Armenia 

clays  (pp.  I99y  30i),  whilst  the  latest  lavas  (oa  the 
north  and  east)  are  either  interbedded  with  lake-marls 
(as  at  Kheddnun,  p.  56,  and  Kherbesor,  p.  58),  or  else 
underlie  lake-marls  with  Dreissensia  polymcrpha  (as 
at  Khinis,  pp.  66,  68). 

V.    The  Volcanic  Succession  of  the  Bingol  Dagh. 

The  following  table  summarizes  my  observations  on 
the  relative  age  of  the  lavas  of  the  Bingol  Dagh» 
set  forth  in  downward  succession.  Doubtless  there 
are  several  gaps,  which  future  investigation  will  fill 
in,  for  the  Bingol  plateau  is  too  vast  an  area  to  be 
covered  in  the  short  space  of  time  at  my  disposaL 
The  lavas  under  each  heading  may  be  regarded  as 
piadtically  contemporaneous : — 

X.   Cindery   Migite-hombiendc    trachyte    of   the    outskirts    of 
the  Bingol  pkteatt,   /.^.»  at  Elhedontm  (No.  39,  p.   56),    Kalaji 


i 


(p.  57),  Ali  Mur  (p.  60),  the  upper  Bingol  Su  (p.  62)»  Khinis  (No.  4J1 
pi.  65),  Kan  Seid  (p.  198)  and  at  Bastok  (p.  aas). 

2*  Pyroxene-hornblende  andesite,  often  with  tridymitet 

(^)  forming  the  central  ramparts  of  the  summit-region :  on 
Demirkala  (Na  1729  p.  an ;  No.  173,  p.  axa ;  ft 
No.  if4t  p.  az3),  widi  its  glaasy  selvage  (No.  Z7I1 
p.  six). 

!ii)  of  the  S.E.  plateau,  peak  jp  (No.  170,  p.  209), 
iii)  of  the  N.E.  plateau,   at  Ali   Mur  (Nos.   36  &  37» 

p.  60). 
0y)  Similar  lava  occurs  further  north  on  the  Madrak  plateau 
(No.  29,  p.  53)  and  on  the  Eyerli  Dagh  (No.  a6y  p.  49). 

3.  Grey,  basaltic  augite-andesite  of  the  S.  W.  plateau  (p.  905).  ' 

4*  OUtine  basalt,  somewhat  andesitic  in  character, 

(i)  of  the  plateau  N.E.  of  Demirkala  (No.  176,  p.  218),  the 
western  cfarque  (No.  175,  p.  2x5),  the  western  pkiteaO 
(No.  177,  p.  220), 
(li)  of  the  S.W.  plateau  (No.   165,  p.  203},  and  of  the 
N»W.  plateau  (No.  179,  p.  922). 

5*  Mica  andflsite  of  the  Goshkar  region  (Na  167,  p.  907). 

6.  Doleiitic  olivine  basalt  of  the  southern  rq^on  (Na   x6S| 


The  Btngdl  Volcano  325 

p.  so9y  and  p.  ao6). 

7.  Staalt  toff  of  the  southern  r^on  (No.  166,  p.  107). 

8.  Rhyolitic  hornblende  andesite  near  the  base  of  the  Btngol 
difi  (No.  164  y  p.  203). 

VI.    Return  to  Erjzerum. 

Our  camp  at  Gugoghlan  (7045  feet)  was  situated  on 
one  of  the  low  mounds  of  travertine  which  border  the 
west  side  of  the  serpentine  hills,  partly  flanked  and 
capped  by  lava  (p.  223).  A  few  of  the  calcareous 
springs  were  still  acftive;  a  species  of  Chara  grew 
luxuriantly  in  the  clear  bubbling  basins. 

We  approached  the  hill  of  Gugoghlan  from  its 
south  side,  riding  at  first  oyer  a  long  flow  of  scoria- 
ceous  lava  (weathering  a  rusty  brown),  which  has 
proceeded  from  the  S»W.,  flooding  the  base  of  the  hill. 
This  lava  has  the  appearance  of  being  older  than 
the  superficial  Bingol  lavas,  and  is  probably  in 
connexion  with  the  conspicuous  volcanic  peak  of 
Sheikhjik,  a  few  miles  further  west.  The  southern 
and  eastern  slopes  of  Gugoghlan  hill  consist  of  brown 
and  green  serpentine — ^an  enstatite-olivine  rock  (No. 
i8o).^  This  ultrabasic  rock  is  intrusive  into  a  cream- 
coloured  limestone,  dipping  30®  N.N.W.,  which  rises 
up  in  bold  crags  on  the  south  and  west  sides  of 
the  hill.    Owing  in  part  to  contadl-metamorphism 

*  The  rock  consists  of  laxge,  lostrous,  crystalline  grains  of  bastite 
(up  to  JI2  mm.)  in  a  matrix  of  rusty  brown  serpentine,  with  veins 
of  pale  green  serpentine.  The  rodL  has  been  slickensided  along 
parallel  joints.  Under  the  microscope,  ENSTATrrs  is  seen  to  form  the 
bulk  of  the  rock;  sometimes  dear  and  freshi  but  mainly  altered 
into  BASTrrs,  and  penetrated  by  veins  of  serpentine ;  parallel  rodlets 
(causing  schillbrization)  are  visible  in  some  of  the  unaltered  grains. 
Some  unaltered  ouvins  b  still  present  in  the  meshes  of  the  yellowish 
serpentine,  which  is  much  stained  by  iron  oxide.  Maonbhtb  is  only 
sparingly  present,  and  spinels  are  almost  absent,  except  for  an 
occasional,  dark  brown  grain  of  ohromite. 


\ 


226  Geology  of  Armenia 

the  limestone  is  almost  barren  of  fossils,  but  traces  of 

ThamnaroM  polymorpha,  Abich, 

establish  its  age  as  Lower  Miocene.  The  least  altered 
part  of  the  limestone  showed  a  loose  texture,  some- 
what simulating  travertine  (No.  181),  very  similar 
to  the  travertinous  appearance  of  the  Miocene  lime* 
stone  of  Khanik  (p.  96)  and  Adeljivas  (p.  161).  In 
this  case,  however,  it  has  evidently  been  caused  by 
percolating  water  dissolving  away  the  shells  of  Fora- 
minifera,  mainly  of  Amphistegina^  the  form  of  which 
is  perfectly  recognisable.  A  fragment  of  a  small 
Peden^  and  a  few  barrel-shaped  joints  of  Crinoid 
stems  were  the  only  other  recognisable  remains. 
This  limestone  has  not  only  been  intruded  by  the 
serpentine,  but  has  also  been  baked  by  the  overlying 
sheet  of  lava — ^an  oli vine-basalt  (No.  182)* — which 
forms  the  adtual  summit  of  the  hill  (c.  8000  feet). 

On  continuing  the  journey  to  Erzerum  I  proceeded 
northwards,  firstly  across  the  alluvial  plain  of  the 
Araxes,  and  then  over  undulating  ground  composed 
of   lava>5ows,    which    have   apparently  come  from 

*  The  rock  is  compact  and  fine-grained,  with  subconchoidal,  rather 
splintery  fracture.  Dark  grey  streaks  alternate  with  purple,  broader 
bands ;  the  rock  weathers  a  rusty  brown ;  it  contains  scattered  brown 
granules  (altered  olivine),  and  many  smaller  black  specks  (augite). 
Under  the  microscope,  olivine  occurs  in  abundant  idiomorphs,  almost 
entirely  altered  into  brown  iddinosits  (slighUy  pleochroic;  finely 
fibrous),  but  under  a  high  power  small  patches  of  unaltered  mineral 
can  still  be  discerned;  very  free  from  inclusions,  but  sometimes 
penetrated  by  groundmass.  Augite  is  present  in  somewhat  less 
numerous  pale  green  idiomorphs ;  often  twinned,  sometimes  repeated- 
ly; occasionally  encloses  olivine  and  magnetite.  Felspar  does  not 
occur  as  a  phenocryst.  The  oaoimDifAss  is  a  felt  of  plagioclase  laths 
(labradorite  about  AbxAni)  in  fluidal  arrangement,  numerous  granules 
and  minute  crystals  of  augite,  and  abundant  magnetite  granules.  The 
darker  streaks  are  caused  b^  a  greater  proportion  of  augite  and 
magnetite  granules. 


The  Bingol  VolcceM  arj 

volcanic  hills  lying  to  the  west  and  north-west.  The 
grey,  basaltic  lava  is  compadl  and  free  from  steam - 
holes. 

On  looking  back  to  Gugoghlan  hill  I  could  see 
that  the  Miocene  limestone  crops  out  also  on  the 
north  and  north-west  sides  of  the  hi]l,  dipping  ^bout 
"1^  N.N.Wt,  as  on  the  south  side.  It  has  been  baked 
brown  by  the  overlying  lava. 

Soon  after  passing  through  a  yaila  of  Zireki  Kurds, 
situated  in  a  hollow  between  two  lava-streams,  I 
began  to  ascend  the  rounded  E.-W.  ridge  which 
overlooks  the  Altun  plain,  and  has  caused  the  Araxes 
to  turn  eastwards.  I  found  this  ridge  to  consist 
of  grey,  unfossiliferous,  marly  clays,  dipping  ^o^ 
N.N.W.  (probiably  continuous  with  the  recent  lalce- 
deposits  of  Khedonun,  a  few  miles  further  east,  p.  56). 
These  strata  are  overlain  by  a  sheiet  of  basaltic  lava, 
which  has  baked  the  clay  to  ai  purple  and  bufi 
porceilanite.  Springs  gu<h  out  at  the  jundlion  of  the 
porous  lava  with  the  impervious  clay. 

The  summit  is  part  of  an  extensive  lava-plateau, 
forming  gently  undulating,  grassy  country,  carved 
out  into  valleys  by  tributaries  of  the  Araxes.  The 
most  considerable  of  these  mountainrstreams  is  the 
Egri-chai,  which  we  had  crossed  lower  down  at 
Duzyurt  on  our  outward  journey  (p.  55).  A  small 
lake  occurs  in  its  upper  course.  The  lacustrine  clays 
and  marls  were  still  visible  in  this  valley,  underlying 
the  lava ;  but  I  could  see,  at  a  distance  of  a  few  miles 
to  the  S.W.,  a  ridge  of  bufiF  limestone  (probably 
Miocene)^  similar  in  appear^ce  to  that  of  Gugoghlan, 
rising  out  of  the  surrounding  lava. 

The  highest  part  of  the  lava-plateau  is  9408  feet ; 
it  is  the  western  continuation  of  the  heights  above 
Madrak  (p.  53).  Here  Palanddken  and  Eyerli  Dagh 
came  into  full  view  on  the  further  side  of  a  broad, 


228  Geology  of  Armenia 

grassy  valley,  which  must  have  been  in  existence  at 
the  time  of  emission  of  the  lava,  for  the  molten  rock 
has  clearly  flowed  down  into  the  valley,  not  only  from 
the  south,  but  also  from  the  Eyerli  Dagh  on  the  north* 

As  I  descended  into  this  valley  serpentine  cropped 
out  from  beneath  the  lava-sheet;  it  weathers  to  a 
greenish  grey,  except  near  the  overlying  lava,  where  it 
is  reddish  owing  to  contadl-adlion. 

Still  further  north  a  limestone  came  into  view — 
doubtless  a  continuation  of  the  Miocene  limestone 
of  Madrak,  only  a  few  miles  to  the  eastward  (p.  52). 

Many  springs  arise  from  beneath  the  relatively  thin 
sheet  of  lava.  In  crossing  the  broad  valley,  now 
inadequately  occupied  by  a  brook,  I  passed  almost 
imperceptibly  across  the  Frat-Araxes  watershed, 
formed  by  grassy  meadows.  It  seems  to  me  not 
impossible  that  this  wide,  unconstricfted  valley  may 
have  been  originally  occupied  by  a  large  left  tributary 
of  the  Araxes — proceeding  perhaps  from  the  Frat- 
Kelkid  watershed — ^and  subsequently  tapped  and 
captured  by  the  Frat  as  the  latter  gradually  ate  its 
way  back  through  the  deep  gorge  now  existing 
between  Erzingan  and  Mamakhatun. 

During  my  ascent  up  to  the  pass  over  the  Eyerli 
Dagh  I  found  the  southern  slope  to  consist  of  grey 
augite-andesite,  with  a  selvage  in  places  of  duU 
black  glass.  The  final  stage  of  my  journey  to 
Erzerum  across  the  Palandoken-Eyerli  cirque  has 
already  been  described  (pp.  49,  50)  in  my  account 
of  the  cirque. 


CHAPTER  XIV 


ERZERUM  TO  TREBIZOND 


From  Erzerum  we  retraced  our  steps  as  far  as  Ilija  (p. 
44),  but  from  this  spot  we  struck  across  the  plain 
to  the  N.W.,  after  fording  the  Frat.  The  pebbles 
of  the  river-gravel  consist  mainly  of  black,  scoriaceous 
lava,  but  I  noticed  also  some  fragments  of  obsidian, 
evidently  derived  from  the  volcanic  heights  on  the 
north  side  of  the  plain  of  Erzerum.  The  low  cliflFs 
north  of  the  Frat  are  composed  of  white  lake-marls, 
dipping  very  gently  to  the  N.W, 

So  far  as  I  could  judge  from  a  distance,  the  rugged, 
brown  cliffs  to  the  S.  and  S.W.  of  the  volcanic  cone 
Sheikhjik,  seem  to  consist  of  the  same  ultrabasic  rock 
which  plays  so  great  a  part  in  the  struAure  of  Palan- 
doken  (p.  46),  on  the  opposite  side  of  the  plain  cf 
Erzerum.  Here  it  appeared  to  be  intrusive  in  a 
limestone. 

After  passing  a  ruined  khan  in  the  plain  we  reached 
the  banks  of  the  Surchemeh-chai,  a  rapid  stream  quite 
as  large  or  even  larger  than  the  Frat;  its  valley 
is    about    i^    mile   in  breadth.    The   cliffs   display 

greenish  grey  strata,  out  of  which  the  same  rusty 
rown  ultrabasic  rock  rises  up  in  island-masses 
wherever  the  enveloping  deposits  have  been  washed 
away  by  the  river.    The  rock  is  evidently  older  in 


330  Geology  of 

date,  for  no  sign  of  any  alteration  is  visible  in   the 
adjacent  strata. 

These  greenish  grey,  sandy  marls  show  current - 
bedding,  and  dip  30°  S.S.E.  At  Maimansur  on  the 
north  side  of  the  valley  the  beds  are  more  gravelly, 
and  contain  frequent  pebbles  of  serpentine,  diallage 
rock  and  especially  of  grey  marble.  They  are  pro- 
bably of  Pleistocene  age.* 

From    Maimansur    we   ascended   steadily   to    the 
W.N.W.,  over  high  downs  of  an  impure,  shaly  lime- 
stone with  pale  greenish  bands,  evidently  composed 
in  part  of  serpentinous  elements;    the  dip  was  35^ 
N.N.W.,  changing  after  a  time  to  S.S.E. ;    there  w^as 
no  sign  of  any  fossils.    Abichf  surmised  that  these 
beds  might  be  of  Eocene  age,  but,  since  no  known 
Eocene  occurs  either  along  this  line  of  strike,  or  at 
any  point  nearer  than   Baiburt   (to   the   N.W.)   or 
Almaly  (S.W.  of  Mamakhatun),  where  it  is  alwa}^ 
well    characterized    by    Nummulites,   his   conjedlure 
(unsupported  by  fossil  evidence)  does  not  seem  to  me 
very  probable.    On  the  other  hand  the  strata  are 
on  the  same  line  of  strike  as  the  Miocene  limestone  of 
Kardarich,   Ashkala  and   Pirnakapan,   and   it   may 
therefore  be  tentatively  assumed  that  they  are  of  the 
same  age — probably  deposited  near  the  north  coast  of 
the  Miocene  sea,  which  was  then  bounded  by  the 
Cretaceous  heights  now  forming  the  Frat-Chorokh 
watershed. 

After  a  long  and  gradual  ascent  we  reached  the 
ridge  overlooking  the  N.E.-S.W.  valley  which  still 
separated  us  from  these  imposing  heights  of  white 
marble.    We  encamped  at  Golviren  (7039  feet),  near 

*  Cp«  Abich,  G^.  Forsch.  in  dm  kauk.  Landem^  ii.  134,  and  see 
Chapter  XXTV  on  the  Pliocene  and  Pleistooene  of  Armenia. 

t  Op.  dt.<,  ii.  107. 


Erzerum  to  TfAizond  231 

the  head  of  this  valley,  almost  under  the  shadow 
of  the  cockscomb  crest  of  the  Akhbaba  Dagh.  The 
banks  of  the  stream  are  formed  of  a  hornblende- 
diorite  (No.  183),^  which  has  imdergone  some  meta- 
morphism. 

Between  GSlviren  and  the  Cretaceous  heights  of 
white  marble  there  lies  a  broad  zone  of  intrusive, 
rusty  brown  olivine- diallage  rocks  and  serpentines, 
decomposing  into  greenish  grey  clays.  The  horn- 
blende-diorite  of  Golviren  has  to  all  appearances  been 
a  later  intrusion  through  these  ultrabasic  rocks, 
which  occupy  the  greater  part  of  the  valley;  the 
diallage  crystals  are  frequently  an  inch  in  length. 
The  building  material  of  the  houses  is  mainly 
serpentine;  marmorized  limestone  (No.  185)  and 
diorite  are  used  to  a  less  extent. 

*  The  rock  contains  abundant  white  felspars  (up  to  5  mm.),  and 
hornblende  needles  (up  to  5  xam.)  in  a  finer  grained  matrix  of  the 
same  minerals ;  there  is  a  little  locsil  efifervescence  with  add.  Under 
the  microscope,  plaoioclass  (andesine  about  Ab4A|i3)  in  allotrio- 
morphic  grains  is  seen  to  form  the  bulk  of  the  rock ;  turbid  and  much 
obscured  by  pale  brown,  dusty  decomposition-products,  and  crowded 
with  tiny  needles  of  white  mica,  and  with  some  calcite ;  it  is  every- 
where undergoing  alteration  into  a  granular  mosaic  of  dear  secondary 
felspar  (albite).  A  little  orthoclasb  is  also  present.  Horkblxsdk 
occurs  in  pale  brown  idiomorphs,  with  well-marked  pleochroism  :  a — 
pale  straw,  almost  colourless,  dumber  brown,  ^^darker  brown  with 
a  greenish  tinge;  sometimes  twinned,  and  often  showing  undulose 
extinction ;  it  is  altering  marginally  to  actinoute,  which  is  altogether 
more  abundant,  occurrmg  in  sheaves  and  radiate  tufts;  its  needles 
often  traverse  the  felspars,  especially  the  clear  secondary  grains;  it 
shows  the  usual  pieoduroism :  a — ^[4k  greenish  yellow,  ^ — ydlowish 
greeu;  ^— <iark  blue-green.  Small  granules  of  xpidote  are  associated 
with  tko  hornblende.  Apatitb  and  lxucoxxnx  (in  irregular  grains) 
are  accessory. 

Another  spedmen  (No.  184)  taken  dose  to  serpentine,  is  a  very 
compact  and  homogeneous,  pale  green,  quartsofelspathic  rock,  with 
sparing,  very  minute  hornblendes,  giving  a  slighUy  silky  lustre  to  the 
parallel  joint-planes. 


334  Geology  of  Armenia 

strata  raised  up  by  faulting,  for  in  another  ten 
minutes,  just  after  they  showed  a  dip  of  ^5^  N.  by  W.# 
I  met  with  the  massive  marble  (also  dipping  N.  by 
W.)»  again  intruded  by  two  zones  of  serpentinous 
rock,  each  about  30  feet  thick.  This,  however,  soon 
gave  place  again  to  the  series  of  limestones  and 
shales,  dipping  25®  N.  by  W.;  the  valley  widened, 
and  a  gentler  charadter  was  imparted  to  the  land* 
scape  by  the  presence  of  these  softer  rocks. 

At  ^azalar  IChan,  surrounded  by  high,  grassy 
mountains,  I  observed  a  syncline,  the  dip  changing  to 
45"^  S.  by  E.;  but  the  whole  series  became  so  greatly 
contorted  and  puckered  (cp.  p.  38)  that  it  was  not 
possible  to  note  all  the  sudden  changes  of  dire(5lion. 
In  many  cases  the  small  folds  have  been  so  sharply 
crushed  that  one  set  of  strata  may  stand  abruptly 
at  right  angles  to  another  set.  Where  adjacent  layer* 
have  become  slightly  separated  there  is  often  a 
deposit  of  rhombohedral  calcite  (dog's  tooth  spar). 
The  prevalent  dip,  however,  of  these  contorted  beds 
is  35®  N.  by  W.  We  were  then  following  the  course 
of  the  Januren  riveij.  Some  travertine  springs  on  thfe 
right  bank  are  still  in  full  adlivity. 

We  made  a  short  halt  about  midway  between 
Jenapa  and  Akhveran.  Here  the  series  of  limestone 
and  shale  has  been  again  sharply  bent  at  a  right 
angle.  The  layers  of  limestone  do  not  exceed  8 
inches  in  thickness;  they  are  much  veined  with 
calcite,  but  the  alternating  shales  retain  their  soft 
charadlen  After  a  close  search  for  fossils  I  merely 
succeeded  in  finding  some  crushed  and  indeterminable 
remains  of  RhynchonellOf  and  a  fragmentary  im- 
pression, which  Mr.  G.  C.  Crick  has  kindly  examined 
and  determined  as  a  Macroscaphites^  not  improbably 

Macroscaphites  Ivani^  d'Orbigny. 


irll 


:*::g: 


Ergerum  to  Trebizand  235 

In  this  connedtion  it  is  interesting  to  note  that 
Batsevitch  *  found  this  fossil  to  the  N.E.,  in  the  lower 
valley  of  the  Ardanuch  river,  occurring  "  near  Gorgo- 
takhan,  in  greenish  grey  limestone,  which  forms  the 
base  of  the  Cretaceous  in  this  region."  Although 
this  locality  is  about  lop  miles  distant,  yet  an  inter- 
mediate observation,  50  miles  N,E.  of  Akhveran,  has 
been  made  by  Hamilton.!  He  remarks  on  the 
similarity  of  "the  thinly  bedded  limestone,  sand- 
stone, and  calcareous  shale  between  Liesgaff  and 
Tortum"  to  the  equally  contorted  series  between 
Baiburt  and  Khoshabpunar,  which  occurs  on  the  same 
line  of  strike.  On  the  other  hand,  AbichJ  found 
near  Halwa  Madejr(about  %6  miles  west  of  Akhveran) 
"  light  grey  limestones  with  a  distindl  impression  of  a 
Scaphites/'  which  unfortunately  he  did  not  figure.  It 
is  probable  therefore  that  this  Neocomian  zone  lying 
between  the  Chorokh  and  the  Kop-Khoshabpunar 
line  of  heights  will  be  found,  on  future  investigation, 
to  have  a  wide  extension  both  to  N.E.  and  W. 

Before  reaching  Akhveran  the  limestone  beds 
became  thicker,  dipping  30°  N.  by  W.;  some  local 
reddening  of  the  shales  perhaps  indicates  the  proxi- 
mity of  still  concealed,  intrusive  rock.  The  valley 
widens  out  on  coming  to  Akhveran,  and  here  we 
obtained  an  uninterrupted  view  of  the  fiat-topped 
Kop  Dagh,  13  miles  away  to  the  south-west. 

The  series  of  limestones  and  shales  here  vanished 
conformably  beneath  a  pale  bufif,  concretionary  lime- 
stone (No.  188),  dipping  60^  N.  by  W.,  and  gradually 
decreasing  to  35®.    No  fossils  were  visible,  but  under 

*  GeffI,  Researches  in  the  Batum  6*  Arttdn  distruts^  Mat  Geol. 
Cauc.  Tiflis,  1887,  Ser.  2,  i.  131,  155.    [In  Russian]. 

t  Researches  in  Asia  Mincr^  i.  2x5. 

X  Ged.  Forsch.  in  den  hank.  Landim^  u.  xix. 


236  Geology  of  Armenia 

the  microscope  the  calcareous  alga  Munieria^  which 
occurs  in  the  Urgonian  of  Bakony,*  is  clearly  present. 

A  zone  of  intrusive  serpentine  occurs  along  the 
strike,  near  the  confluence  of  the  Kurt  river,  and  some 
marmorizing  of  the  limestone  has  ensued.  A  yellow, 
compadl  sandstone  (No.  189)  of  considerable  thick- 
ness overlies  the  massive  limestone,  which  rises  in  tall 
crags  and  readily  weathers  into  caves.  Just  before  we 
reached  the  Chorokh  the  dip  changed  from  35®  N.  by 
W.  to  60°  S.  by  E.  At  the  confluence  I  noticed  an  old 
river-terrace,  20-30  feet  above  the  present  level.  As 
we  proceeded  down  the  Chorokh  valley  the  limestone 
consisted  of  thinner  beds,  dipping  45°  S.  by  E.,  but 
the  approach  of  night  precluded  any  further  ob- 
servations, and  we  had  to  ride  for  2  hours  more  in 
total  darkness  before  reaching  Maden  Khan  (5434 
feet),  where  we  struck  our  outward  route  (p.  36). 

On  the  following  morning  (September  2nd)  I 
noticed  that  the  escarpment  on  the  south  side  of 
the  broad  Chorokh  valley  shows  a  dip  of  28°  S.  by  E., 
whilst  the  dip  on  the  north  bank  is  25°  N.  by  W, 
Hence  the  valley  (as  already  noted,  p.  36)  follows  an 
anticlinal  axis.  The  rock  is  a  massive,  grey,  un- 
fossiliferous  limestone,  traversed  by  veins  of  calcite. 

The  Chorokh  has  cut  through  extensive  terraces  of 
travertine,  stretching  along  the  base  of  the  limestone 
cliffs.  For  a  considerable  distance  down  the  river  no 
igneous  rocks  occurred,  and  on  looking  up  the  valley 
of  a  left  tributary  nothing  but  the  bold,  rounded 
contours  of  the  limestone  were  visible  for  several 
miles  to  the  south. 

Still  further  down  the  Chorokh  valley  a  series  of 
yellow,  calcareous  sandstones  rises  up  m  thin  layers, 

*  See  M.  Hovelacque's  Album  des  microphotographies^  1900,  Plate 
xlvi,  2. 


Erzerum  to  TrAuumd  ayj 

frequently  contorted,  dipping  25®  N.  by  W.  on  the 
north  bank  and  80^  S.  by  £.  on  the  south  bank.  The 
alluvial  terrace  of  old  river-gravels  lies  here  about  30 
feet  above  the  present  level  of  the  river. 

As  we  proceeded,  the  sandstones  were  seen  to  over- 
lie a  series  of  grey  limestones  and  shales,  alternating 
in  thin  layers,  and  exhibiting  a  considerable  amount 
of  contortion.  It  is  probably  the  westerly  repetition 
of  the  similar  (Neocomian)  beds  already  described  (pp. 
37,  38,  233,  234,  235),  lying  between  the  Chorokh  and 
the  Kop-Khoshabpunar  line  of  heights.  Owing  to 
the  softer  nature  of  these  strata  the  valley  widens  out 
considerably. 

Shortly  before  the  Chorokh  makes  its  great  bend 
round  to  the  north,  the  strata  on  the  right  bank 
are  much  reddened  owing  to  the  baking  of  the  shales 
by  an  intrusion  of  olivine-diallage  rock  in  a  broad 
zone  along  the  strike. 

As  the  Chorokh  bends  round  in  its  wide  valley,  the 
grey  limestones  and  shales  could  be  seen  to  continue 
westwards,  dipping  30°  N.  by  W.,  and  overlain 
conformably  by  calcareous  sandstones.  Occasional 
reddening  by  intrusive  rocks  was  still  visible. 

At  Baiburt  (5075  feet)  these  strata  are  overlain 
by  barren,  grey  marble.  According  to  Abich*  tliis 
marble  is  of  Turonian  age,  but  I  could  find  no  traces 
of  fossils  in  the  grey  marble  on  the  west  side  of  the 

*  Geol.  Farsch.  in  ien  kauk.  Zandem,  ii.  120:— "The  Upper 
JRudistes  zone  prevails  on  the  heights  between  Baiburt  and  Sipikor, 
just  as  at  Khoshabpunar  and  Baiburt/'  and  on  p.  109  he  also  states 
that  ''  indistinct  and  very  sparing  traces  of  fossils  point  to  the  lime- 
stone of  the  Kop  Dagh  and  Khoshabpunar  system  being  Hippuritic 
limestone."  However,  my  fossils  of  Neocomian  (Hauterivian)  age 
(PP*  39>  ^3^9  334)  show  that  Abich's  surmise  was  inconrect  with  regard 
to  the  Kop  Dagh  and  Khoshabpunar  pass,  and  it  is  therefore  not 
improbable  that  he  was  equally  mistaken  with  regard  to  Baiburt,  for  in 
no  case  did  he  mention  any  fossils. 


338  Geology  of  Annmia 

bridge — the  only  place  which  I  was  able  to  examine. 
Ve^tation  is  extremely  scanty,  and  the  bare  rock 
frequently  appears  to  view,  weathering  to  a  pinkish 
brown,  on  the  high,  rounded  hill  (surmounted  by 
a  turbek)  just  west  of  the  bridge  and  north  of  our 
road.  The  strata  of  this  hill  appeared  to  show  an 
anticline,  the  dip  changing  from  35**  S.  by  E,  to  the 
same  angle  N.  by  W,  They  are  overlain  on  the  N,W. 
and  W.  by  strata  which  are  presumably  Eocene,* 
The  junction  was  not  clear  at  a  distance,  but  this 
marble  seemed  to  be  conformably  overlain  by  grey 
shales,  and  these  in  turn  by  yellow,  calcareous  sand- 
stone with  massive,  mural  jointing.  This  sandstone 
forms  low,  undulating  downs — the  commencement 
of  the  wide  plain  of  Varzahan  (p.  3&).  I  examined^ 
small  quarry  of  this  excellent  building-stone,  but 
I  could  find  no  fossils.  Probably  the  Nummulites 
which  Abichf  colledled  near  Baiburt  came  from  a 
locality  north  of  the  town.  Here  the  rock  is  close- 
grained,  with  occasional,  parallel,  brown  bands  of 
iron  oxide.  It  dips  30^  S.  by  E.,  soon  changing  to  20*^ 
N.  by  W.,  and  is  then  conformably  overlain  by  a  pale 
brown,  sandy  limestone  in  thin  beds. 

On  looking  backwards,  I  could  see  that  a  wide 
zone  of  the  barren,  rounded  hills  of  marble  (pre- 
sumably Lower  Cretaceous)  extends  to  the  north  and 
east  of  Baiburt.  The  thinly-bedded,  yellow  or  brown 
sandy  limestone  became  more  massive  as  we  pro- 
ceeded westwards ;  it  dips  25*^  S.  by  E.,  and  continues 
in  gently  undulating  downs,  with  occasional,  low 
escarpments,  as  far  as  Varzahan.    No  fossils  could  be 

*  Cp.  Abich's  section  of  the  Cretaoeons  and  Eocene  at  Baiburt 
GeoL  Forsch.  in  den  kauk,  Ldndem^  ii.  1x4. 

t  Op.  dt^  ii.  115,  341-249,  and  Plates  ix,  sd;  and  see  my  Chapter 
XXI  on  the  Eocene  of  Armenia. 


Erzerum  to  Trebizand  239 

seen  anywhere ;  greenish  grey  concretions  were  some- 
times present.  Just  before  we  descended  to  the  adlual 
plain  of  Varzahan  we  crossed  an  anticlinal  ridge 
of  greenish  sandstone — the  dip  changing  to  20®  N.  by 
W.    It  yields  a  grey  soil. 

At  Varzahan  (5620  feet)  we  left  our  previous  route 
(p-  3^)y  ^^d  rode  northwards  over  the  gently  sloping 
plain,  passing  across  a  broad  syncline  of  pale  green 
sandstone— the  same  rock  which  has  been  used  for  the 
keystones  over  the  doorway  of  Surp  Khach  at  Var- 
zahan (p.  36). 

On  reaching  Kerzi  (5214  feet),  on  the  north  side  of 
the  plain,  I  found  an  outcrop  of  Nummulitic  Lime- 
stone, dipping  35*^  S.S.E.  and  rising  up  from  beneath 
the  green  sandstone.  The  nodular,  bulf  limestone 
contained  very  abundant  individuals  of  a  variety  of 

Nummulites  perforata^  de  Montfort.* 

Ill-preserved  specimens  of  Spondylus^  sp.  indet.,*^ 
were  the  only  other  fossils  visible.  The  Nummulite 
appears  to  me  to  closely  resemble  d'Archiac's  var.  B. 
columbresensis,  subvar.  E.  It  seems  to  be  the  same 
variety  which  Abich  figured  f  merely  as  N.  perforatay 
stating  that  "  it  is  the  species  of  Nummulite  which  is 
the  most  widely  distributed,  not  only  in  the  Araxes 
valley,  but  in  Armenia  generally.''^ 

As  a  result  of  my  discovery  of  Nummulites  on 
the  northern  border  of  the  Varzahan  plain,  it  may, 
1  think,  be  deduced  that  this  plain  (or,  more  broadly 
speaking,  the  Kelkid-Lower  Chorokh  basin)  about  20 
miles  in  diameter,  should  be  regarded  as  a  part  of 
a  geosyncline,  for  it  is  only  in  the  upturned  strata  on 

*  Kindly  confirmed  by  Mr.  R.  BuUen  Newton,  F.C.S. 
t  Ver^eichende  geoi.  Grutukuge^  etc.,  Pl*te  III,  iff,  «,  b. 
%  Op.  at.  168. 


240  Geology  of  Armenia 

the  northern  edge  at  Kerzi,  and  on  the  southern 
margin  at  Sadagh,  Plur,  Warlisne,  and  Baiburt  *  that 
the  Nummulitic  strata  rise  up,  whilst  the  central 
portion  is  occupied  by  overlying,  unfossiliferous, 
calcareous  sandstones. 

The  Nummulitic  Limestone  at  Kerzi  is  in  turn 
underlain  by  unfossiliferous,  calcareous  flagstones, 
rising  up  to  form  a  small  but  steep  escarpment  facing 
the  north-west. 

Large,  waterwom  boulders  of  quartz-diorite  were 
numerous  in  the  river-bed ;  they  had  evidently  been 
brought  down  from  the  Pontic  Range,  to  which  the 
Eocene  beds  form  the  southern  selvage. 

On  proceeding  to  the  north-west,  the  dip  of  the 
flagstones  increased  to  75°  S.S.E.,  but  before  reaching 
Ostuk  it  changed  to  25°  N.N.W.  The  limestone  is 
reddened  locally,  probably  by  still  hidden  intrusive 
rocks.  At  Ostuk  we  ascended  a  left  tributary;  the 
ravine  has  been  cut  through  a  |frey  marble,  very 
massive  and  evenly  bedded,  dipping  50®  S.S.E.  It 
is  quite  different  in  charadter  to  the  overlying  Eocene 
limestones,  and  it  may  with  probability  be  regarded 
as  Cretaceous,  for  it  occurs  on  the  same  line  of  strike 
as  the  very  similar,  massive  limestones  (presumably 
Cretaceous,  pp.  30,  33,  35)  of  Balakhor,  a  few  miles  to 
the  S.W.  An  anticline  was  soon  noticeable,  the  dip 
changing  to  20°  N.N.W. 

A  little  higher  up  the  valley  I  noticed  an  intrusive 
sill  of  hornblende-porphyrite  (No.  190),!  reddening  on 

*  Abich,  Geoi.  Forsch.  in  den  kauk.  Zandem^  ii.  11 5-1 17. 

t  The  rock  is  pale  greenish  grejr,  compact,  with  somewhat  fissile 
fracture.  It  contains  white,  striated  felspars  (up  to  3  mm.),  and  dull 
green,  chloritic  pseudomorphs  after  hornblende  (up  to  3  mm.) ;  the 
rock  effervesces  briskly  with  add.  Under  the  microscope,  the  plaoio* 
CLASS  (ANDBsm  about  Ab4An3)  is  rather  sparingly  present  in  smalls 
tabular  idiomorphs,  which  have  been  very  laigdy  replaced  by  caldte^ 


L 


Erzerum  to  Trebizand  24J 

both  sides  the  limestone,  which  here  dips  35^  S.S.E. 
Another  intrusive  sill .  of  a  similar,  greenish  grey 
porphyrite  soon  came  into  view ;  in  this  case  it  had 
caused  some  local  contortion  in  the  strata. 

Huge  boulders  of  diorite  became  more  frequent 
in  the  river-bed,  and  had  evidently  not  travelled 
far.  At  Torosos  we  came  to  an  andesitic  agglo- 
merate (No.  191),**  which  overlay  a  quartz-mica- 
bat  a  narrow  margin  of  clear  felspar  always  remains;  sometimes 
2onedy  with  alternate  layers  of  dear  felspar  and  caldte.  Pseudo- 
morphs  after  hornblende  are  abundant,  down  to  rather  small  dimen- 
sions, all  with  the  characteristic  outlines,  but  none  of  the  original 
mineral  has  escaped  alteration ;  the  crystals  now  consist  of  a  very  fine 
aggr^ate  of  magnetite  granules,  with  caldte  and  chlorite.  A  few 
pseudomorphs  of  caldte  and  quartz,  or  of  calcite  and  chlorite  show 
outlines  characteristic  of  augits,  and  may  perhaps  represent  that 
mineral.  The  oroundmass  b  rather  fine-grained  and  orthophyric, 
consisting  of  stumpy,  rectangular  prisms  of  orthoclase  with  more 
frequent,  narrow  laths  of  oligoclase,  magnetite  granules  and  some 
interstitial  quartz;  abundant  calcite  part  ides  are  also  present,  with 
a  little  chlorite ;  but  there  is  no  augite  now  present  in  the  groundmass. 

*  The  rock  is  dark  green  and  compact,  with  numerous,  angular, 
greenish  black  rock-fragments  (up  to  7  mm.),  containing  frequent  pink 
and  white  felspars  (up  to  5  mm.),  sometimes  with  a  green  core,  and 
also  small  green  hornblende  prisms  (up  to  2  mm.),  with  silky  lustre. 
The  rock  is  slightly  magnetic  ^4"").  Under  the  microscope,  the  lava- 
fragments  consist  of  the  following  andesitic  t|rpes  : — 

(i)  A  very  fine  microfelt  of  felspar  microlites  (oligoclase)  with 
abundant,  interstitial  chlorite  granules  (perhaps  after  augite),  and 
magnetite  granules.  The  phenocrysts  consist  of  plaoioglasb  (an- 
DESDfS  about  AbsAnj)  in  idiomorphs  of  prismatic  habit,  often  con- 
taining epidote  and  chlorite ;  pseudomorphs  after  augitb  of  chlorite 
and  epidote ;  and  in  particular  large  crystals  of  pale  green  fibrous 
HORiiBLBirDS,  oftcn  paler  internally  and  patchy  in  colour;  mostly 
quite  fresh,  but  occasionally  partly  altered  centrally  to  a  minutely 
fibrous,  serpentinous  aggregate  with  epidote ;  pleochroism :  a — ^pale 
straw,  ^ — ^brown-green,  ^—blue-green ;  sometimes  twinned ;  the  ex- 
tinction-angle is  up  to  13^ ;  no  resorption-border,  although  sometimes 
with  melted-down  outlines ;  however,  well-marked  prism-terminations 
sometimes  occur ;  endoses  magnetite  sparingly.  In  some  cases  these 
andesitic  fragments  show  perlitic  markings,  now  outlined  in  chlorite. 


242  Geology  of  Armenia 

diorite  (No,  192).* 

(ii)  StUl  finer  microfeltSy  rhyolitic  and  crjrptooTstallme,  with 
felspar  miorolites  and  nuignetite  dust  and  granules ;  the  dear  matrix 
has  a  pale  bluish  tinge;  they  contain  phenocrysts  of  plagioclasx 
(andbsins)  and  also  of  OBTHOCLASSy  both  somewhat  decomposed, 
as  well  as  chlorite  pseudomorphs  after  augitb. 

(iii)  Black,  angular  lapilli  occur  more  seldom;  vnder  a  high 
power  Uie  groundmass  consists  of  densely  crowded  magnetite  granules 
and  felspar  microlites  (oligodase),  with  phenocrysts  of  plagioclask 
(andbsins  about  Ab4An3),  green  hornblbndb,  pseudomorphs  of 
chlorite  and  epidote  after  augitb  ;  the  round  steam-cavities  are  now 
filled  with  chlorite. 

The  sparing  oroundmabs  in  which  all  these  fragments  lie  consists  of 
the  same  crystals  and  finer  fragments,  with  mudi  interstitial  chlorite 
and  epidote ;  a  rather  large  magnetite  grain  encloses  pinkish  brown 
apatite  crystals. 

*  The  rock  is  holocrystalline  and  rather  fine-grained ;  about  2/3  of 
the  rock  consists  of  abundant,  white  plagioclase  crystals  (up  to  2 
mm.)  in  a  matrix  of  pink  orthoclase  and  quartz  ;  the  remainder  is 
composed  of  dark  ffreen,  lustrous  crystals  of  hornblende  and  pyro* 
xene  (up  to  6  mm.),  with  biotite  flakes  (up  to  5  mm.).  Under  the 
microscope,  the  plagioclasb  (labradokitb  about  AbiAni)  is  for  the 
most  part  idiomorphic,  with  very  close  and  fine  albite  twinning ; 
zoning  is  frequent  in  the  clear  periphery,  whilst  the  central  part  is  not 
zoned,  and  contains  many  tiny  flakes  of  white  mica;  sometimes 
encloses  augite  in  crystals  or  snudl,  rounds  grains ;  there  is  a  clear 
instance  of  idiomorphism  towards  biotite.  Obthoclabe  occurs  in 
allotriomorphic  grains,  quite  subordinate  in  amount,  rather  less  than 
the  quartz,  and  turbid  with  brown,  dusty  decomposition-products* 
When  unaltered,  the  augitb  is  a  very  pale  green,  but  nearly  all  of 
it  has  been  converted  into  urautic  hornblbndb;  this  is  present 
in  allotriomorphic  grains ;  the  alteration  is  rarely  complete,  and  a  few 
scraps  of  augite  generally  remain;  occasionally  intergrown  with 
biotite ;  the  pleochroism  of  the  hornblende  is :  a — pide  greenish 
yellow,  d — brownish  green,  c — blue-green;  granules  of  epidote  are 
sometimes  present  in  these  paramorphs.  A  few  groups  of  rounded, 
crystalline  grains  of  hypbrsthbmb  are  also  prerent ;  quite  fresh,  with 
marked  pleochroism :  a — brownish  pink,  ^ — ^brownish  yellow,  c — ^blue* 
green.  Biotitx  is  abundant  and  allotriomorphic,  enclosing  apatite. 
Quartz  is  interstitial  and  allotriomorphic,  with  fairly  numerous,  small 
fluid-inclusiohs  containing  a  bubble,  often  ranged  in  strings.  Mao* 
NETiTB  and  APATiTB  are  accessory  associated  widi  the  ferro-magnesiao 
constituents.    A  little  secondary  hornblende  and  magnetite  has  formed 


Erxerum  to  Trehizond  243 

quartz-mica-pyroxene-diorite  shows  much 
similarity  to  the  quarte-mica-augite-diorites  described 
by  Thost  *  from  the  South  Karabagh,  excepting  that 
in  these  rhombic  p3n:oxene  is  absent. 

A  turn  in  the  valley  brought  us  to  Mezereh  Khan, 
the  site  of  a  picturesque  mill.  Now  that  we  iiad  left 
the  region  of  limestones  I  noticed  that  the  river  no 
longer  contains  the  calcareous  alga  Chara^  which  had 
luxuriated  from  Kerzi  upwards.  Our  course  here  lay 
nearly  due  west  through  the  diorite,  which  rises  up  in 
lofty  precipices,  crowned  by  sharp  crags.  It  decom- 
poses to  a  yellow  loam,  which  seemed  to  be  well 
suited  for  corn -growing. 

At  Upper  Mezereh  we  left  the  bed  of  the  torrent, 
and  rose  up  obliquely  on  the  left  bank  to  Ali  Mezer, 
skirting  a  bold  cliff  of  a  coarse,  andesitic  agglomerate 
(Nos.  i93,t  i94-i96)>  which  has  weathered  into  the 

along  cracks.  The  order  of  crystallization  seems  to  have  been  : 
I.  Apatite  and  magnetite;  2.  Plagiodase;  3.  Augite  and  hyper- 
ithene;  4.  Biotite;  5.  Orthodase;  6.  Quarts.  The  augite  had 
commenceid  tc  crystallize  out  before  the  plagioclase  had  finished,  and 
the  biotite  simikurly  overlaps  the  augite. 

*  Abhandl.  Scncktnb.  Natuff.  Ges.  Frankfurt  a/M,  XVIII,  ii.  227- 
«99,  his  Nos.  134,  143, 151. 

t  The  rock  is  coarsely  fragmental.  The  lapilli  are  black,  some- 
times  ^th  a  purplish  tinge  ;  up  to  12  mm.  in  the  hand-spedmen,  but 
often  much  larger  in  the  mass,  up  to  several  inches;  frequently 
highly  amygdaloidal ;  the  amygdales  are  mostly  spherical  (up  to  3 
mm.),  chiefly  composed  of  a  white  zeolite  with  a  lining  of  chlorite, 
sometimes  entirely  of  chlorite.  Striated  felspars  (up  to  3  mm.), 
and  lustrous  augites  (up  to  6  mm.)  occur  both  in  the  lapilli  and  in  the 
green,  chloritic  groundmass,  in  which  patches  of  greenish  yellow 
epidote  are  also  visible  with  the  lens.  Under  the  microscope,  the 
LAPILLI  consist  of  brownish  black  glass,  with  felspar  microlites,  and 
circular  steam-cavities  of  all  sizes — some  filled  with  dilorite,  others 
with  a  zeolite  in  pale  brown,  radially  fibrous  aggregates,  polarizing  in 
low  greys ;  others  again  viih  epidote  in  radial  aggregates ;  in  some 
cases  aU  three  are  present,  the  chlorite  forming  the  lining,  and  the 
seolite  being  later   thlm  the  epidote.    Plaoioclasb  (uibradoritx 


244  Geology  of  Armenia 

charadleristic  *  woolpack  *  manner,  with  rounded  crags 
and  cones.  The  rock  is  nodular,  porphyritic  and 
amygdaloidal,  showing  a  rough  stratification  in 
purple  and  green  layers,  dipping  30^  N.N.W,  It 
weathers  to  a  pale  grey,  the  black,  amygdaloidal 
lapilli  standing  out  conspicuously.  It  overlies  fine- 
grained, shaly  tuffs,  which  are  also  occasionally  inter- 
calated in  the  coarse  agglomerate.  The  microscope 
revealed  the  interesting  facft  that  minute  (probably 
stunted)  Forarainifera  (Globigerina)  are  present  in  this 
tuff  (No.  197)  * 

There  is  a  little  burial-ground,  called  Ali  Mezer 
Vezerli  (7484  feet),  situated  at  the  western  end  of 
the  cliff  of  agglomerate,  where  it  joins  on  to  the 
Kitowa  plateau,  but  here  we  became  enveloped  in  a 

about  Ab5  An6)  occurs  in  idiomorphs  of  prismatic  habit,  much  cracked 
and  more  or  less  completely  altered  to  a  colourless  or  pale  brown 
zeolite,  espedallj  along  the  cracks ;  sometimes  with  a  central  network 
of  brown  glass  or  of  chlorite.  Augitx  is  less  abundant,  in  pale  green 
idiomorphs.  quite  fresh,  sometimes  with  twin-lamellation  or  with  faint 
'herring-bone'  twinning;  encloses  magnetite.  A  few  small  crystals 
of  BRONziTE  are  present,  sometimes  altered  to  a  serpentinous  mineral. 
There  is  a  gradation  in  the  lava-fragments  from  nearly  opaque  or  dark 
brown  to  a  streaky,  dusty,  glassy  matrix,  partially  devitrified  and 
chloritized,  containing  the  same  phenocrysts.  These  rock-fragments, 
together  with  crystal- fragments  of  felspar  and  augite,  lie  in  a  brownish 
dusty  GROUNDMAS8 — the  comminuted  dust  of  the  same  lava,  nearly 
isotropic,  with  much  chlorite,  epidote  and  zeolite. 

*  The  rock  is  a  dull  greenish  grey,  extremely  fine-grained  and 
compact;  occasional,  thin  streaks  of  calcite  occur  along  bedding 
planes ;  it  effervesces  readily  with  add.  Under  the  microscope,  the 
texture  remains  fine-grained  even  with  a  high  power  (1/4^),  but  the 
stratification  is  clearly  evident.  The  rock  mainly  consists  of  dusty 
CALCITE,  but  is  also  much  saturated  with  chlorite;  some  epiuotx 
and  a  zeolite  are  also  present.  Besides  the  minute  tests  of  G/oM" 
gerina^  which  are  filled  with  chlorite  or  with  limonite,  there  also  occur 
numerous,  tiny,  angular  fragments  of  plagioclase  (andesine  about 
AbsAnj),  pale  green  augite,  brown  hornblende  (pleochroism : 
0— straw,  r-— greenish  brown),  icaonktiti  and  quartz,  aU  quite  fresh. 
Clear  caldte  frequently  occurs  in  Uttle  veins. 


'^^mm 


Erzerum  to  Trdnssond  245 

dense,  white  mist,  whi^  unfortunately  obliterated  all 
the  surrounding  features  of  the  country,  and  greatly 
restridled  my  observations.  It  persisted  until  we  had 
crossed  the  Kazikly  Dagh  on  the  following  day. 

From  Ali  Mezer  Vezerli  we  rode  over  high  downs 
(8044  feet),  skirting  the  Kitowa  Dagh.  The  rock 
is  a  volcanic  agglomerate  with  small,  angular  frag- 
ments similar  to  those  in  No.  191  (p.  241).  It  is 
itly  decomposed,  yielding  a  brown,  sandy  soil, 
m,  however,  we  passed  the  western  base  of  some 
black  cliffs,  which  consist  of  a  banded  pitchstone — sl 
glassy  hornblende-bronzite-andesite  (No.  198].* 

This  was  succeeded  by  another  outcrop  or  volcanic 
agglomerate  similar  to  No.  193  (p.  243),  and  then 
overlain  by  hard,  shaly  beds  (similar  to  No.  197,  p. 
244),  in  layers  about  2  inches  thick,  forming  a  shallow 

*  The  rock  is  a  black  glass,  weathering  a  dull  brown,  especially 
along  the  numerous  perlitic  cracks;  it  contains  frequent  striated 
felspars  (up  to  3  mm.),  and  lustrous  hornblendes  (up  to  5  mm.),  which 
stand  out  conspicuously  on  the  weathered  surface.  Under  the  micro- 
scope,  the  plaoioclasb  (andbsinb  about  Ab4An3)  occurs  in  large, 
tabular  idiomorphs,  often  in  groups ;  frequently  zoned ;  cracked,  but 
very  dear  and  fresh ;  inclusions  sometimes  occur  centrally,  consisting 
of  bronzite,  apatite,  pale  brown  glass  and  occasional  grains  of  brown 
hornblende.  Hornblbndb  is  sparingly  present,  in  large,  brown 
prisms  with  well-marked  terminal  faces;  sometimes  in  groups; 
occasionaUy  twinned;  marked  pleochroism:  a — straw,  ^— norange- 
brown,  c — dark  umber-brown ;  quite  fresh ;  encloses  apatite ;  some- 
times intergrown  with  pll^B;ioclase.  Bronzits  is  rather  lesft  abundant, 
in  much  smaller,  pale  yellow,  nearly  colourless  idiomorphs,  with  the 
characteristic  cross-partings ;  pleochroism  :  a — ^pink,  ^ — ^yellow  brown, 
c — blue-green,  all  in  pale  tints  ;  quite  fresh,  but  sometimes  showing  a 
minute  fibrillation.  Maonbtits  and  apatfts  are  accessory.  The 
OROUNDMASS  cousists  of  a  pale  brown  or  colourless  glass,  but  this  has 
been  for  the  most  part  devitrified  and  become  cryptocrystalline,  with 
perlitic  markings  outlined  by  iron  oxide ;  it  is  densely  crowded  with 
belonites,  tricmtes  and  dust,  scattered  microlites  of  oligodase  in 
fluidal  arrangement,  hornblende  needles,  minute  granules  of  augite, 
and  sparing  magnetite  granules.  Irregular,  jagged  cracks  are  now 
filled  with  chalcedony. 


346  Geology  of  Anuinia 

syncline  with  a  dip  of  30^  N.N.W.  to  30^  S.S,E.  A 
sUffht  fault  on  our  left  showed  a  downthrow  of  about 
3  feet  at  40^  S.S.E.  These  fissile  tuffs  were  soon 
overlain  by  amygdaloidal  agglomerate. 

So  far  as  the  mist  would  permit  me  to  see,  our 
route  lay  over  broad,  grassy  downs  forming  the 
watershed  between  the  ChoroKh  and  the  Black  Sea. 
Another  outcrop  of  a  pitchstone — a  glassy  hjrper- 
sthene-andesite  (No.  199)* — occurred  on  the  north 
side  of  the  watershed,  forming  a  low  cliff  on  our  left, 
in  columns  8  inches  in  diameter,  at  the  head  of  a 
steep  ravine.  Here  we  rapidly  descended  to  the  bed 
of  a  stream  flowing  northwards — a,  left  tributary  of 
the  river  which  enters  the  Black  Sea  at  Surmeneh, 

*  The  rock  it  a  Uack  glass,  with  perlitic  cracks.  It  contuns 
numerous  striated  felspan  (up  to  3  mm.]^,  lustrous  augites  (up  to  2 
mm.),  and  still  smaller  black  crystals  with  a  txronze  lustre  (hyper- 
sthene).  Under  the  microscope,  the  pl40ioclasx  (aNDssmE  about 
Ab4An3)  occurs  in  large,  tabular  idiomorphs,  often  beautifully  zoned 
and  usually  very  dear  and  fresh ;  in  some  cases  partly  melted  down ; 
sparing  augite  and  magnetite  granules  may  occur  oentxalhr,  and  there 
is  sometimes  a  peripheral  network  of  twown  ^lass  (often  showing 
bubbles) ;  brown  du^t  is  more  seldom  enclosed,  either  centrally  or  in  a 
broad,  peripheral  zone.  AuGrra  occurs  in  large,  pale  green  idio- 
morphs, often  in  groups;  sometimes  showing  lamellar  twinning; 
small  inclusions  of  apatite,  magnetite  and  brown  glass  are  not 
infrequent  Hypebsthenb  is  present  in  smaller  idiomorphs,  as- 
sociated with  the  augite;  markedly  pleochroic:  a — salmon  pink, 
^ — reddish  yellow,  r— pale  green ;  encloses  apatite.  Hobnblende  is 
aocessoiy,  only  present  in  a  few  individuals;  very  dark  brown,  and 
much  resorbed,  the  smaller  crystals  being  rendered  quite  opaque. 
Maonbtits  is  chiefly  associated  with  the  pyroxenes.  Apatite  is 
accessory,  mostly  in  small  needles,  but  a  few  larger  crystals  are 
present.  The  oroundmass  is  a  colourless  or  pale  brown  gi.ass, 
densely  crowded  with  belonites  and  dust,  forming  lighter  and  darker 
streaks,  with  oligodase  in  sparing  laths  and  abundant  roicrolites  with 
fluidal  arrangement,  and  a  few  augite  and  magnetite  granules.  The 
rock  is  travened  tluK>ughout  by  perlitic  cracks,  stained  with  iron  oxide. 
A  small  angular  frvf^ent  is  present  of  a  more  crystalline  felt  of 
oligoclase  laths,  augite  and  mtisignetite  granules,  with  only  a  little, 
interstitial  glass. 


Erzerum  to  Trehtzond  247 

after  carving  out  the  Kara  Dereh.  This  lava  overlies 
the  greenish  grey,  andesitic  agglomerate,  greatly 
decomposed,  which  in  turn  rests  on  shaly  tuffs, 
dipping  35**  S.  by  E. 

The  gorge  had  already  deepened  considerably  at 
YowuUy  Khan.  Here  the  rock  is  a  massive,  greenish 
white,  earthy  tuff  (No.  200),  weathering  to  a  buff 
colour,  and  similar  to  the  greenish  tuffs  between 
Trebizond  and  Jevizlik  (p.  30).  A  very  similar,  de- 
composed tuff  of  a  rhyolitic  nature  has  been  described 
by  Becke  *  from  the  Beden  Heights  in  Somketia,  in 
the  Eastern  Border-ranges  of  Russian  Armenia.  This 
rock  has  been  too  deeply  kaolinized  and  chloritized  to 
arrive  at  a  positive  determination  concerning  its 
original  composition.  It  is  well  stratified,  thin  beds 
alternating  with  thick,  massive  beds,  dipping  35^  N. 
In  the  crags  which  tower  above  the  hamlet  this  tuff 
seemed  to  be  overlain  by  a  black  lava,  probably  the 
same  flow  which  I  had  observed  at  the  head  of  the 
gorge. 

Below  Yowully  Khan  the  valley  deepened  rapidly, 
becoming  a  gloomy  canyon  with  nearly  vertical  walls, 
over  1000  feet  high.  The  torrent  seemed  to  be 
remarkably  small  to  have  had  the  power  to  carve  out 
such  a  deep  and  precipitous  ravine,  but  the  rapid  fall 
in  level — ^3000  feet  in  about  3^  miles = a  gradient  of  i 
in  6 — ^must  give  it  great  erosive  power.  Doubtless 
a  large  bodv  of  water  descends  earlier  in  the  year 
from  the  melting  of  the  snows  on  the  Kitowa  plateau. 
The  tufe  become  more  shaly  in  their  lower  beds,  and 
have  imdergone  much  disturbance;  rapid  changes 
of  dip  occurred  within  a  very  short  distance,  cg.^  35** 
N.,  45*"  S.,  vertical,  50''  N.,  30°-7o°  N.,  40^  S. 

Shortly  before  we  came  in  sight  of  Keshishkdi  I 
noticed  a  purple  homblende-porphyrite  (No*  201),* 

*  The  rock  is  compact  and  effervesces  generaUy  with  add.    It 


348  Geology  of  Armenia 

in  a  broad  intrusive  vein,  which  has  reddened  the 
strata  on  both  sides.  It  soon  became  the  prevalent 
rock  in  the  lower  part  of  the  valley,  passing  into  a 
green  variety.  We  left  Keshishkoi  on  our  right,  and 
rode  westwards  up  the  main  stream  for  a  short 
distance  as  far  as  Chorak  Khan  (5763  feet). 

From  this  village  we  began  our  ascent  of  the 
Kazikly  Dagh,  which  belongs  to  the  outer  line  of  the 
Pontic  Range,  by  proceeding  up  a  left  tributary  of 
the  main  river.  The  rock  in  the  immediate  proximity 
of  the  khan  is  a  dark  green,  nodular,  amygdaloidal 
augite-andesite,  in  places  presenting  a  brecciated 
appearance.  The  building-stone,  however,  is  a  dark 
purplish  grey  marble,  which  occurred  in  situ  (No.  202) 

contains  abundant,  white,  striated  felspars  (up  to  3  mm.),  and  less 
frequent  chloritic  crystals  (up  to  3  mm.)  with  the  outlines  of  horn- 
blende. Under  the  microscope,  the  plagioclase  (andesinb  about 
Ab5An3)  is  present  in  abundant  idiomorphs  of  prismatic  habit,  turbid 
with  brown,  dusty  decomposition-products,  and  also  often  partially 
replaced  by  caldte;  zoning  is  still  evident;  occasionally  encloses 
apatite  and  magnetite.  Pseudomorphs  are  abundant,  showing  in 
most  cases  the  characteristic  forms  of  hobkblekde;  the  original 
mineral  has  been  either  entirely  resorbed,  or  else  replaced  by  an 
aggregate  of  quartz,  calcite,  pale  green  chlorite,  epidote,  and  limonite. 
The  resorbtion-border  has  the  appearance  of  ink  which  has  slightly 
'  run';  a  high  power  (1/4O  shows  that  this  *  washed-out '  appearance  is 
due  to  innumerable,  tiny  needles  of  magnetite  surrounding  the 
resorbtion-border.  Some  of  t|ie  pseudomorphs  show  no  resorbtioo- 
border,  and  may,  by  their  outlines,  indicate  the  original  presence  of  a 
RHOMBio  PYBOXENE ;  a  fcw  othcTS,  now  Consisting  only  of  calcite  and 
chlorite,  may  similarly  be  referted  to  augitb.  Magnetite  is  fairly 
frequent.  Apatite  is  accessory,  in  small  crystals  of  a  pinkish  tinge, 
slightly  pleochroic  {E — violet,  6^-— pale  yellow).  The  groukdmass  is 
a  pale  brownish,  dusky,  micropoecilitic,  quartzofelspathic  felt  of 
oligodase  microlites  in  fluidal  anangeBoent,  with  skeletal  needles  of 
magnetite,  which  are  crowded  along  streaks,  and  round  the  horn- 
blende pseudomorphs.  Portions  of  the  groundmass  are  isotropic,  and 
appear  to  have  been  opaliced.  Calcite  is  generally  disseminated 
through  the  rock,  and  quarts  i8.^frequent  in  little  veins  and  nests 
of  granules. 


Erzerum  to  TrMztmd  ^49 

a  little  higher  up  the  valley,  interbedded  in  the 
andesite,  and  dipping  30^  S.  by  £.  No  fossils  were 
visible  to  the  eye,  out  the  microscope  revealed  sparing 
remains  of  a  Lithotkamnion^  and  a  se(^ion  of  an 
Echinoid  spine.  This^  rock  is  quite  different  in 
charadler  to  the  Miocene  LithothamnicmAimesiotie  of 
the  Armenian  plateau,  and  it  may  perhaps  find  a 
parallel  in  the  ^^  brecciated  limestones,  with  numerous 
microscopic  remains  of  Lithothamnion  and  fragments 
of  shells,"  which  Inostranzeff  •  records  on  the  southern 
slope  of  the  Caucasus,  regarding  it  as  a  littoral 
deposit  of  Eocene  age,  contemporary  with  the  Num- 
mulitic  Limestone.  A  comparison  may  also  be  made 
to  Gumbel's  *  granite-marble,'  with  Lithothamnion^ 
from  the  Lombardy  Alps  (Montorfano). 

A  vertical  dike,  20  feet  thick,  cuts  through  the 
strata  a  little  further  up  the  valley.  Still  higher  up 
we  passed  through  a  considerable  thickness  of  dark 
grey  or  even  black  marble  with  white  veins;  it 
overlies  the  andesite,  and  forms  an  anticline,  the  dip 
changing  from  50®  S.  by  E.  to  25°  N.  by  W.,  and  tlien 
back  again  to  55°  S.  by  E.  In  some  places  a  purplish 
tinge  has  been  induced  by  contact  with  an  overlying 
sheet  of  andesite.  This  lava  became  in  turn  overlain 
by  a  dark  green,  andesitic  tuff  (No.  203),t  which 

*  Au  trovers  de  ia  ehaine  prindpdU  du  Cauaae^  947,  St  Peten- 
bmg,  1896.    [In  Russian,  widi  a  Firench  summary]. 

t  The  rock  is  full  of  white  particles  (less  than  i  mm.)  in  a  dark 
green  matrix.  It  weathers  a  rusty  brown,  and  readily  eflfervesces  with 
add.  Under  the  microscope,  the  fragmental  nature  of  the  rock  is 
at  once  evident ;  the  most  abundant  fh^ments  are  angular  portions  of 
a  lava,  cryptocrystalline  and  almost  isotropic;  a  high  power  (1/4'} 
shows  it  to  be  a  micropoecilitic,  dusky  felt  of  oligodase  microlites, 
with  sparing  granules  of  augite  and  magnetite,  but  most  of  Uw 
fragments  are  saturated  with  dilorite  and  caldte.  Lapilli  also  occur, 
nearlf  or  quite  opaque ,  with  crystals  of  plagioclasb  (andesine  about 
AbsAn3),  and  magnetite  in  gnaules  and  needles;  the  steam-pores 


250  Geology  of  Armenia 

forms   the   summit   of   the   watershe(L     It    readily 

and  forms  bare,  sweeping  slopes,  recalling 


^«i)iiii:« 


the  scenery  of  the  Zigana  pass  (p.  2^). 

A  rapid  descent  brought  us  to  "fashkSpri,  situated 
on  a  stream  flowing  northwards  to  join  the  river  of 
the  Kara  Dereh.  &ld,  rugged  crags  of  a  hematitic 
quartzite  (No.  204)  *  form  &e  left  bank,  in  marked 
contrast  to  the  gentle  sloi>es  which  we  had  just 
traversed.  The  face  of  the  quartzite  clifiEs  is  much 
slickensided,  and  the  rock  is  accompanied  by  a 
compact  fridtion-breccia  (No.  205).!    It  is  therdfore 

are  filled  with  chlorite.  Numerous  crystals  and  crystal-fiagments  are 
also  present,  chiefly  of  plagioclasb  (andbsinb  about  Ab5An3)  in 
idiomorphs  of  prismatic  habit,  sometimes  fresh,  but  mostly  decom- 
posed aad  containing  caldte  and  chlorite;  orthoclase  is  less 
abundant,  usually  somewhat  decomposed;  occasional  pseudomorphs 
occur  after  hornblbkdBi  with  the  characteristic  outlines  marked 
out  in  iron  oxide,  and  the  interior  filled  with  caldte  and  chlorite ; 
grains  of  icAGNSTm  are  fairly  numerous.  The  interstitial  cement 
consists  of  CHLORiTB ;  in  the  larger  interstices  the  interior  is  a  bluish 
green,  with^  the  usual  steel-blue  interference-colours,  and  often 
spherulitic,  whilst  the  margin  or  lining  is  a  yellow-green.  CALcm 
is  also  abundant,  usually  associated  with  the  chlorite,  and  filling 
small  cavities. 

*  The  rock  is  purple,  compact,  and  quartzose,  with  a  dull  lustre  on 
the  joint-planes ;  the  cut  surface  shows  that  the  quarts  has  a  greenish 
tinge,  with  interstitial,  purple  hematite,  and  grains  of  lunonite. 
Under  the  microscope,  the  rock  is  seen  to  consist  almost  entirely 
of  QUA&TZ  in  roughly  parallel  zones  of  fine  mosaics  and  larger  grains, 
/./.,  the  laiger  grains  are  often  surrounded  by  minute,  crushed  up, 
interlocking  granules;  it  endoses  very  fine  dust  and  minute  fluid- 
indusions.  Hekatitv  is  present  in  abundant  strings  and  irregular 
networks,  with  iCAaNKTrrs  in  its  meshes,  and  with  umonftb  in  laiger, 
but  less  firequent  grains  than  the  magnetite.  The  magnetite  has 
apparently  been  altered  into  hematite.  Patches  of  sericitic  wHtrs 
MICA  are  associated  with  the  hematite,  occurring  in  scales  so  minute 
that  no  deavages  are  distinctly  visjUe. 

t  The  rode  consbts  of  angular,  greenish  or  piuple,  quartzoee 
fragments,  with  finely  disseminated  hematite,  which  gives  the  rode  its 
dull  purple  colour  ,as  in  die  quartzite.    Under  the  microscope,  the 


Erjmrum  to  Tifid»fi(m^  an 

probable  that  Tashkdim  is  situated  on  the  line  of 
a  considerable  fault;  it  is  abo  noteworthy  in  this 
respect  that  I  did  not  meet  again  with  tiie  inter- 
bedded  limestones  until  about  halfway  between 
Sumelas  and  Jevizlik  (vide  infra). 

A  steep  ascent  brought  us  to  the  high-downs  of 
Tashkopri  Yaila.  Here  a  slight  break  in  the  clouds 
^ave  us  a  sight  of  a  magnificent  bluff  on  the  other 
side  of  the  valley,  to  the  S.E.;  the  cliff  showed 
several  thick  lava-flows,  lenticular  in  sedtion,  piled 
one  above  the  other.  The  mist  rapidly  becsime 
winter  and  more  opaque,  only  allowing  us  to  see 
about  lo  feet  ahead.  Rugged  masses  of  green 
porphyrite,  often  greatly  decomposed,  projeAed  here 
and  there  out  of  the  grassy  mountain-side. 

After  crossing  a  rocky  col  (8292  feet),  we  emerged 
upon  the  broad  downs  of  the  dioritic  Kazikly  Dagh. 
Tlie  solid  rock  seldom  appeared  to  view,  for  it  readily 
decomposes  into  a  brown  loam,  forming  rich  pastures, 
with  occasional  peat-mosses  in  the  hollows.  Plates 
of  biotite,  up  to  an  inch  in  diameter,  were  still 
visible  in  the  decayed  rock  (No.  206).  A  little  further 
north  the  rock  was  finer  in  grain  but  fresher;  it  is 
a  quartz-mica-diorite  with  augite  (No.  207),*^  rather 

fingments  ar!e  seen  to  show  the  same  characters  as  the  last  rock, 
but  the  quarts  is  even  in  a  finer  state  of  division,  and  the  mica 
is  a  little  more  alnmdant,  often  filling  cracks. 

*  The  rock  is  holocrystalline,  the  numerous  striated  felspars  (up 
to  5  mm.),  together  with  subordinate,  pale  pink  orthodase  and  a 
little  quartz,  forming  about  half  the  mass ;  hornblende  crystals  (up 
to  3  mm.)  and  biotite  flakes  ^up  to  6  mm.)  are  associated  together. 
The  rock  is  slighdj  magnetic  (s^).  Under  the  microscope,  the 
dominant  felspar  is  flaoioclass  (LABRADoarni  about  AbiAni)  in 
large  idiomorphs,  mostly  of  prismatic  habit  >  sometimes  dusky  with 
pale  brown,  dusty  decomposition-products  and  frequent  specks  of 
white  mica;  peridine  twinning  is  sometimes  visible  in  addition  to  the 
albite  twinxdng,  which  is  often  very  fine ;  sometimes  xoned ;  grains  of 


252  Geology  of  Armenia 

similar  in  diaradler  to  the  diorite  near  Mezereh  Khan 
(No.  193,  p.  242). 

On  the  north  side  of  this  dioritic  zone  of  the 
Kazikly  Dagh — the  true  axis  of  the  Pontic  Range — 
ive  commenced  our  descent  into  the  Meiriman  Dereh. 
The  contrast  was  stupendous.  The  valley  suddenly 
fell  away  thousands  of  feet  below  us,  its  precipices 
seamed  by  torrents,  and  the  roar  of  the  rushing 
waters  was  almost  deafening.  The  remarkable  dif- 
ference in  level  of  about  5000  feet  in  4  miles  (from 
the  edge  of  the  Kazikly  plateau  to  the  bed  of  the 
river  at  Sumelas)  means  an  approximate  gradient  of  i 
in  4,  which  is  amply  sufficient  to  explain  the  vast 
erosive  power  of  the  torrent. 

On  reaching  the  rhododendron  zone  the  majestic 
crag  of  Sumelas  came  into  view.  Here  the  quartz- 
mica- diorite  (No.  208)*  had  become  finer  in  grain. 

^reen  hornblende  are  occasionally  enclosed.  ^  Orthoclask  is  present 
in  allotriomorphic  grains,  often  quite  turbid  with  brown  decomposition 
products,  and  sometimes  intergrown  with  quartz.  Auorrs  occurs 
in  allotriomorphic  grains,  pale  green  and  malaoolitic ;  when  enclosed 
in  biotite  or  quartz  it  has  escaped  alteration,  but  it  has  for  the 
most  part  been  altered  into  ubalitic  bobnbuende;  even  when  the 
hornblende  is  compact  a  core  of  augite  frequently  remains^  or  a 
transition  is  noticeable  in  the  hornblende  from  a  compact  centre 
to  a  reedy  periphery ;  the  pleochroism  of  the  hornblende  is :  a — pale 
straw  with  a  greenish  tinge.  ^ — sap-green,  r— dark  ^een.  Biotitb 
is  somewhat  less  abundant  than  the  hornblende,  with  which  it  is 
intexgrown.  A  fair  amount  of  magnktitb  is  associated  with  the  last  s 
minerals.  Apatits  is  accessory.  Quartz  occurs  in  interstitial 
grains,  not  very  abundant;  containing  numerous,  small  fluid* 
inclusions,  often  with  a  bubble.  The  order  of  crystallization  seems  to 
have  been:  i.  Apatite  and  magnetite;  2.  Augite  and  biotite;  3. 
Plagioclase ;  4.  Orthoclase  and  quartz. 

*  The  rock  shows  numerous,  greenish  grey,  fine-grained  areas 
(about  12  mm.),  lying  in  a  coarser  grained  aggregate  composed  of 
striated  felspars  (up  to  3  mm.),  often  stained  a  pale  brown  by  iron 
oxide,  and  lustrous  hornblende  prisms  (up  to  6  mm.),  with  a  less 
amount  of  interstitial  quartz.    Under  the  microscope,  the  dominant 


Erxerum  to  TrMzond  353 

It  varies  somewhat  in  texture — fine-grained  segre- 
eation-nodules  lying  in  a  coarser  matrix.  Still 
further  down  the  valley  the  diorite  (No.  209^  *  shows 
essentially  the  same  composition,  but  is  still  finer  in 
grain.  At  this  point  we  were  just  above  the  zone 
of  deciduous  trees. 

We  had  kept  to  the  right  bank  until  we  were 
nearly  opposite  the  monastery  <)f  Sumelas ;  here  the 
rock,  which  was  greatly  decomposed,  appeared  to 
be  transitional  between  a  very  fine-grained  diorite 
and  a  compact  porphyrite.  The  river,  which  had 
already  carved  out  a  gorge  of  considerable  depth,  now 

fdspar  is  plagioclasx  (andesinb  about  Ab3An2)  in  idiomorphs  with 
a  tendency  to  a  tabular  habit ;  occasionally  a  little  turbid  with  dusty 
decomposition-products,  and  with  minute  flecks  of  white  mica ;  faintly 
coned ;  sometimes  with  a  peripheral  line  of  dust ;  encloses  hornblende 
and  epidote.  Orthoculse  is  subordinate  and  somewhat  turbid. 
Hornblende  is  frequent  in  large,  green  idiomorphs,  often  twinned, 
and  enclosing  apatite  and  magnetite ;  pleochroism :  a — pale  yellow, 
^ — olive  green,  c — ^blue-green ;  it  also  occurs  in  small,  allotriomorphic 
grains,  but  it  is  always  idiomorphic  to  quartz.  Numerous  small  plates 
and  flakes  of  green  BioirrE  occur  in  the  finer-grained  areas ;  some- 
times quite  altered  to  chlorite  with  qpidote;  pleochrobm  from  pale 
Straw  to  deep  green.  Quartz  is  frequent,  with  small  fluid-inclusions ; 
the  larger  grains  are  clear  centrally,  enclosing  large  idiomorphs  of 
hornblende,  but  show  an  irregularly  pegmatitic  margin,  intergrown 
with  dusty  orthodase  and  hornblende ;  some  of  the  fluid-inclusions 
^dose  a  cube  of  salt  as  well  as  a  bubble.  MAONBrrrE  is  accessory. 
Secondary  epidote  in  pale  yellow  grains  occurs  both  separately  and 
filling  a  vein.  The  finer-grained  areas  merely  reproduce  the  above 
structiure  on  a  smaller  scale,  but  the  plagioclase  is  more  fre<|uently 
idiomorphic  and  of  prismatic  habit ;  in  fact  these  parts  show  diabaslc 
^aracters. 

*  The  rock  is  remarkably  similar  in  almost  all  particulars,  both  in 
the  characteristics  and  arrangement  of  the  constituents  to  die  finer- 
grained  portions  of  the  4ast  rode ;  none  of  the  constituents  exceed 
I  mm.  The  plagioclase  (aiidbsini  about  Ab4An3)  is  entirely 
idiomorphic  and  of  prismatic  habit.  The  hornblende  is  very 
abundant,  nearly  equal  in  amount  to  the  felspar.  Micrographic 
Inteigrowths  of  orthoclase  and  quarte  also  occur. 


a54  Geology  of  Armenia 

plunged  into  a  vast,  elliptical  amphitheatre,  walled 
in  by  precipitous  heights,  and  clothed  in  dense  forest. 
With  some  difficulty  we  descended  down  the  steep, 
slippery  bank  to  the  level  of  the  torrent  (3680  feet). 

A  long,  zigzag  path  up  the  equally  steep,  left  bank 
brought  us  to  the  ancient  monastery  of  Sumelas  (4457 
feet),  built  up  against  the  precipitous  face  of  the  lofty 
crag,  which  partly  overhangs  it.  The  rock  is  a  dark 
green  augite-porphyrite,  inclined  45°  N.  by  W. 

On  resuming  the  journey  to  Trebizond,  I  descended 
again  to  the  river,  through  the  most  luxuriant 
vegetation,  and  crossed  over  to  the  right  bank.  The 
augite-porphyrite  still  preserved  much  the  same 
inclination,  30''  N.  by  W.,  but  it  was  soon  interrupted 
by  an  intrusive  outcrop  of  a  granitite  (No.  210).* 
Several  boulders  of  similar,  but  finer-grained  rock  lie 
in  the  river-bed,  and  have  apparently  fallen  from  the 
heights  above.  This  outcrop  seemed  to  be  a  large, 
intrusive  boss. 

After  leaving  the  granitic  area  I  ejitered  again  into 
the  region  of  altered  augite-  and  homblende-andesites, 

*  The  rock  is  very  pale  brown,  with  white  or  pale  pinkish  brown 
felspars  (up  to  3  mm.),  and  sparing  flakes  of  biotite  (up  to  2  mm.),  in 
a  matrix  of  quartz  granules.  Under  the  microscope,  orthoolasi 
is  next  in  abundance  to  the  quartz,  with  which  it  often  forms  grano- 
phyric  intezgrowths ;  allotriomorphic ;  turbid  with  brownish,  dusty 
decomposition-products,  or  else  clearer,  but  with  abundant  tiny  scales 
of  white  mica  and  granules  of  epidote.  Plagioclasb  (andbsimb 
about  AbsAnj)  occurs  only  sparingly,  in  rather  laj:ge  crystals  often 
approaching  idiomorphism,  sometimes  zoned ;  very  fresh  and  free 
from  inclusions.  Green  biotitb  is  present  in  small,  rath^  sparing 
flakes,  partly  altered  to  chlorite  and  epidote,  as  in  the  last  two  rocks ; 
occasionally  a  small  flake  of  unaltered,  brown  biotite  can  be  detected 
with  a  high  power  (1/4O  enclosed  in  a  qoartis  grain.  Quabtz  is  the 
most  abundant  constituent,  occurring  in  interstitial  grains  with  minntt 
fluid-inclusions  (often  with  a  bubble).  Magnstitx  and  apahtx  are 
accessory.  The  order  of  crystallization  seems  to  have  been : — x. 
Apatite  and  magnetite;  a.  Biotite;  3.  Plagiodase;  4.  Orthodase 
and  quartz. 


Erzerum  to  Trebisumd  255 

but  from  this  point  onwards  I  observed  niunerous 
bands  of  marmorized  limestone  and  also  of  ^ale, 
ibterbedded  with  the  volcanic  rocks  and  evidently 
contemporaneous  (as  near  Chorak  Khan  on  the  south 
side  of  the  range,  p.  249,  and  in  the  Machka  Derdi,  to 
the  west  of  my  route,  pp.  21,  22,  where  I  have  stated 
my  reasons  for  considering  these  strata  to  be  of 
Eocene  age).  On  proceeding  down  the  Meiriman 
Dereh  it  was  clear  that  I  encountered  successively 
higher  beds,  for  thev  were  all  inclined  about  30®  N. 
by  W.  The  first  (and  lowest)  of  these  beds  is  a 
white,  massive,  saccharoidal  marble,  with  the  above 
dip,  interbedded  in  green  or  purple  augite-andesite, 
which  forms  pidhu^que  qrags,  weathering  to  browns 
and  greys.  This  was  the  only  instance  in  which 
the  marble  of  this  series  was  pure  white,  probably 
owing  to  the  proximity  of  the  granite  bc6S.  A  layer 
of  shales,  about  12  feet  thick,  dipping  35*^  N.  by  W., 
was  the  next  sedimentary  bed  to  be  noticed;  they 
had  been  baked  right  through  to  a  purple  colour 
by  the  overlying  andesite. 

After  crossing  over  to  the  left  bank  by  a  wooden 
roofed  bridge,  I  observed  a  band  of  black  marble, 
dipping  40"*  N.  by  W.,  interbedded  in  compaA,  green 
andesite.  Here  the  narrow  track  left  the  river-bed, 
and  ascended  to  a  wayside  shrine,  called  Merdosh, 
situated  on  another  (higher)  band  of  black  marble 
(No.  211).  The  microscope  revealed  numerous,  small 
Foraminifera,  chiefly  MilioUna  and  Textularia^  some- 
what similar  to  those  in  the  limestones  of  the  Machka 
Dereh  (p.  21)  and  the  Kharshut  valley  (p.  26),  but 
here  they  are  much  dwarfed,  probably  owing  to  the 
unfavourable  conditions  of  submarine  volcanic  erup- 
tions (cp.  p.  244). 

Then  I  descended  again  to  the  river,  passing 
another  layer  of  similar  marble,  dipping  25®  N.  by  W. 


256  Geology  of  Armenia 

These  inteibedded  sediments  are,  however,  infrequent, 
and  of  slight  thickness  compared  to  the  andesites. 
The  lavas  probably  flowed  along  the  sea-bottom  at 
some  distance  from  the  coast,  for  I  could  see  no  trace 
of  sandstones  or  conglomerates.  The  cliffs  are  mainly 
composed  of  the  altered  hornblende*pyroxene-ande« 
site  (No.  2ia)  ;*  near  this  locality  it  begins  to  assume 
a  columnar  strudture.  Another  band  of  marble, 
about  50  feet  thick,  dipping  40**  N.N.W.,  interrupted 
the  uniformity  of  the  green  andesite,  crowded  with 
white  felspars. 

About  a  mile  above  the  confluence  of  the  Meiriman 
and  Larkhana  rivers  numerous  dikes  seamed  the  cli& 
from  top  to  bottom ;  they  consist  of  a  more  compadfc 
and  finer-grained  rock,  but  time  did  not  permit  of  an 
examination.    The  andesite  in  their  neighbourhood 

*  The  rock  is  dark  green  and  compact,  with  abundant  idiite 
striated  felspars  (ap  to  3  mm.^,  eadi  with  a  dark  green,  chloritic  core, 
and  a  less  number  of  chlontic  pseadomorphs  (up  to  s  mm.)  after 
hornblende  and  pfrroxene.  Under  the  microscope,  the  plagioglabk 
(andesikb  about  AbjAni^  occurs  in  fairly  laige  and  numerous  idio- 
morphs,  mostly  of  prismatic  habit,  sometimes  in  groups ;  often  roned  ; 
twinning  much  obscured  by  dusty,  brown  decomposition-products; 
the  central  part  is  full  of  inclusions  of  bluish  green  chlorite,  often 
ranged  along  the  zonal  planes,  and  it  often  polarizes  in  a  different 
sense  from  Sie  peripheral,  relatively  clearer  felspar.  Pseudomorphs 
after  hornblkndb  are  fairly  frequent,  with  sharp  crystal-outlines; 
it  is  seldom  that  any  scraps  of  the  original,  brown  hornblende  still 
remain,  showing  pleochroism  from  a  Veqr  pale  straw  to  an  umber  with 
a  greenish  tinge ;  it  has  for  the  most  part  been  replaced  by  caldte, 
chlorite,  and  brownish  leucoxene;  it  encloses  apatite.  Several 
pseudomorphs  also  occur,  which  by  their  outlines  point  to  a  pyro- 
XBNB ;  they  consist  entirely  of  chlorite  (sphenilitic  internally,  but  with 
a  radially  fibrous  margin)  and  irregular  grains  of  epidote.  Magnbtitb 
(partly  altered  to  hematite),  iLKBNrra  (partly  altered  to  leucoxene), 
and  APATITE  are  accessory.  The  oroundmass  is  a  micropoecilitic  felt 
of  oligodase  microlites,  in  fluidal  arrangement,  together  with  abunduit 
chlorite,  epidote,  and  magnetite  grlmules ;  sxnall  grains  of  quartz  are 
filirij  frequent.  Sometimes  the  groundmass  approaches  a  finely  ortho* 
phyric  appearance.    Calcitb  is  present  in  cavities. 


Erxerum  to  Trdnzond  257 

had  been  much  disturbed  and  veined.  I  crossed 
the  stream  by  another  covered  bridge,  and  the  path 
thenceforward  kept  to  the  right  bank.  Below  the 
confluence  a  black,  unfossiliferous  limestone  (No.  213), 
containing  iron  pyrites,  occurs  associated  with  shales. 
The  strata  dip  yf  N.  by  W.,  and  form  a  zone  of 
considerable  thickness,  iiiterbedded  in  the  green  an- 
desite.  Further  down  the  valley  another  series  of 
shales  was  visible,  partially  baked  to  a  purple  colour. 

The  volcanic  rock  became  more  regularly  columnar 
as  I  approached  Jevizlik,  and  the  lofty  river-cli&  well 
display  the  ancient  lava-flows,  piled  one  above  the 
other  and  lenticular  in  sedlion.  The  columns  are 
nearly  vertical,  sometimes  of  considerable  height,  but 
only  about  8  inches  in  thickness.  Large  augite 
crystals  were  often  conspicuous,  but  the  lavas  near 
Jevizlik  are  usually  much  decomposed. 

The  few  remaining  observations  which  I  was  able 
to  make  (owing  to  a  violent  thunderstorm  and  the 
approach  of  night)  have  been  incorporated  in  my 
account  of  the  outward  journey  between  Trebizond 
and  Jevizlik  (pp,  19,  20). 


8 


PART  n 


THE 

GEOLOGICAL    RECORD 

OF 

ARMENIA 


CHAPTER   XV 

THE  PRE-DEVONIAN  ROCKS  OF  ARBCENIA 

L     TBS  VOBIHBltN  MXD  XASIUN  tXOIONS. 

TkE  Zitula  (or  Meschic)  dome  forms  the  conneding 
link  between  the  acutely  folded  Central  Caucasus 
on  the  one  hand,  and  the  Thrialetic  Range  (the 
northern  border-range  erf  Armenia)  on  the  other.  The 
importance  of  this  intervening  dome  is  evidenced 
by  the  facft  that  while  the  folds  of  the  Caucasus 
have  been  bent  over  to  the  south,  those  of  the 
Thrialetic  Range  are  curved  over  to  the  north.  It 
seems  therefore  evident  that  the  Zirula  dome  has 
played  the  part  of  a  horstj  offering  an  inert  resistance 
to  the  tangential  forces,  which  first  of  all  adted  (as 
already  indicated,  pp.  9,  10)  from  south  to  north  upon 
the  Armenian  ranges  in  the  Lower  Permian,  ante- 
Tithonian  and  Middle  Miocene  periods,  and  then 
adled  upon  the  Caucasus  from  north  to  south  at  a 
period  immediately  subsequent  to  the  Upper  Miocene. 
It  is  therefore  not  surprising  to  find*  that   the 

^'^  Zirula  dome  is  composed  of  granites,  gneisses,  and  crystalline 
schists.  It  is  elliptical  in  plan,  tbe  long  axis  of  the  ellipse  running 
west  and  east.  It  is  bounded  on  the  west  and  south-west  by  the 
synclinal  basin  of  the  Upper  Rvizila  and  Cherimela,   on  the  east 

*  Foumier,  Dtscriptian  gtd.  du  Catuau  tmtral,  256. 


262  Geology  of  Armenia 

by  the  syndiiul  bAsin  of  die  Rim*  and  on  tlie  iMXth  by  the 
Syikh-Labcrta  tmassif.  At  one  point — between  Varvarin  and  Tapa- 
h  meets  tangentially  the  northcm  folds  of  the  littk  Csnosos," 
ia.^  the  Thrialetic  Range. 


Fournier  *  also  states  that 

**it  is  a  dome  which  emeiged  at  a  very  ancient  period.t   Very 
quaitzose  gneisses,  ^mnUtjiig  mica-schists^  appear  in  the  northeni 
part  of  the  dome;   near  Korbo-uli  they  are  intimately  penetrated 
by  granites,  granulites,  amphibolites,  and  syenites ;   some  paxts  pass 
into  protogine.   .   .    Near  Jiatura,  in  the  Talley  of  die  Kvirila,  micro- 
granulites  with  veins  of  quarts  occur  on  the  N.N.W.  border;  some 
parts  present  the  appearance  of  true  granophyres.     In  the  Tslley 
of  the  Kvirila,  in  that  of  the  Zirula,  near  Poni  and  neir  Mardisi 
the   granites   and  granulites   are   traversed  by  numerous    veins  of 
melaphyres,  diabases,  and  poiphyritcs/' 

In  addition,  FoumierJ   points  out  that   the  an- 
tiquity of  the  Zirula  dome  is  shown  by  the  fad;  that 

*'  all  the  formations,  from  the  Lias  onwards,  contain,  where  they  abat 
on  the  Zirula  dome,  beds  of  conglomerate  formed  from  its  rocks,  and 
beds  of  lignite,  which  become  thicker  as  the  dome  b  approached. 
Certain  marly  or  argillaceous  formations,  on  nearing  the  dome,  pass 
into  sandstones,  gravels,  and  arkoses,  arising  from  the  alteration 
of  the  granitic  rocks.  The  Ostraaa-heds  of  the  Neocomian,  Aptian, 
Cenoroanian,  and  even  of  the  Chalk,  and  the  jReptuma-heds  oif 
the  Urgonian  all  attain  their  maximum  thickness  round  this  dome. 
The  beds  of  the  Aptian  and  the  Chalk,  which  at  Kutais  still  contain 
Cephalopods,  enclose  on  the  N.£.  flank  of  the  dome  Ostracea^ 
LUhophagi^  and  even  PaiellcR,^* 

Ancient  crystalline  rocks  underlie  the  Cretaceous  of 
the  Thrialetic  Mountains,§  perhaps  in  still  hidden 

•  Op.  cit.  83. 

t  Abich  iVergleicJfende  geol.  Grundsuge^  etc.,  55)  had  previously 
mentioned  tnat  the  granitic  rocks  of  the  Meschic  (or  Meskian)  Mts. 
are  the  oldest  in  Armenia. 

}  Op.  cit.  as7. 

%  Abich,  MUth.  aus  dem  Kaiikasus^  Verb.  k.-k.  geol.  Reichsanst., 
1877,  30. 


'  ■»  A 


r     •  •»   • 


I? 


!•■• 


Pre-Devonian  Hocks  263 

connedtion  with  the  Zirula  dome ;  they  also  occur  in 
the  latitudinal  ranges  of  the  Somketian  Mountains.* 
In  the  valley  of  the  Upper  Khram  (or  Kzia)  they 
diredlly  underlie  Cenomanian,  and  consist  of  "  grani- 
tic gneiss,  mica-schist,  and  chlorite-schist."|  Pro- 
bably several  of  the  areas  of  crystalline  rocks  in  the 
eastern  border-ranges  of  Armenia  (the  so-called  Little 
Caucasus)  will  be  found  on  closer  examination  to 
consist  of  ancient  gneiss.  Abich:|:  indeed  gives  a 
ore-Devonian  age  for  the  "partly  confusedly  strati- 
ned,  partly  unstratified  ma^es,  forming  the  adlual 
crystalline  fimdament  of  the  mountain  range  which 
branches  out  to  the  N.E.  from  the  Alagaya  Dagh," 
t.e.,  the  masses  of  "  felspar-porphyry "  on  the  south 
side  of  the  Terter  valley,  in  the  North  Karabagh 
region.  The  "massive,  porphyritic  quartz-felspar 
rocks  at  the  bottom  of  the  deep  Uchdara"  (S.W. 
of  Shusha)  belong  to  "the  oldest  formations  of  the 
Shusha  mountains/*§  and  appear  to  be  of  similar 
antiquity. 

On  the  inner  (west)  side  of  the  eastern  border- 
ranges  metamorphic  rocks  occur  more  frequently, 
but  their  precise  age  is  doubtful.  On  the  north 
side  of  the  Aglagan  Dagh  (to  the  north-east  of 
Alexandropol)   Abich||    found    that 

**  a  very  extensive  and  thick  fonnation  of  crystalline  limestone  like 

*  Abich,  AfiUA.  aus  dem  Kdukasm^  Verh.  k.-k.  geol.  Reichsanst 
1877,  30. 

t  Md.\  his  GeoL  F&nch,  in  dm  katti.  Ldndem^  ii  125,  and  hb 
Ver^teuhende  gad.  Grundzuge^  55 ;  in  his  Mangasuru^  Bull.  phys. 
math.  Acad.  Imp.  Sci.  St  Petersbiug,  xvi.  316,  he  specifies  the  ro^ 
as  "  granite,  quartz-poiphyry,  eurite,  and  duk,  schistose  rocks.'' 

X  Geol.  Beobachi.  ww.  Kur  6*  Araxes^  126. 

§  Ibid.  126. 

II  Ibid.  6. 


264  Geology  of  Armenia 

Carrara  marble,  and  a  considerable  formation  of  usefiil,  although 
somewhat  odcareous,  roofing-slates.  The  deep  ravine  of  the  Ssurp 
river  (a  right  tributary  of  the  Kamenka-chai)  well  displays  the  slates, 
which  Uke  a  fundamental  part  in  the  composition  of  Aglagan  (9833 
feet).  The  marble  is  here  associated  with  extremely  thick  masses 
of  serpentine  with  its  usual  companion-rocks ;  lead  and  copper  ores 
occur  abo  in  this  r^on."  These  metamorphic  rocks  underly 
beds  of  Turonian  age. 

Further    to    the    south    Abich*    discovered    that 

"homblendic,  chloritic,  talcose  and  micaceous  schists,  with  ser- 
pentine and  gabbro,  occasion^y  occur  in  disconnected,  lateral  zones 
of  the  central  masses  of  the  South  Pambak  Mountains ;"  this  rather 
general  statement  is  amplified  by  his  statement  t  that  in  the  Miiskhana 
valley  (Darachichek  region) ''  a  broad  zone  is  occupied  by  a  formation 
of  chlorite-  and  mica-schists,  and  quartzite  associated  with  white 
marbles,  serpentines  and  schistose  gabbros.  Granite  forms  the  aids 
of  the  S.£.  spur  of  the  Pambak,  between  the  Abaran  and  Miiskhana 
valleys.  .  .  On  descending  to  the  S.W.,  from  the  obsidian  heights  of 
Sinyerly  into  the  Akhveran  valley,  the  zone  of  serpentines  and  schists 
appears  again  in  increasing  development,  together  with  schistose 
marbles  and  calcareo-chloritic  rocks,  as  far  as  the  syenite-porphyry, 
hornblende-diorite,  and  felspar-porphyry,  wl^ch  prevail  up  to  Arsa- 
kent,  at  the  entry  of  the  Zanga  valley." 

Oa  the  west  side  of  the  Darachichek  Mountains, 
facing  Alagoz,  Abich  has  indicated  on  his  large  map  ± 
a  zone  of  "  chloritic,  talcose,  and  horablendic  schists/ 

To  the  south  of  Lake  Gokcha  (Sevanga),  in  the 
valley  of  the  Daralagoz  Arpa-chai,  the  Devonistn 
strata,  according  to  ADich,§  overlie  **dark,  chloritic 
and  aphanitic  schists  in  connecStion  with  dioritic 
schists,  which  probably  belong  to  the  Silurian." 

Similar  schists  are  mapped  again  by  Abich  |1  in  the 
South  Karabagh  districft,  w>.,  in  the  headwaters  of 

*  Geal.  Beobacfit.  vw.  Kur  6f  Araxes^  13. 

t  IHd.  15,  16. 

%  Geol.  Rnch.  in  dm  kamk,  LSniertu  Atlas. 

{  Ibid,  ii.  25. 

II  Ilnd.  Atlas. 


Pre-Devoman  Rocks  265 

the  Okhchi^  river,  on  the  east  side  of  the  granitic 
Kapujik.  It  is  probable  that  this  series  is  continued 
across  the  Araxes  to  the  S.E.  in  the  contorted  and 
faulted  clay-slates,  which  extend  between  Iri  and 
Aliar  in  the  Karadagh  region.* 

Metamorphic  rocks  occur  here  and  there  in  the 
Talish  mountains — the  eastern  border- ranges  of  Azer- 
baijan. According  to  Felkner  "f  calcareous  clay-slates 
and  quartzose  schists  often  rise  up  on  the  east  side 
of  the  watershed;  their  strike  is  on  the  average 
N.N.W.-S.S.E.,  with  a  dip  of  15^-20^  Schists  with 
an  easterly  strike  [?  S.S.E.]  and  gentle  dip  were 
noticed  by  Raddef  in  ascending  the  Astara  river, 
whilst  at  Kashbinsk,  on  the  Russo- Persian  frontier, 
•*  the  schists  dip  steeply." 

XL      THS  PONTIC  REGION. 

Early  Palaeozoic  beds  seem  to  occur  in  the  Lower 
Chorokh  region  in  the  neighbourhood  of  Artvin.  In 
proceeding  down  the  Ardanuch  valley,  Abich  §  noted 
at  the  confluence  of  the  Imer-khevi  the  occurrence  of 

"  homogeneous,  bluish  black,  pelitic  rocks,  much  recalling  Palaeozoic 
slates,  standing  in  the  closest  connection  with  syenite,  quartzite,  and 
felsite.  ...  In  the  lowest  r^on  of  the  valley  at  Artvin  bluish  grey 
clay-slates  rise  up,  presenting  the  appearance  of  Palaeozoic  rocks. 
The  slates  are  characterized  by  frequent  inclusions  of  iron  pyrites, 
thus  recalling  the  slates  of  die  Terek  valley." 

Batsevitch||  has  confirmed  the  Palaeozoic  age  of 
these  strata,  and  has  prolonged  their  outcrop  for 
some  distance  to  the  S.W.  along  the  strike.    It  is  not 

*  Grewmgk,  Giognost,  &•  orogr,  VerMUn.  nordl.  Fersicns^  47. 

t  Russian  Journal  of  MineSf  1837,  342-363.     [In  Russian]. 

X  Talysch  &  sei$u  Bewohner^  137. 

§  GeoL  Forsch.  in  den  kauk.  Laniem^  iL  184. 

II  GeoL  Researtkis  in  the  Batum  &  Artoin  Districts^  Mat  Gto\ 
Cauc.  Tifiis,  1887,  Ser.  a,  i.  77-79.    [In  Russian]. 


266  Geology  of  Armenia 

improbable  that  the  same  series  of  rocks  extends 
along  the  inner  border  of  the  Pontic  Range;  two 
out(  lops  of  rather  similar  strata  were  traversed  by 
Koch^  when  he  crossed  the  range  from  Pertekrek 
to  Atina.  He  travelled  up  the  Pertekrek  valley 
almost  to  its  commencement ;  a  grey  limestone  (pro- 
bably Cretaceous)  gave  place  to  dark  clay-slate 
(causing  a  narrowing  of  the  vallej'),  broken  through 
by  porphyries.  Turning  to  the  S.W.,  the  clay-slate 
prevailed.  The  dirty  grey  limestone  appeared  again 
in  the  valley  of  Khingemek,  opposite  Kiskin.  After 
traversing  this  massive  limestone  as  far  as  Khodudjur, 
Koch  proceeded  up   the  main  valley  to  the   N.W. 

"  The  rock  at  Khodudjur  [c.  8000  feet]  was  a  grey,  often  red  limc- 
stone,  probably  belonging  to  tne  Primary  rocks.  It  had  been  greatly 
fractured  and  contorted.  .  .  Iron  pyrites  occurred  in  cubes  in  the 
limestone.  .  .  The  lofty,  syenitic  Khachkar  Dagh — ^the  most  im- 
portant nlountain  in  the  Pontic  Range — lay  close  at  hand,  to  the 
north-west,  t" 

Turning  eastwards  to  the  Kur-Chorokh  watershed, 
similar  metamorphic  rocks  appear  to  form  the  chief 
heights,  flanked  by  Cretaceous  beds,  in  the  western 
part  of  the  Goleh  distridl  of  the  Upper  Kur.  Here,  in 
travelling  from  Ardahan  to  Olti,  Koch  J  found  that 
the  heights  consisted  of  clay-slates,  which  began  to 
narrow  the  valley  of  one  of  the  western  branches 
of  the  Kur,  S.W.  of  Dort-kilissa.  Old  silver  mines 
occur  among  these  strata  at  Giimush-djur,  and  there 
are  also  disused  gold  workings  at  Altiin-bulak  (7 
hours  east  of  Giimiish-djur,  and  5  hours  S.E.  of  Urut). 

III.      THE  TAURIC  BBGIOH. 

Just  as  the  Zirula  dome,  to  the  north  of  Armenia, 

*  J^nu  impontischen  GMrge^  iL.  83I 
t  Ibid.  ii.  ^. 
X  lUd.  ii.  231. 


Pre-Devonian  Rocks  267 

has  adled  as  a  horst^  stemming  the  crustal  folds,  first 
from  the  south  and  afterwards  from  the  north  (p.  261) 
— so  in  the  south  of  Armenia  a  large  block  of  ancient 
metamorphic  rocks  has  (it  seems  to  me)  resisted  the 
mountain-folding,  and  now  forms  the  axis  of  the 
Taurus,  south  of  Lake  Van  (see  ray  Strudlural  Map  of 
Armenia).  When  these  schists,  slates,  and  marbles 
were  still  plastic  enough  to  be  thrown  into  folds,  they 
had  a  N.E.-S.W>  strike  impressed  upon  them.  This 
probably  occurred  at  the  close  of  the  Carboniferous 
period,  for  this  orientation  is  that  of  the  original 
strike  of  the  Devonian  and  Carboniferous  in  the 
region  of  Ararat  and  the  Daralagoz,^  and  also  in 
the  AntitauruSjf  and  in  both  these  localities  they  rest 
on  schists  (pp.  no,  264,  and  Chapter  XVI).  Since 
that  time  the  Tauric  schists  have  resisted  all  later 
folding,  and  when  the  Caucasus  in  post- Miocene 
times  received  its  final  folding  by  a  tremendous 
pressure  from  the  N.E.,  the  result — so  far  as  the 
Armenian  area  was  concerned — ^was  to  produce  a 
series  of  N.W.-S.E.  fradlures  (pp.  10,  no),  and  in 
particular  the  great  fault,  with  this  direction,  which 
boimds  the  plain  of  Mush  and  Lake  Van  on  the 
south,  cutting  obliquely  across  the  oW  strike  of  the 
Tauric  horst.  Hence  the  S.W.-N.E.  ridges  appear  in 
^helon  along  the  line  of  this  fault. 

The  isolated  little  peninsula  of  Tadvan  (pp.  100- 
102)  remains  as  the  outlying  summit  of  a  mountain- 
crest  to  bear  witness  to  this  comparatively  recent 
sinking-down  of  the  Tauric  schists  beneath  Lake 
Van.  The  promontory  of  Tadvan  consists  of  two 
parts :  the  castle-hiR  is  the  larger,  and  is  conneAed 
by  a   narrow,   sandy   isthmus  with  a    much    lower 

•  Abich,  V^r^chende geoL  Grundsuge^  ax,  81. 
t  Tchihatchefif,  Asie  Mineurif  Ghlogie^  679-699. 


1 


268  Geology  of  Armenia 

and  smaller  mass,  jutting  out  into  deep  water.  An 
anticlinal  axis  nms  obliquely  through  tiie  castle-hill, 
and  a  synclinal  axis  through  the  small,  outer  pen- 
insula (to  E.S.E.),  both  running  W.  5"^  S -E.s^'N. 

The  downward  succession  of  the  castle-hill  on  the 
south  side,  near  the  rude  jetty,  down  to  the  water's 
edge,  is  as  follows : — 

I.  Dark  grey  marUe,  ooartdj  foliated,  dqppiog  so*  N.  s*  ^m 
and  full  of  veiiis  of  rhombohedral  ttldte.  It  formi  the 
summit  of  the  hill,  and  the  bj^ol  of  the  anticline..^  25  feet 

9.  Palequartz-mica.schist        ^  20  feet 

3.  Pale  green,  chloritic  mica-schist  in  thinner  lamims ;  quarts  in 

relativelj  large  folia  ^.  15  feet 

4.  Dark  grey,  foliated  marble,  becoming  more  massive  at  its 

base ;  with  white  veins  of  calcite  •••        ••«        .••^.  30  feet 

The  synclinal  axis  only  just  cuts  the  northern  shore 
of  the  smaller  hill.  A  low,  rugged  cliff  forms  the 
extremity  of  this  mass ;  it  consists  of  a  silvery, 
micaceous  limestone-schist  (described  on  p»  loi), 
dipping  35^  N.  5®  W.,  and  much  traversed  by  veins  of 
rhombohedral  calcite  (dog's  tooth  spar).  It  is  over- 
lain conformably  by  massive,  dark  grey  marble. 
Higher  beds  occur  on  the  north  side  of  the  hill,  which 
yields  altogether  the  following  succession  down  to  the 
water's  edge,  each  bed  being  about  10  feet  in 
thickness : — 

I.  Pale  mica-schist^  with  irregular  layers  and  '  eyes  *  of  quarts. 
9.  Fissile,  black  marble,  with  obscure  maridngs  perhaps  fuooidaL 

3.  Foliated,  grey  marble,  with  scattered  mica->flakes  along  tht 

foliation-plfljies ;  passing  into 

4.  Massive,  dark  grey  marble. 

5.  Silvery,  micaceous  limestone-schist  (described  on  p.  loi),  with 

veins  of  rhombohedral  calcite. 

Th^  rocks  present  a  slightly  different  characfler  from 
those  of  the  castle-hill,  and  probably  represoit  a 
higher  horizon,  if— as  it  seems  to  me — these  outliers 
at  Tadvan  of  the  Taurie  Heights  are  merely  the 
summits  of  blocks,  which  have  sunk  down  by  step* 


PrC'Dewmian  Ri>cks  1169 

faults  to  a  lower  level,  and  have  then  been  surrounded 
by  the  lavas  of  Nimrud  and  the  waters  of  Lake  Van. 

A  visit  to  the  Gdzel  Dereh,  at  the  S,W.  corner 
of  the  lake,  gave  me  a  further  insight  into  the 
structure  of  the  Tauric  Heights  (pp.  107,  108).  In 
company  with  Colonel  G.  S.  Elliot  I  ascended  a 
conspicuous  peak  {c.  8000  feet),  on  the  west  side  of  the 
valley.  My  observations  showed  the  following  sue- 
cession : — 

I.  Pure  white,  sacdianridal  tnai6le,  fonning  the  summit  of  the 

peak, 
a.  Brownish,  micaceous  quartste,  with  occasional  streaks  of 

mica-schist;  passing j^aduaDy into 

3.  Silvery  brown  mica^si^ist  (desorftied  On  p.   107),  occurring 

about  halfway  down  the  peakf  and  weathering  to  a  brown 
sand. 

4.  Grey  marble,  greatly  veined  with  caldte,  forming  the  lower 

slopes. 

5.  Mica-schist  of  the  river-bed  and  the  east  baak. 

In  our  ascent  we  crossed  the  axis  of  an  anticline, 
Nos.  4  &  5  dipping  80^  S.S.E.,  and  Nos.  i,  2,  &  3,  50^ 
N.N.W.,  but  exposures  were  few.  Looking  from  the 
summit,  these  rugged  heights  of  marble  and  schists 
were  seen  to  extend  southwards  to  the  horizoii  for 
fully  10  miles ;  another  anticlinal  ridge  occurs  a  little 
to  tiiie  south  of  this  peak,  with  a  dip  of  about  50^  to 
N.N.W.  and  S.S.E. 

This  metamorphic  zone  extends  westwards  to  form 
the  southern  boundary  of  the  plain  of  Mush.  Whilst 
the  highest  peaks  and  ridges  are  composed  of  barren, 
grey  or  white  marble,  the  foot-hills  consist  of  mica- 
schist.  Just  west  of  Tadvan  the  southern  part  of  the 
watershed  between  Lake  Van  and  the  Bitlis*chai 
is  formed  by  a  spur  of  mica-schist  (dipping  40^ 
N.N.W.),  extending  from  the  base  of  a  high  peak 
of  foliated,  grey  and  white,  micaceous  marble  above 
Urtab(pp.  109,  III). 


ayo  Geology  of  Armonia 

« 

Further  west,  at  Norshen  (p.  xog^i  I  found  that  the 
mountain-side  (next  to  the  plain)  is  composed  of 
mica-schist,  dipping  about  50^  N.N.W.,  with  seams  of 
quartz.  Further  south  the  higher,  barren  mountains 
are  clearly  composed  of  grey  and  white  marbles. 
Indeed  Ainsworth*  states  that  the  Zirzira  Dagh 
(south  of  Norshen)  is  a  '*  giant  limestone  precipice  on 
primary  schist/*  and  that  ^*  just  west  of  Khaskoi  a 
cape  of  limestone  juts  out  into  the  plain  *'  of  Mush. 

Still  further  west,  near  Mush,  Kochf  relates  that, 
although  he  was  unable  to  ascend  the  mountains 
south  of  Mush,  yet  so  far  as  he  went  *' mica-schist 
much  traversed  by  quartz-veins,  was  the  only  kind  of 
rock  composing  the  heights  on  the  south  side  of  the 
Mush  plain.*'  He  further  states  that  the  church  of 
Khardsor  (between  Arinj  and  Khoronkh,  west  of 
Mush)  is  built  of  mica-schist,  and  that  ^'the  plain 
is  strewn  with  much  gravel  of  this  rock/* 

These  schists  certainly  extend  as  far  as  the  41*^ 
meridian,  and  probably  form  the  continuous  axis 
of  the  Taurus  to  Palu  and  Lake  Goljik;  Kotschyt 
noticed  that  the  mountains  on  the  south  side  01 
the  Murad  at  Noreg  and  Angag  (S.W.  of  Boghlan, 
which  lies  about  10  miles  W.  of  Surp  Karapet) 
consist  of  schists,  rising  up  in  precipitous  heights, 
and  he  confirms  Koch  in  stating  that  the  mountains 
behind  Mush  consist  also  of  mica-schists.  On  the 
south  side  of  the  Taurus,  at  Darakol  (almost  due 
south  of  this  point),  Taylor  §  noticed  that  all  the 
houses  were  built  of  clay-slate.  It  is  possible  that 
the  limestone  which  he  observed  at  Nerjiki  (Sasun 

*  Travels  and  Resiordus  m  Asia  Mtnar^  6r^,  iL  357. 

t  Reiu  impifkiisekia  GMrgt^  ii.  396,  399,  407. 

X  Reise  van  Thi/egmit,  kc.^  Petermai^a't  Mitth.  vL  69,  Gotfaa,  t86o. 

{  Jaum,  Rcy,  d^tgr.  Ar.  x.  357. 


PYe-Drnxmem  Rocks  ayi 

distridl),  and  at  Lijeh  (W.  of  Darakol)  is  also  meta- 
mofphic.  Iron  (specular  iron  ore,  p.  112)  is  worked  at 
Talori  (Sasun  oistrid),  and  at  Sivan  (8  hours  S.E. 
ofPalu). 

From  this  point  nothing  as  yet  is  known  con- 
cerning the  axis  of  the  Taurus  until  we  come  to  Lake 
Goljik.  Viquesnel*  (who  examined  Hommaire  de 
Heirs  rocks)  states  that  metamorphic  schists  occur  on 
the  shores  of  this  lake,  and  they  are  describe  in 
Hommaire  de  Hell's  narrative t  as  "friable  clay- 
slates,  alternating  with  a  disturoed,  schistose  sand- 
stone." This  is  corroborated  by  the  observations  of 
Wunsch:|;  that  the  Hasarbaba  Dagh  (the  range 
boimding  Lake  Goljik  on  the  south]  "seems  to 
consist  almost  entirely  of  clay-slate,"  whilst  the  range 
to  the  north  of  the  lake  "  consists  of  a  very  loose  and 
rapidly  weathering  rock,  immense  masses  of  loose 
rubbish  falling  into  the  lake/' 

De    Chancourtois  §    also    states    that 

**  beyond  Argana  [probably  just  south  of  Lake  Goljik]  the  mountains 
generally  present  the  diaracter  of  sacchaioidal  limestone,  resting 
on  very  quartzose  mica-achists,  and  the  two  rocks  are  frequently 
cut  by  porphyritic  veins/' 

Separated  from  the  Goljik  area  by  the  Eocene 
basin  of  Kharput,  a  large  outcrop  of  metamorphic 
rocks  occurs  at  the  confluence  of  the  Frat  and  Murad. 
In    general    terms    Viquesnel  |[    describes    it    as 

**  a  system  of  talc4cfaist,  day-slate  (paasing  into  phyllite  and  mica- 
schist),  and  limestone,  sometimes  granular,  sometimes  very  compact 
and  translucent  along  the  edges.  In  this  system  the  mines  of  argenti- 
ferous galena  of  K^an  Maden  occur,  in  contact  with  igneous  rocks.'' 

*  Bull.  Soe.  GM.  Fiwue^  wti.  s,  viL  499. 
t  Voyage  m  TWywir,  etc.^  GMogU^  iv.  83. 
%  Mitth.  CM.   G4S.  (s),  xzviii.  1885,  i-ai. 
{  CompUs-nnimi^  Aaii.  da  Sa\  Paris,  xviii.  830. 
II  Of.  at.  499. 


273  Geology  of  Amunia 

Hommaire  de  Hell,*  however,  in  his  narrative,  merely 
speaks  of  ''the  qnartzose  rocks  with  obscure  strati- 
fication/' and  "the  white,  very  hard  quartzite, 
constituting  entire  hills,*'  in  which  the  silver  mines 
are  situated. 
Smjrth,!  in  his  account  of  Keban  Maden,  states  that 

"  the  mountains  round  the  mines  in  general  show  bare  surfaces  of  gr^ 
compact  limestone,  or  of  aigiUaoeons  and  chloiitic  slates.  On  both 
sides  of  the  valley  of  the  town  rise  sharp  peaks  of  a  hard,  felspathic 
porphyry,  with  lai^ge  aystals  of  pink  felspar,  and  sometimes  wiA 
a  slaty  texture,  with  the  oystallixed  parts  so  ill  defined  that  where 
it  occurs  in  contact  with  the  day-slates  it  is  difficult  to  assign  to  each 
its  proper  boundary." 

Ainsworth's  account  J  of  this  locality  is  some- 
what more  explicit.    It  runs  as  follows : — 

''The  mines  of  Keban  Maden  lie  on  the  south  side  of  white, 
saccharoidal,  or  granular  limestone,  with  high  dip  to  S.E.,  resting  on 
a  formation  of  blue, 'compact  or  granular,  but  slaty  limestone,  with 
the  same  dip.  The  two  limestones  rest  on  mica-schist  (mica-slate), 
which  appears  in  the  valley  about  i  mile  south  of  Keban  to?m, 
and,  tilted  up  by  granitic  rocks,  rises  to  near  the  summit  of  the  hills 
to  the  S.W.,  separating  there  also  the  superincumbent  limestone  from 
the  granitic  rocks,  which  occupy  the  hill-slopes  below.  .  .  Near 
Keban,  particularly  to  the  south,  the  formations  aie  very  various. 
The  fundamental  rock  is  a  highly  crystalline  aggr^;ate  of  quartz, 
felspar,  and  mica,  but  in  other  places  there  occurs  a  crystalline 
aggregate  of  large,  white  felspar  crystals  (albite),  on  which  lies  an 
ordinary  gneiss,  capped  by  chlorite-schist,  through  which  the  fel- 
spathic rock  protrudes  in  dikes.  The  mica-schists  are  also  accom- 
panied in  places  by  carbonaceous  and  ferruginous  talc-schists  and 
by  chlorite-schists.  .  .  The  first  hills  north  of  Keban  Maden  are 
composed  of  granite  at  the  base,  with  superimposed  gneiss  and 
chlorite-schist." 

Similar  rocks  occur  further  to  the  N.W.,  on  the 
north  side  of  Arabkir,   which  (according  to  Ains- 

♦  op.  ^.  78,  80. 

t  ^iKin.  JouMm.  CM.  Boe*  i.  jjy. 


i 


Pre-Devonian  Rocks  273 

worth  *)  is  "  backed  by  high  mountains,  composed  at 
the  base  of  mica-schist,  sometimes  anthracitous/' 
overlain  by  horizontal  beds  of  "chalk,"  alternating 
with  sheets  of  basalt — 2l  series  which  is  probably 
of  Miocene  age  (see  Chapter  XXIII).  These  schists  of 
Arabkir  are  not  improbably  continuous  with  the 
"clay-slates,  blue  limestone,  and  mica-schist,"  which 
Tchihatcheff  t  found  to  the  W.S.W.,  on  the  same  line 
of  strike,  at  Belen,  in  the  Antitaurus ;  at  this  place 
'  (between  Albistan  and  Kaisariyeh)  the  schists  are 
overlain  by  beds  of  Middle  and  Upper  Devonian  age. 
The  metamorphic  heights  of  Lake  Goljik  are 
probably  continued  to  the  S.W.  through  the  Mastikan 
and  Chembek  ranges  to  the  neighbourhood  of  Er- 
kenek  (S.S.W.  of  Malatia,  on  the  road  to  Besneh). 
Here  Ainswortht  relates  that  colossal  limestone  cliffs 
hem    in    the    Gok    Su    at    first,    but    tliat 

''after  Erkenek  [Le.,  S.E.  of  it]  the  valley  changes,  the  rock  changing 
from  hard  limestone  to  shingly  schists,  which  here  as  elsewhere 
constitute  the  axis  of  the  Taurus." 

In    another    place  §    he    states    that 

*'  the  Taurus  consists  of  a  central  nucleus  of  granite,  gneiss,  and  mica- 
schists,  associated  with  limestones,  diorites,  and  disdlage-rocks,  with 
lateral  formations  of  diallage-rocks,  serpentines,  actinolite-rocks,  stea- 
sdusts,  and  slate-days,  and  of  outlying  sandstones  and  limestones." 

The  following  collated  observations  of  travellers 
will  show  that  the  Tauric  region  of  schists  and 
marbles  extends  eastwards  along  the  south  coast  of 
Lake  Van,  and  southwards  as  far  as  the  Bohtan 
valley,  forming  a  broad,  highly  mountainous  zone. 
My  own  observations  (p.  104)  along  the  coast  only 

*  op.  cit.  aSa. 

t  Asie  Mifuun^  GUogie^  i.  679-699. 

X  Thtvels  and  Researches  in  Asia  Minor ^  d«^.,  i.  259. 

{  Researches  in  Assyria^  etc.^  19. 


374  Geology  of  Armenia 

extend  from  the  Gdzel  Dereh   to   the   Sheikli    Ora 
cratefi  but — so  far  as  the  failing  light  would  perm  i  t — 
I  oould  see  from  our  boat  that  the  comparatively  low 
cliffs   were   composed  of  banded,    grey   and     ^^'^hite 
marbles,  overlying  mica-schists.    Several  faults   ^vere 
apparent,  and  the  strata  dipped  about  60^  to   S.S.E., 
and  then  (further  west)  to  N.N.W.:   an  anticlina.1  axis 
apparently  runs  along  the  shore- line  of  this  pa.rt  of 
the  lake.     In  general  terms,  de  Chancourtois  *^  states 
that  the  watershed,  which  runs  parallel  and    close 
to   the   south   coast  of   Lake  Van  is  composed    of 

**  a  grey,  compact  limestone,  overlying  micaceous,  or  greenish  taicose 
schist,  passing  sometimes  into  true  mica-schist,  and  dipping  steeply  to 
the  north." 

Hommaire   de   Hell,t   who   travelled   from    BitJis 
along  the  south  coast  of  the  lake,  noticed  that   the 
mountains  east  of  the  road  from  Bitlis  to  Tad va/i 
consisted    of    "micaceous    schists,   charadlerized     by 
masses  of  white  quartz,"  underlying  limestone,  which 
formed    the   summits.    This   agrees   with    my    own 
observations  of  this  spur  of  Urtab  (p.  aSg).    A  little 
to  the  south  of  Urtab,  on  the  road  between  Alaman 
and  Tukh,  Colonel  G.  S.  Elliot  coUeAed  a  specimen 
of  grey,  silky  mica-phyllite,  weathering  brown   (p. 

In  proceeding  eastwards  along  the  coast  from 
Gdzel  Dereh  through  Elmali  Dereh,  Hommaire  de 
Hell  (I.e.)  found  that  the  rocks  still  consisted  of 
"  mica-schists  with  quartz,"  and  Aat  the  dark  tuflEs  of 
the  Sheikh  Ora  crater  ^p.  103,  104)  rest  on  "mica- 
ceous and  quartzose  schists."  However,  I  noticed 
(p.  104)  that  the  tuff  contained  blocks  of  grey  and 
white,  banded  marble. 

*  Can^tes^renduSf  Aiod.  des  Set.  Paris^  zviii.  Szf. 
t  Vifyage  en  Titrqme^  Gtol^^  vr.  95,  tt  sqq. 


Pre-Devonian  tioeks  275 

According    to    the    same    authority    (Ix.) 

**a  blue  limestone  forms  clUTs  juttiiig  out  into  the  lake  between  Unsus 
and  Gharzil  [=Garzik].  .  •  It  is  much  disturbedi  and  occasionally 
encloses  very  hard  white  masses  [probably  quartz].  .  .  This  limestone 
continues  in  clifb  along  the  shore  to  Surp." 

In  connedlion  with  the  disturbance  of  these  strata, 
mention  may  here  be  made  to  the  specimen  of  fine* 
grained  diabasic  rock,  with  pyrites,  which  Colonel 
Elliot  collected  from  "  a  cliff  at  Surp  "  (p.  112), 

From  Surp  Hommaire  de  Hell  turned  inland, 
crossed  the  plain  of  Garchigan,  and  ascended  over 

"blue  limestones,  with  occasional  white  masses  [?  quartz],  ac- 
companied by  greenish,  micaceous  schists,  which  often  pass  insensibly 
into  the  limestone.*' 

It  seems  to  me  probable  that  this  was  the  locality 
whence  Colonel  Elliot  obtained  a  specimen  of  a 
micaceous,  chloritic  limestone-schist  (described  on 
p.  Ill)  which,  together  with  a  serpentine  (described 
on  the  same  page),  both  labelled  "south  coast  of 
Lake  Van,'*  he  kindly  pla,ced  at  my  disposal. 

Still  proceeding  eastwards,  the  pass  between 
Goli    and    Pelo    is,    according    to    Earl    Percy,^ 

'*a   low   crest   (6600    feet),    strewn   with    grey    slabs    of    slate.'' 

Hommaire  de  Hell  (I.e.)  relates  that  "  the  schists 
form  the  lower  heights,  and  continue  as  far  as  Anzek 

fe=Enzakh],  while  the  summits  are  of  limestone." 
arl  Percy  (Ix.)  states  more  explicitly  that  the  pass 
(7300  feet)  Detween  Pelo  and  Enzakh  is  "a  steep 
shoulder  of  mica-schist.'*  Chantre,t  however,  refers 
to  this  locality  as  *'the  difficult  pass  of  Intzakiars, 
consisting  of  lustrous  clay-slatts "  (see  also  p.  276), 
Interrupted  by  basalts  for  a  time  near  Sarik,  the 
schists  were  found  by  Hommaire  de  Hell  (Ix.)  to 

♦  Highlands  of  Asiatic  Turkey^  155. 

t  Beyraut  d  7\/lis,  Tour  da  Monde,  Iviii.  255. 


276  Geology  of  Armenia 

soon  re- appear  on  the  coast,  but  near  Vostan  a  more 
recent  lintestone  replaces  it  and  forms  cliJSFs  along  the 
shore  to  Artemid.  It  is  similar  to  the  limestone  of 
the  rock  of  Van,  and  is  probably  Eocene  in  age  (see 
Chapter  XXI). 

The  lofty  Agherov  Dagh  (lying  between  this  part  of 
the  coast  and  Mukus),  together  with  the  Amos  Dagh 
(east    of    Mukus)    consist     according    to    Kotschy,* 

**  entirely  of  mica-schist,  yet  nearer  the  lake  it  alternates  with  lime- 
stone. .  .  The  hospice  of  Khana  Putkieh  (7000  feet),  on  the  south 
side  of  the  Agherov  Dagh,  towards  Mukus,  lies  deep  in  a  funnel 
of  precipices  of  mica-schist  and  other  primitiTe  rocks.'' 

Thence  the  road  to  Mukus  and  its  spring- grotto  lies 

<'  over  slippery  marble — a  fine-grained,  ash-grey  limestone  (Alpenkalk)^ 
which  forms  the  prevailing  ground  from  half  an  hour  below  Khana 
Putkieh  through  the  whole  of  the  Mukus  dutrict.  At  Mukus  the 
limestone  gives  place  for  a  short  distance  to  a  porous,  coarse-grained 
sandstone,  readily  weathering,  extending  right  across  the  valley,  but 
the  limestone  predominates.    It  is  intruded  by  syenite." 

Kotschy,  in  travelling  from  Mukus  to  the  Shirvan 
valley,  met  with  the  ^ists  again  at  Masca  Uanobeni 
{9ic\  west  of  the  Hesher  Mountains  [=Gia-i-Hestiri], 

From  some  geological  MS.  notes  of  a  journey  from 
Van  to  Sert,  kindly  communicated  to  me  by  Major 
F,R.  Maunsell,  R.A.,  I  am  able  to  extradl  the  follow- 
ing details  of  a  traverse  of  this  Tauric  metamorphic 
area,  from  N.E.  to  S.W.: — 

'^The  col  between  Enzakh  and  Pclo  [1.^.,  the  Lake  Van — Ghindig 
Su  watershed]  is  formed  of  bedded  quartzose  rock,  dipping  80^  N.E. 
[cp.  p.  375].  Proceeding  down  the  Ghindig  valley  to  the  W.S.W., 
the  rocks  s  or  3  miles  below  Pdo  are  dark  shales  with  W.N.W. 
strike,  dipping  steeply.  At  a  village  7.5  miles  from  Pdo  the  rocks  are 
fliggy  or  shaly,  much  contorted,  but  apparently  with  a  general  S.W. 
strike,  dipping  to  N.W.  .  .  The  rocks  on  the  right  bank  are  flaggy, 
striated  and  banded  in  colours^  the  layers  being  often  much  contorted. 
The  rocks  on  the  other  side  appeared  more  schistose  and  rugged.  At 
Saris  the  rock  is  mica-schist.   .   .    In  the  bend  of  the  Ghmdig  Su^ 

*  Riifi  wm  Tyapmme,  eU.,  Peternann's  Mitth.  vL  73,  Gotha,  x86o. 


Pre^Devonian  Rocks  277 

lieyond  the  teHeh  of  Sheikh  Jelhl«ed-dby  there  stands  a  fine  rocky 
peak  of  massivQ  crystalline  limestone.  pThis  loodity  is  probably 
indicated  by  Earl  Percy  *  in  his  reference  to  a  height  between  Khoros 
and  Tekke  (4500  feet),  of  which  ''the  ^lift  crowning  die  summit 
are  bare  and  jagged,  apparently  of  dolomite"].  Half  an  hour  beyond 
this  point  shale  and  mtca*schist  again  appear,  dipping  85"*  S.,  strike 
W.  by  N.  .  Be3rond  Karasu  the  road  fint  crosses  the  range  of  low 
hills  of  thinly  bedded  sandstones,  strike  £. — ^W.,  dip  neariy  verticaL 
Many  blocks  of  quartz  are  seen  about  the  summits  of  these  hills.  .  . 
Descending  into  the  valley  of  Khisan,  the  rock  becomes  finely 
laminated  and  banded  in  colour,  with  quarts  interstiadfied  in  places. 
Lower  down  it  becomes  micaceous,  with  a  cleavage  running  across  the 
bedding  in  a  N.W  by  W.  directioui  /.«.,  at  an  aogle  of  about  40*^ 
with  the  bedding.  In  the  valley,  near  a  fine  stream  crossed  before 
reaching  the  MiiA  of  another  Sheikh  Jellal-ed-din,  the  rock  is  still 
ouca-sdhist,  dip  4s'*  N.,  strike  about  E.^ — ^W.  Cros^dng  a  low 
lidge  beyond  Khizan,  the  mica-schist  dips  70^  S.W.  [These  ob- 
servations corroborate  Kotschy's  statement  t  that  the  valley  at  Khizan 
lies  in  weathered  schist].  The  river  then  runs  through  a  fine, 
precipitous  gor^e  of  dark  grey  limesttHie  with  white  veins,  dip  20^ 
S.W.  Underlying  the  limestone  is  still  mica-schist."  This  massive 
limestone,  "in  vast  mural  precipices,"  is  evidently  unconformable 
and  of  later  date,  for  the  underlying  mica-schists  show  ''distinct 
stratification,  dip  4S**te*  N.W.,  nearly  at  right  angles  to  that  of 
the  limestone,  and  with  S.W.— N.K  strike" — the  same  strike  in 
fact  as  I  noticed  at  Gosd  Dereh  (p.  269)  on  the  north  side  of  this 
metamorphic  zone.  Shortly  before  reskching  the  Shirvan  valley, 
''these  underlying  rocks  change  to  a  grey  porphyry.  About  7 
miles  up  this  valley  [or  Mad^n  Su]  the  porphyry  gives  place  to 
diale  and  trap."    The  sditsts  cle.irly  do  not  extend  fmther  south. 

Several  travellets  have  made  observations  on  this 
metamorphic  area  further  west,  between  the  Keser 
Su  and  Bitlis.  The  Chelifteh  Dagh  (dividing  the 
valleys  of  the  Keser  and  Bitlis  riveif  between  Yeni- 
khan  and  Dukhan)  cx)nsists  accor^ng  to  Loftus^ 
of  "blue,  altered,  contorted  limestone."  Proceeding^ 
up   the   Bitlis  river,   he  found   this  blue,  compaA 


*  Highlands  of  Asiaik  TMrkiy^  153. 

t  Reise  von  TIrapetnmtt  tf^.*  Petermann's  Mitth.  vi.  74,  Gotha,  z86o. 

I  Quart.  Joum.  GooL  Jbr.  xL  385. 


278  Geology  of  Armenia 

limestone  to  overlie  conformably  "micaceous  slate, 
both  dipping  away  from  the  ravine/*  Between  Shall- 
tek  [=Shatakh]  and  Bitlis  the  '* micaceous  slate*' 
predominates,  much  contorted,  forming  lofty  ranges 
on  either  side  of  the  valley.  According  to  Hom- 
maire  de  Hell*   the  rocks  in  this  part  consist   of 

''qmrUosey  micaoeoos  day-datei  with  much  ulute  quartz.  Among 
them  there  occurs  inddentally  a  bluithachistoie  rocki  very  micaceous, 
aaalofOtts  to  grauwacke.'' 

The  mountains  to  the  east  of  Bitlis  are  com- 
posed,   according    to   the   same  authority   (Z.c),   dt 

'*  micaceous  schistSt  characterised  by  masses  of  white  quarti, 
uaderiying   limestone,   which    formed   the   summits." 

Loftus  also  states  ({.c.)  that  these  heights  on  both 
sides  of  the  head  of  the   Bitlis  valley  consist   of 

*'grey,  quartxose  rock,  underlain  by  soft  mica-schist  The  beds 
dip  away  at  a  slight  angle  from  the  ravine.'' 

Ainsworthf  travelled  from  Sert  to  Bitlis  by  a 
slightly  more  easterly  route  from  Yenikhan  through 
Ulek  [Olek]  into  the  Bitlis  valley ;   hd  considered  that 

'*the  central  axis  of  the  primary  schists  is  the  AU  Dagh ''  [S.  of  Olek]. 

To  the  north  of  Olek  he  again  crossed  '^  hills  of 
chlorite-schist,*'  and  descended  into  the  vale  of  Inip, 
leaving  it  by  a  narrow  pass  in  **  limestones  resting 
on  mica-schists,*'  and  thus  came  into  the  Bitlis 
valley,  evidently  near  Pakhund,  a  few  miles  below 
Bitlis.  His  statement  that  ''  the  mountains  around 
Bitlis  are  of  limestones  on  mica-schists  **  corroborates 
the  above  quotations.  De  Chancourtois  ^  noticed 
that  these  metamorphic  rocks  of  the  Bitlis  valley 

**  have  for  the  most  part  an  E. — ^W.  strike,  although  sometimes  quite 
different ;"  (cp.  pp.  876,  277,  where  similar  Tariations  in  strike  are 

*  Vfyagf  m  Titrquii^  Gtd&gi^  tv.  94,  95. 

t  TVaods  wul  Rmofd^  in  Am  Mmor^  6^^.,  ii.  36a 

X  Compi$$-f9nAiSf  Ami.  ies  Sa,  Paris^  xviii.  837. 


\ 


Pre-Devonian  Rocks  279 

recorded  in  the  Ghindig  vallej  further  east 

Kotschy*  rode  eastwards  from  Bitlis  over  the 
mountain-crest  into  the  Gozel  Dereh,  and  found  that 

''on  the  summit,  6J167  Paris  feet,  a. 5  hours  from  BitHs,  schist 
alternated  with  limestone;"  cp.  my  observations  in  the  Gozel 
Dereh  (p.  S69). 

Only  meagre  indications  are  at  present  extant  with 
regard  to  the  easterly  extension  of  the  Tauric  schists 
beyond  Mukus»  but  dolomites  are  perhaps  to  be 
inferred  by  Earl  Percy's  reference  t  to  "  the  whitened 
line  of  spires  overlooking  Shattakh  (5300  feet)/* 
According  to  Ainsworth,t  "quartz-rock  and  schist'" 
compose  the  Zoma  Suwarri  (with  its  glaciers  of  Itha)« 
which  is  the  highest  range  of  the  Kuriki  Dagh,  on  the 
west  side  of  the  Zab.  Perhaps  the  granitic  area 
of  Eslaya  (6258  feet)  and  Tur  Burju-llah,§  lying 
between  these  mountains  and  Julamerk  may  prove  to 
be  as  ancient  as  the  neighbouring  schists,  which 
are  flanked  on  both  sides  by  Cretaceous  rocks  (see 
Chapter  XX). 

On  the  east  side  of  the  Zab  the  lofty  Jelu  Dagh 
is  perhaps  a  continuation  of  the  Tauric  metamorphic 
belt.  Its  serrated  peaks  rise  to  over  13,000  feet, 
and  probably  consist  to  a  large  extent  of  dolomites, 
with  clay-slates,  but  we  possess  no  geological  account 
of  this  interesting  region.  Major  E.  Clayton  ||  has 
given  an  outline  sketch  of  these  mountains  and  states 
that  their  forms  reminded  him  of  "the  Tyrolese 
dolomites/*    Earl  Percy  ^  describes  them  as  "  a  chain 

*  Heiu  wm  TWtftmmtf  «fr.,  Petermann's  Mitth.  vL  76,  Gotha>  z86o. 

t  Highlands  of  Asiatic  Titrkejf^  z6a 

X  Travds  and  Rtssarekss  in  Asia  Mmar^  &*c,f  ii.  S15. 

{  Idid,  ii.  830. 

II  AiftHs  Jmnm.^  1887,  995. 

\  Op.  cii.  164,  &  frontispieoe. 


1 


a8o  Geology  •/  AttMma 

of  jagged  needles  like  the  spires  of  Milan  Cathedral/' 
and  gives  a  photograph  showing  that  the  pinnacled 
precipices  consist  of  thickly  bedded  strata,  dipping 
30*^  S.,  apparently  of  limestone  or  dolomite,  and 
certainly  not  granitic,  as  Ainsworth^  supposed.  In 
commencing  to  cross  the  Jelu  range  south  of  Jula- 
merk,  Earl  Percy  f  states  that  the  ascent  from  the 
confined   Zab    valley    began    up    a    narrow    gorge 

"between  naked  walls  of  crambUng  datei  the  road  twisting 
and  turning  over  the  splinters  and  slutvingi'' — a  description  quite 
applicable  to  a  clay-slate* 

Further  on,  to  the  south  of  the  village  of  Madis, 

-'blocking  the  valley  to  the  south-west  are  two  strange  clifi  like 
sugar-loaves,  resembling  the  Tre  Sorori  of  the  Austrian  Tyrol.  The 
nearest  is  called  Suringa  Kaleh  ;"t  these  probably  consist  of  dolomite. 

On   the  south  side  of   the   Jelu  range  he  found  § 

''huge  grey  boulders.  Stupendous,  greyish  white  precipices  stretch 
eastward  to  meet  the  darker,  red-streaked  peaks  of  Oramar  Dagh.'' 

The  latter  are  perhaps  intrusive  in  character,   for 

'*  the  banks  of  the  Ishtaain  valley  [the  N.E.— S.W.  river  just  south  of 
Oramar]  dxt  littered  with  brilliant  red  and  green  dabs  of  jasper  and 
poiphyry."|| 

Earl  Percy's  photograph^  of  this  locality  shows 
these  cliffs  to  be  boldly  jointed  like  the  intrusive 
porphyrites  of  the  Khaishut  valley  (p.  28). 

The  following  additional  notes  on  the  Jelu  Dagh 
have  been  kindly  commimicated  to  me  by  Major  F.K. 
Maunsell : — 

''The  very  deep  Ishtasin  gorge,  only  8  feet  wide  [evidently  in 

*  Tmoels  Cr  Ruiorekis  in  Asia  Mmor^  Cfc.  ii.  taS. 

t  Bighiands  rfAsiaiu  TMkjf^  183. 

X  IbU.  185. 

{  Ibid.  206. 

N  INd.  ao8. 

\  lUi.  109. 


Pre^Devoman  Rocks  38 1 

its  upper  part],  is  cut  throqgh  gtty  fifnestone;  the  blue  dilTs  ef 
tfie  Geliashin  peak  (x3»5oo  feet)  show  sheer  precipices  on  all  sides ; 
from  this  point  westwards  a  razor-edge  of  limestone  ends  in  Suppa 
Derek ;  Gsdianu  is  very  steep  and  rocky,  composed  of  a  friable  shale ; 
its  snows  feed  a  glacier;  Tura  Danil  [W.  of  Galianu,  and  S.W.  of 
Geliashin]  is  a  precipitous  limestone  ridge.''. 

To  the  S.E.  of  the  Jelu  Dagh  the  schistose  zone 
is  continued  into  the  Zagros  through  Rowanduz, 
which,  according  to  Ainsworth's  account,*  exhibits 

**  bold  and  sharp,  rocky  pinnacles  of  grey  and  green  quartz,  precipices 
of  brown  and  blue  schists,  and  giant  precipices  of  limestone." 

In  the  Zagros  itself  Rodlerf  found  that  the 
Devonian  beds  rest  on  a  banded,  blue,  unfossiliferous 
limestone,  overlying  black,  calcareous  clay-slate. 
Since  schists  underly  Devonian  beds  in  the  z^agros, 
the  Antitaurus  (p.  273),  and  the  mid-Araxes  region 
(p.  264),  it  is  at  any  rate  a  strong  presumption  that 
the  schists  within  the  triangle  formed  by  these  points 
(i.e.,  the  Tauric  and  Armenian  area)  are  all  of 
pre-Devonian  age. 

IV.  THE  CENTRAL  REGION. 

Metamorphic  rocks  evidently  occur  north  of  the 
Jelu  Dagh,  along  the  Turco-Persian  frontier,  for 
Pohlig:}:    mentions,    in    a    general    manner,    that 

''to  the  west  of  Lake  Urmi  mountains  occur  with  a  N.N.W.  strike, 
composed  of  red  granites,  rocks  like  augen-gntiss,  knotted  schists, 
and  day-slates,  overlain  by  Cretaceous  beds." 

The  silvery,  graphitic  schists  together  with  the 
overlying,  dark  bluish  grey  marble,  which  I  found  (p. 
61)  at  the  west  end  of  the  Ak  Dagh,  N.W.  of  Khinis, 
are  probably  also  of  pre-Devonian  age.    The  whole 

*  Thwds  and  Raearches  in  Asia  Minor^  dr'r.,  ii.  syt. 

t  Sihungsber.  k.-k.  Akad.  Wiss.  Mathem.-naturw.  CI.  xcviii.  a6. 

X  Verk.  k.-k.  gtol.  JUicksanst  1884,  384. 


aSa  Geology  of  Armenia 

series  dips  40®  N.N.W.  at  the  Kaiakaya  col  (7500- 
feet);  the  graphitic  schists  contain  thin  seams  of 
white  marble  in  their  upper  part,  and  are  greatly 
contorted  and  puckered.  The  marble  is  banded  and 
foliated,  and  extends  northwards  into  the  barren, 
furrowed  hills  of  the  Kurreh  Hassan  Dagh  (p.  58), 
to  sink  westwards  beneath  the  lavas  of  the  Bingdl 
Dagh ;  the  last  outcrop  is  the  rounded  hill  of  Gabefeh 
(p.  58),  composed  of  grey  and  white,  foliated  marble, 
dipping  50°  N.W.,  with  a  steep  escarpment  facing 
S.E.,  and  rising  like  an  island  out  of  the  surrounding 
lava.  At  the  col  (734.0  feet)  between  Gabeleh  and  Ali 
Mur  the  marble  has  been  intruded  by  a  serpentine  (a 
pyroxene-olivine  rock),  which  shows  an  interesting 
cataclastic  structure  (p.  59). 

The  main  mass  of  the  Ak  Dagh  consists  of  strongly 
folded  marble  (p.  69);  it  is  not  a  single  range,  but 
is  composed  of  a  series  of  parallel  ridges,  with 
W.S.W.-E.N.E.  strike,  closely  ranged  together  in 
^helon,  and  dissedled  by  denudation  into  numerous 
peaks.  Ainsworth*  crossed  the  Ak  Dagh  (or,  as  he 
called  it,  the  Chekmah  Dagh)  by  the  Akhveran  pass ; 
here  he  observed  that  lava  has  risen  up  "  in  dikes 
through  indurated  limestones  on  mica-schists'*  to 
form  the  sheet  of  black  basalt  which  has  given  the 
heights  west  of  the  pass  the  name  of  Karakaya. 

On  the  south  side  of  the  plain  of  Khinis  Ains- 
worth (Ix.)  found  that  the  lowest  rocks  exposed  in  the 
Basbkent  col  between  Khamur  and  the  Bingdl  Dagh 
consist  of  "  mica-schist  and  clay-schist.*'  It  seems  to 
me  probable  that  the  marbles  and  schists  of  the  Ak 
Dagh  rise  up  again  on  the  west  of  the  Bingol 
Dagh  to  form  the  serrated  peaks  of  the  Javresh  and 
Shaitan  Dagh,  lying  on  the  same  line  of  strike. 

*  TfWfUs  and Xismuxkes  m  Asia  Minor^  ft^c.^  ii.  385. 


CHAPTER  XVI 

THE  DEVONIAN  AND  CARBONIFEROUS  OF  ARMENIA 

I.     6INBIUL    DimUBUTION. 

Since  there  is  no  break  in  continuity  between  the 
the  l)evonian  and  Carboniferous  of  Armenia,  either  in 
sedimentation  or  in  organic  life,  it  seems  natural  to 
place  the  two  systems  in  the  same  chapter. 

Recently  our  knowledge  of  these  systems  in  the 
mid-Araxes  region  has  been  greatly  increased  and 
placed  on  a  &rm  footing  by  the  investigations  of 
rrech  and  von  Arthaber.*  Their  observations  com- 
mence with  the  lower  limestones  [Calceola  stage)  of 
the  Middle  Devonian.  Abich,t  however,  distinctly 
states  that  the  Devonian  limestones  overlie  quartzitic 
sandstones  and  conglomerates  with  porphjrry  pebbles, 
similar  in  facies  to  the  Old  Red  Sandstone ;  and  it 
is  not  improbable  that  these  fragmental  beds  may 
mark  a  great  transgression  of  the  sea  in  the  Lower 
Devonian  epoch.  The  entire  system  was  seen  by 
AbidiJ  to  rest  on  "dark,  chloritic  aiid  aphanitic 

*  Das  PalSamncum  in  Hacharm^ttcn  6*  Persien^  Bcitr.  Pal.  (Estr.- 
Ung.  ziL  i6i. 

t  GeoL  Fonch.  in  den  tumk.  JLandem^  ii.  25 ;  and  his  Vergleichtnde 
t»L  Grundnigi  €i€.  76. 

%  G»i.  Forstk.  ac.  ii.  2$. 


a84  Geology  of  Armenia 

schists  in  connecJtion  with  dioritic  schists,  probably 
belonging  to  the  Silurian." 

The  chief  outcrop  of  the  Devonian  and  Carboni- 
ferous occurs  to  the  south  and  east  of  Ararat,  between 
the  Maku  river  and  the  Araxes ;  it  increases  in  thick- 
ness in  the  Daralagoz  mountains^,  N.E.  of  the  Araxes 
plain.  It  is  continued  down  the  Araxes  valley  to 
Julfa,  and  extends  southwards  into  Azerbaijan.  It  is 
well  exposed  (according  to  Abich  ^)  above  Gerger,  on 
the  way  from  Julfa  to  Tabriz;  between  Caravan- 
serai and  Marand  the  Carboniferous  Limestone  is 
associated  with  manganese  ores ;  and  between  Sofian 
and  Tabriz  the  dark  limestones  are  well  exposed 
in  the  high  Merov  Dagh,  where  they  rest  on  meta- 
morphic  schists  and  thick  conglomerates. 

The  Palaeozoic  beds  are  continued  west  of  Gerger 
underlying  Cretaceous,  and  they  are  traversed  by  the 
Araxes  between  Karmirvank  and  Julfa.  Towards 
Khoi  the  Palaeozoic  is  overlain  by  Cretaceous,  and 
at  Khoi  by  Tertiary  beds;  "the  strongly  folded 
Devonian  and  Carboniferous  only  occasionally  be- 
come exposed  to  view  in  the  valleys."  At  Shahi,  and 
on  the  coast  and  islands  of  Lake  Urmi  the  dark  grey 
Carboniferous  limestone  is  direcftly  overlain  by  Mio- 
cene limestones.  A  Palaeozoic  age  is  probable  also 
for  the  dark  limestones  with  indistindt  fossils,  asso- 
ciated with  light  grey  dolomite,  and  with  dark 
schists,  quartzites,  gabbros,  &c.,  which  occur  in  the 
mountains"  on  the  west  coast  of  Lake  Urmi  and  near 
the  hot  borax  springs  of  Salmas.f 

These  Devonian  and  Carboniferous  limestones  have 
been  traced  far  to  the  east  through  the  Elburz 
(Alburs)  Mountains  to  Asterabad  and  Shahrud«  and 

*   Vergieickende geoL  Grundziige  etc.  77. 

t  Abich,  GeoL  Foruh.  in  dm  kauk,  LanJkm^  i.  a»4. 


Devonian  and  CarbMiferons  285 

thence  into  Khorasan  and  Afghanistan.  Some  traces 
of  these  systems  occur  also  in  the  southern  Iranian 
arc,  viz.y  in  the  Zagros  Mountains  south  of  Lake 
Urmi.  In  these  ranges  Rodler*  considers  that  the 
Palaeozoic  shows  the  following  succession  : — 

1.  Foraminiferal  Caifoonifefont  limestone. 

2.  Devonian  Limestone. 

3.  Banded,  blue  limestone,  unfOBsiUferous. 

4.  Blacky  calcareous  chy-slate. 

Probably  Loftus*  "blue,  foetid  limestones/'f  which 
form  a  zone  east  of  the  Cretaceous  of  the  Zagros,  are 
also  Palaeozoic. 

Finally,  to  the  west  of  the  Armenian  area,  Tchi- 
hatdieffi  found  Devonian  and  Carboniferous  to  be 
well-developed  in  the  Antitaurus,  at  Chatal-oglu, 
Belen,  and  Fekeh  (between  Kaisariyeh  and  Albistan). 
An  analysis  of  his  lists  of  fossils  shows  that  the  same 
sones  occur  as  in  the  mid-Araxes  and  Daralagoz 
regions,  viz ; — 

Lower  Carbomferow  ^  ^  ^one  of  StiriJ^  ianuumsis  (Transitional) 
Upper  Devonian  3^  2k>ne  of  Sfiriftr  Vtmeuiii. 

/  4.  Upper  stage,  with  Cyathophyllum  quadri- 
Middle  Devonian       i  gemimm. 

(5.  Cti/r^Ai-stage. 

It  is  evident  that  there  can  have  been  no  barrier 
of  land  between  the  Armenian  and  Antitauric  areas 
during  these  periods. 

It  is  difficult  to  reconcile  Abich's  statements  with 
those  of  Freeh  with  regard  to  the  strike  of  the 
Devonian  and  Carboniferous  strata  in  the  mid- 
Araxes    region.     According    to    Abich§ 

*  Sihmngsher.  t.^i.  Akad.  Whs,^  JfatAem^-naturw.  CI.  xcviii.  28. 

t  Quart.  Jaum.  Geoi.  Sac.  xL  247. 

X  Asit  Mineure^  Ghtogii^  i.  679-699. 

{   VergUichende  GruMdufgi  6v.  »i,  81 ;  and  GioL  Forsch.  drv.  ii.  8. 


286  Geology  of  Armenia 

"the  S.W. — N.E.  direction  has  everjrwhore,  without  exception^ 
affected  the  Palaeozoic  of  Armenia.  .  •  There  is  no  doubt  about  the 
original  S.W. — N.E.  strike  of  the  Devonian  and  Carboniferous 
Limestone,  which  extend  in  a  broad  cone,  partly  covered  by  later 
strata,  from  the  southern  foot  of  Ararat  diagonally  across  Armenia. 
In  spite  of  th:  in:redible  disturbances  which  these  strata  have  under- 
gone, as  a  result  of  successive  elevations  in  other  directionsi  this 
S.W. — N.E.  strike  is  still  clearly  apparent." 

Yet  Abich  *  himself  observed  that  in  the  DaralagSz  it 
passes  through  S.S.W.-N.N.E.  very  distindlly  into 
S.E.-N.W.,  e.g.y  in  the  Gyneshik  valley.  Freeh  t  also 
noticed  a  variation  in  the  strike  between  the  Arpa- 
chai  (Daralagoz)  and  the  Beli  Dagh :  where  the  dip 
was  slight  the  strike  underwent  frequent  changes, 
but  wherever  the  dip  was  steep  the  strike  corres- 
ponded to  what  Freeh  maintains  is  the  general 
direction,  viz.^  S.E.-N.W.  Freeh,  however,  noticed 
the  S.W.-N.E.  and  W.S.W.-E.N.E.  variations  just 
west  of  Sardarak,  ^n  the  left  bank  of  the  Araxes,  and 
this  agrees  with  Abich's  observations  J  of  the  W.S.W. 
-E.N.E.  strike  in  the  small,  parallel  Palaeozoic  chains, 
which  lie  across  the  plain  to  the  south  of  Ararat,  and 
are  traversed  by  tbe  Araxes  at  Degma  Danga  (a  few 
miles  south  of  Sardarak). 

It  is  noteworthy  that  tl;ie  Palaeozoic  rocks  in  this 
region  have  nowhere  .been  subjected  to  any  very 
acute  folding  or  overthrusting  although  local  dis- 
turbances may  occur,  owing  to  the  intrusion  oi 
igneous  rocks.  In  facft,  Freeh  (I.e.)  was  "irresistibly 
reminded  of  the  regular  troughs  and  saddles  of  the 
outer  Jura."  It  is  evident  that  the  folding  (which 
occurred  between  the  Lower  Carboniferous  and  Upper 
Permian)  must  have  been  of  a  comparatively  gentle 

*   Vergleichende  g€9l,  Grundxuge  etc,  80. 
t  Op.  cit.  X71. 
X  Op.  cit.  as. 


.  ^1 


Middle  Devonian  287 

nature ;  it  was  unaccompanied  by  any  contemporary 
volcanic  action,  and  probably  tiiie  resultant  folds  had 
a  S.W.-N.E.  strike ;  this  folding  was  perhaps  the 
final)  and  feebler  effort  of  the  force,  which  had 
previously  impressed  the  same  strike  on  the  Tauric 
schists  and  marbles  (p.  267).  Now  a  Triassic  or  post- 
Triassic  folding  took  place  after  the  deposition  of  the 
Permo-Triassic  beds  of  Julfa  (see  Chapter  XVII),  and 
it  was  this  folding  which  caused  the  more  strongly 
pronoimced  S.E.-N.W.  strike ;  it  was  not  only  more 
powerful,  but  it  was  accompanied  by  many  intrusions 
of  igneous  rocks.  All  later  folding — e.g.^  m  the  ante- 
Tithonian,  post-Cretaceous,  and  mid-Miocene  periods 
— ^has  proceeded  along  the  same  lines  in  this  distridt. 
This  S.E.-N.W.  line  of  strike  occurs  in  contemporary 
folds  from  the  Hercynian  chains  of  Silesia,  through 
the  parallel  folds  of  the  Donetz  and  the  abraded 
primeval  mountains  or  South  Russia,  through  the 
Palaeozoic  ranges  of  the  mid-Araxes  into  Persia. 

II.      MIDDUE  DBVONUN. 

Whilst  in  Central  Asia  the  Middle  Devonian 
transgression  begins  everywhere  with  the  higher 
stage,  the  typical  Calceola  beds  were  found  by  Freeh  ^ 
at  Daheha  (east  of  Sardarak),  with  fossils  often 
identical  with  those  of  Torquay,  the  Eifel,  the 
Eastern  Alps,  and  the  Antitaurus.  Well-known  Eifel 
corals  and  brachiopods  weather  out  of  a  rock  agreeing 

EBtrographically    with    the    marly    Upper    Calceola- 
imestones  of  Gerolstein.     They  comprise  the  fol- 
lowing forms : — 

Calceola  sandalina^  Linn.;  Cyathophyllum  Araxis, 
Freeh ;  C.  vennicularej  Gf.  mut.  prcecursor^  Fredi 
(very  frequent  and  of  mratiud  site) ;  C.  coespitosum^  Gf. ; 

•  Of.  di.  173,  i8s-t86» 


^88  Geology  of  Armenia 

Endophyllum  Nicolai-Michaelidh,  Freeh  {\txj  fre- 
quent); Mesophyllum  maximum^  Schlut  var.  dam- 
noniensisj  M-E.  &  H.;  Af,  cylindrtcum:  Schlut; 
Cystiphyllutn  cristatum^  Frch,  C.  tiestcuZosum,  Gf. 

Favosites  Goldfussi,  M-E.  &  H.;  Striatopora  cf. 
denticulata^  M-E.  &  H.;  i42woZ/tes  subor6tcuIans, 
Lam.  var.  minor,  Freeh. 

Heliolites  porosus^  Sch. 

Fistulipora  favosa^  Gf. 

Stromatapora  concentrica^  Gf.  &  Nicholson ;  Stroma- 
toporella  eifeliensis^  Nich.;   i4(fi/nostroma  steZItt- 
latuniy  Nich. 
Brachiopods    are    somewhat    rarer,   and   comprise 

Produdiella  svbaculeataf  Murcfa.;  Leptcena  depressa. 
Sow.;  Spirifer  speciosus^  Freeh,  S.  svbcuspidatus, 
Schl.;  Cyrtina  heteroclitafDefr.;  Merista  plebeia^ 
Sow.;  Atrypa  reticulariSy  Linn.,  i4.  aspera^  SehL 
Pentamerus  Sieberi^  von  Bueh,  (the  mid-Devonian 
mutation). 

Bryozoa. 

Paracyclas  proavia^  Gf. 

The  upper  beds  of  the  Middle  Devonian  were  foimd 
by  Freeh  *  in  two  neighbouring  localities ;  both  the 
limestones  show  a  W.N.W.  strike  : — 

(i)  a  coral  limestone,  about  a  verst  west  of  Daheha» 
full  of  Cyatkophyllum  cf.  quadrigeminumfif. 

(ii)  a  brachiopod  limestone,  west  of  Sardarak,  with 
no  corals,  but  studded  with  Spirifer  inflatus^  Sehnur, 
and  Chonetes  cf.  Bretziy  Schnur. 

The  abundance  of  the  first  species,  and  the  oc- 
currence of  the  following  rarer  forms,  Spirifer  medio- 
textiis^  Arch.  Vem.,  and  Rhynchonella  letiensis^  Goo., 
charaAeristic  of  the  upper  stage  of  the  Middle  Devon- 
ian, leave  no  doubt  of  the  exacc  horizon  of  the  bed. 

♦  Op.  cit.  171,  174. 


Middle  Devonian  289 

The  Middle  Devonian  is  clearly  indicated  by  the 
following  fossils  which  Abich*  found  at  Khorvirab, 
higher  up  the  Araxes,  rising  like  an  island  out  of  the 
great  loss-plain : — 

Cyathophyllunt  quadrigeminum^  Gf.;   CampophyU 

lum  flexuosumj  M-E. 
Favosites  spongites^  Gf.,  F.  polymorpha^  Gf. 
Spirt fer  speciosiis^  Murch.;  Orthis ;  Lingula. 

These  Devonian  beds   are  conformably  overlain  by 
Carboniferous  Limestone  {vide  infra). 

Another  block  with  Devonian  fossils  (not  specified) 
"as  at  Khorvirab'*  occurs  on  the  lower  slopes  of 
Little  Ararat,  immediately  beneath  thick  lava-streams 
which  have  also  flowed  round  this  fault-scarp  of  the 
Araxes  depression,  3500  feet  below  the  Tambali  pass 
(6600  feet).  Here  both  the  Devonian  and  the  Car- 
boniferous have  been  traversed  by  "red  quartz- 
porphyries  and  green  labrador-poiphyries,"  especially 
at  the  Giisgiin  Dagh,  where  a  red,  micaceous  sand- 
stone like  Old  Red  Sandstone  is  present,  apparently 
imderlying  Devonian  limestones.t 

To  the  N.E,,  2  miles  north  of  Yaiji  in  the  Daralagoz 
distriA,  the  following  Middle  Devonian  fossils  were 
collected  by  Tsulukidse,  and  are  now  in  the  Till  is 
Museimi  X  \ — 

Cyathophyllum    obtortum,    M-E.;     Cystiphyllum 
vesiculosum^  Gf. 

Favosites  gracilis^  Sandb.;   Alveolites  suborbicula- 
Ws,  Lam. 

Tentaculites  cf.  annulatus. 

*  VergUichende  giol.  GtundMugteU.  78;  Bull.  Soc.  Ghl.  Ihmn,  (s) 
iii.  138;  Qttart.Jmm.  Gtpl*  Soc  iL  93. 

t  Grewingk,  Dr.  CCA.,  Gtegnast  &»  wrogra^.  VerAdlM.  des  nonB. 
PenUnSy  41. 

X  Giol.  CaM.  TiJKs  Mmmm^  92. 

V 


290  Geology  of  ArmMta 

The  Middle  Devonian  (overlain  by  Upper)  is  also 
indicated  by  the  presence  of  Favosites  Goldfussi^  M-E.  & 
H.|  among  the  fossils  collected  by  TsuluKidse  *  from 
the  hill  of  Ashaga-buruni  near  A^ga-dansik^  on  the 
right  bank  of  the  Daralagdz  Arpa-chai. 

in.     UPPIR  DIVOMUK. 

The  steep  hill  of  Davalu,  on  the  left  bank  of  the 
Araxes,  rises  boldly,  like  Khorvirab,  out  of  the  wide 
loss-plain.  It  is  composed  of  pure,  reddish,  thickly- 
bedded  limestones,  with  marly  intercalations,  and 
of  black  clay-slates.  These  contain  a  brachiopod 
fauna  of  Upper  Devonian  age.  Here  the  strike 
of  the  much  disturbed  limestone  is  W.N.W — 
E.S.E.  (just  as  on  the  opposite  mountains),  with 
a  dip  of  30^-50^  N.N.E.  The  strike  cuts  the 
main  direction  of  the  Araxes  valley  at  a  very  aeute 
angle.  The  strata  overlie  grey,  marly  limestones 
of  the  Middle  Devonian  with  the  same  strike,  and 
were  foimd  by  Freeh  f  to  contain : — 

Spirifer  Vemeuili^  Murcfa.,  S.  Archiaci^  Murch. ,  S. 
tenticulum^  Murch. ;  Rkynchonella  sp.  indet. 

At  Khorvirab  Upper  Devonian  (overlying  Middle 
Devonian  limestone,  is  indicated  by  the  presence  of 
Spirifer  Archiaci,  March.,  S.  aperturatus,  Schl.,  and  S. 
ostiolatus^  Schl.,  in  Abich's  list  of  fossils.:]: 

In  the  Dsynlerly  Dagh,  whidi  presents  a  fault- 
scarp  on  the  left  bank  of  the  Araxes,  the  same  Upper 
Devonian  limestone  occurs  with  Spirifer  Archiad^ 
Mmch.,  and  S.  Vemeuili^  Moroh.,  overlain  by  Carboni* 
ferous. 

*  (7W.  QOtti.  lyiU  Mmmm^  X09. 
t  Op.  Hi.  162, 171, 17s. 

t  Vtrginekemdi gi$l.  GfmnM^$k.  78;  BM.  See.  GM.  Fhma^  (*) 

*  138;  Qftart.jmm^  CM  Ar;  ii.  93. 


Upp0r  Dewmian  agf 

Further  east^  a  miles  aorth  of  Yaiji  in  the  Darala- 
gdz  distrid,  the  fossils  Spivijet  Vemeuili  (disjundtus)^ 
Murch.9  and  Acertmlaria  pentagona^  M-E.,*  ^ow  the 
presence  of  Upper  Devonian  in  addition  to  the  Middle 
oeds  already  referred  to  (p.  289).  Carboniferous 
Limestone  overlies  the  series* 

In  pebbles  of  the  Lower  Arpa-chai  (Daralagoz) 
Freeh  t    coUedled    the    following    Upper    Devonian 


Spirifer    Vemeuiliy  Hnrcb.,    S.  tenticulum^  Sow.; 

Praductella. 
Primitia. 
Bellerophon  Verce^  Frch.;  Naticopsis  Marthas^  Frdi. 

The  Spirt fer  aperturatuSy  Schlut.,  and  the  Middle 
Devonian  Euotnphalus  WcAlenbergi,  gathered  by  Dr. 
Radde«  and  now  in  the  Breslau  Museum,  were  also 
derived  from  pebbles  in  the  Arpa-chai.:}; 

It  is  not  improbable  that  these  fossils  originally 
came  from  Ashaga-burun  higher  up  the  river,  where 
not  only  Middle  Devonian  fossils  have  been  recorded 
(p.  290),  but  the  following  Upper  Devonian  forms 
sdso  occur  § : — 

Acervularia  cf.  limitata^  M-E. 
Rhynchanella  (Wilsonia)  cubaides^  Sow. 

The  following  Upper  Devonian  beds  also  under- 
lie the  Spirifer  tamacensis  zone  at  the  confluence  of 
the  Daralagoz  Arpa-chai  with  the  Araxes  || : — 

I.  Black  daj-slate  and  mail,  with  Spirifer  VtrnemH^  Mmch., 
and  Spinftr  Vtrmmli,  var.  BrM^  Wenj. 


*  Gtol.  Cdtal,  lyUs  Jfuatm,  92. 

t  Op.  nt.  197. 

t  Fredi,r^.  ai.  173,  and  LMaa  GtopnosHa^  Bd.  3|  lief.  L  145. 

(  Gtol.  CaM.  Tiflis  Mu$mm^  109. 

J  LUhaa  GtognoiU  .••  L  196. 


aga  Geology  of  Armenia 

9.  Similar  beds,  with  much  zoQed  iprdmens  of  Sftn/er 
Vtnumiii^  MmdL 

Some  Upper  Devonian  and  Carboniferous  bedfr 
come  to  view  beneath  the  Permian  of  the  Karmirvank 
-Julfa  gorge  of  the  Araxes  (Chapter  XVII).  Abich  * 
found  here  the  Upper  Devonian  forms  Spirifer  Archiaci^ 
MiHch.)  and  S.  Verneuili^  Much.  The  list  of  fossils 
from  Karmirvank  (now  in  the  Tiflis  Museum!)  shows 
not  only  the  presence  of  the  following  Upper 
Devonian  species,  but  also  indicates  the  overlying 
Lower  Carboniferous  {vide  infra) : — 

Spirifer  Archiad^  Murch.,  S.  Vemeuili  [disjunc- 
tus]^  Murch.;  Rhynchonelia  (Wilsonia)  cuboides^ 
Sow.,  JR.  inauritay  Sandb.,  R.  VersilofH^  Vera. 

Phacops  bretncauda,  Sandb.;  Cypridina. 

Pterinea;  Area;  Cardiomorpha^ 

IT.      LOWKR  CAMBOmmOVS. 

The  Carboniferous  limestones  are  very  similar  in 
colour  and  strudlure  to  the  Devonian  limestones. 
But,  according  to  Abich,^  lighter  colours  usually 
prevail,  and  they  are  always  more  or  less  bituminous, 
yielding  the  chara(5leristic  odour  when  struck.  Alum 
also  is  sometimes  present.§  In  general,  the  Carboni- 
ferous beds  are  rather  poor  in  fossils. 

There  is  a  gradual  transition  between  the  Devonian 
and  the  Carboniferous;  the  lower  boundary  is  cha- 
radlerized  by  Spirifer  tenticulum^  Sow.,  and  was  found 
by  Freeh  II  to  be  well  exposed  at  Norrashem,  in  the 

*  Vergkkhenie  geol.  Grundsugi  Crt.  78. 
t  Geoi.  Catal.  TifiU  Muswm,  176. 
X  op.  Hi.  79. 

S  Giewingk,  Dr.  C.C.A.,  Giogtmt.  6forpgra^.  Verkalin.  its  wnrU. 
PirsiiHSy  37. 

II  Das  Palaaaoicum  in  acharmemm  *•  Pmim^  :Beitr.  Pal.  (EstTt- 
Vng.  xiL  179,  zSs,  198. 


Lower  Carboniferous  293 

transverse  valley  of  the  Daralagoz  Arpa-chai,  at  its 
confluence  with  the  Araxes  (cp.  p.  291). 

This  Lower  Carboniferous  series  of  brachiopod- 
limestones  may  be  divided  into  a  lower  zone  with 
Spirifer  tomacensiSf  and  an  upper  zone  with  Productus 
giganteus. 

A.  Zone  of  Spirifer  tornacensis : — Marly  limestones^ 
marls,  and  slates,  with  Spirifer  tornacensis^  de  Kon., 
and  S.  tenticulum^  Saw.  (this  is  rare  here). 

(i)  Upper  part  with  Dalmanella  Michelini,  l'Ev., 
in  great  abundance : — 

Numerous  Crinoid  stems — PachylocrinuSy  Platy- 

crinus. 
Productus  perlonguSy  Pand.,  mut.  armenica^  Freeh,; 
Orthothetes  crenistria,  M'Coy  (frequent) ;    Athyris 
ambigua^  Sow. 
The  following  occur  more  rarely  in  this  bed : — 

Cyathaxonia  cornu^  Mich.;  Syringopora. 
Griffithides  longiceps. 

Chonetes  hardrensis^  Phill.;  Leptoena  analoga^  Phill.: 
Spirifer  aff.  rotundato,  Sow.;    S.  aff.  pleno.  Hall ; 

Spiriferina    Moelleri^    Kon.;    Athyris    Royssii, 

L'Ev.,  mut.  tornacensis^  Freeh. 
Temnocheilus  armeniacunty  Freeh. 

(ii)  In  the  lower  part  Dalmanella  Michelini  is  rare 
or  wanting.  Productus  perlonguSy  Pand.,  mut.  armentca, 
Freeh,  P.  perlongus  cf.  spinulosus^  Sow.,  and  Athyris  am- 
bigua^  Sow.,  are  frequent.    The  following  are  rarer  : — 

Produ£ius  Raddeanus^  Freeh.;  Orthothetes  cren- 
istria^  var.  Kelli^  M'Coy ;  Leptoena  analoga^ 
Phill. ;  Spirifer  tenticulum^  Sow.  ;  Rhynchonella 
pleurodonj  PhiU.,  &  var.  Davreuxiana^  de  Kon. 

Bellerophon. 

This  Spirifer  tornacensis  zone  is  evidently  represent- 


294  Geology  of  Armenia 

ed  by  the  fossils  found  by  Tsulukidse  *  in  bituminous 
limestone  at  Bash  Nurs^in  (apparently  the  same 
locality) : — 

Alveolites  ramosus^  Stefaun. 

Spirigera  concentrica^  d'Orb.;    Athyris  Herculea^ 

Barrande. 

Pleurotomaria  cf.  subclathrataj  Sandb. 

Higher  up  the  Arpa-chai,  the  Spirifer  tomacensis 
zone  is  evidenced  by  the  occurrence  of  this  fossil  (the 
Spirifer  calcaratus^  Sow.,  of  Abich)t  in  a  limestone 
at  Ashaga-burun,  near  Ashaga-dansik,  on  the  right 
bank  of  the  river,  overiying  Upper  and  Middle  De- 
vonian beds  (pp.  290,  291). 

A  specimen  f  of  Spirifer  tornacensis  (calcaratuSy  Sow., 
of  Abich)  from  the  Dsynleriy  Dagh  (p.  290)  shows  the 
presence  of  this  zone  further  west. 

To  the  south,  this  zone  clearly  appears  in  the  gorge 
of  the  Araxes,  at  Karmirvank,  overlying  the  Upper 
Devonian  (p.  292).  Here  Abich  §  found  Spirifer 
tornacensis,  de  Kon.,  (S.  calcaratuSj  Sow.),  and  Leptcena 
Dutertri;  and,  in  addition  to  the  first-named,  the 
following  out  of  the  list  of  fossils  coUedted  from  the 
same  locality  by  Radde  and  Sievers||  belong  to  the 
same  zone : — 

Orthothetes  crenistria,  Phfll.;   Spirigera  concentric 
ca,  d'Orb.;  Leptcena ;  TerAratula. 

It  may  be  noted  that  this  zone  with  Dalmanella 
Michelini,  L'Ev.,  and  Orthothetes  crenistria,  Phill.,  occurs 
in  Palestine  and  the  Sinai  peninsula.^ 

*  Geal.  Catai.  Tiflis  Museum,  109. 

t  Ibid.  109. 

t  VergUichendi  ged.  GrumdaugibH.  78. 

\  Ibid.  78. 

II  Geol.  Catal.  lijlu  Museum,  176. 

T  Tate,  R.,  Qfiari.  Jauru.  Geol.  Sec.  zzfiL  404. 


Lower  Carboniferous  295 

B.  Zone  of  ProduAw  giganteus: — ^Thickly  bedded, 
partly  bituminous,  and  somewhat  gypsiferous  lime- 
stone at  the  new  Aipa-chai  bridge  at  Norrashem. 
Freeh  *  found  it  to  contain  the  following : — 

Numerous  corals,  e.^.,  Lonsdalia  Araxis^  Freeh. 

Echinid  spines  and  Crihoid  stems. 

Productus  semireticulatus^  Mart,  P.  Keyserling- 

ianus^  de  Kon. 
Euomphalus  pentangulatuSf  Sow. 

Both  the  last  two  appear  now  for  the  first  time. 

It  was  evidently  from  these  beds  that  Tsulukidse  t 
colledled  the  following  fossils^  They  occurred  in  a 
dark,  bituminous,  foetid  limestone,  overlain  by  Fusu- 
Una  limestone,  between  Bash  Nurashin  and  Yaiji,  on 
the  Arpa-chai : — 

Lithostrotion  Martini^  M-E.,  L.  Portlocki^  M-E.; 

Lonsdalia  papillatay  M-E.,  L.  floriformiSj  M-K 
Michelinia  favosa^  Gf. 
Euomphalus  pentangulatus,  Sow.;  Bellerophon. 

At  Ashaga-dansik,:]:  higher  up  the  river,  the  same 
horizon  occurs  in  bituminous  limestone  (overlying  the 
Spirifer  tornacensis  zone,  p.  29A),  with  Lonsdalia 
fhrijormisj  M-E.,  and  Bellerophon.  In  the  same  distridl, 
a  miles  north  of  Yaiji,  Tsulukidse  §  found  a  coral- 
limestone  with  Zaphrentis  cylindrical  Seoul.,  evidently 
overlying  limestones  with  Upper  Devonian  fossils 
(P-  290), 

At  Karmirvank,  in  the  Araxes  gorge,  the  following 
fossils  colleded  by  Radde  and  Sievers|{  show  tlite 

t  GioL  Catal.  lYJUs  Musmm^  99. 
%  JbUL  no. 
{  Ibid.  92. 
R  IbU.  176. 


296  Geology  of  Armmia 

presence  of  this  zone,  overlying  the  previous  zone : — 

Lithodendron   f(isciculatum^    Reys.;     Calamopora 

polymorpha. 
Productus  pustulosuSj  Phill.,  P.  produ£toides^  Murch., 

P.    sulcatus^    Sow.,     P.    siibocttieatus,    Murcfa., 

P.  caYhonarius^  de  Kon. 
Pleurotomaria. 
Goniatites. 

It  occurs  also  on  the  south  side  of  the  Araxes, 
underlying  the  FttSttWna-limestone — both  being  partly 
c:overed  up  by  the  lavas  of  Little  Ararat.  According 
to  Abich*  the  compadl,  often  marmoraceous  lime- 
stone contains  corals,  apparently  in  reefs,  chiefly  of 

Lonsdalia  floriformisj  M-E.;    Lithostrotian  incon- 
fertum ;  ana  several  species  of  Michelinia. 

Numerous  smooth  species  of  TerAratula  were  also 
found  by  Abich  in  these  beds,  together  with  Spirigera 
planosulcata^  d'Orb.,  occasionally  in  great  numbers.  In 
the  hemicrystalline'  limestone  of  the  Maku  Mountains 
^•E.  of  Ararat)  this  species  occurs  together  with 
lerebratula  Salteri,  Abich. 

At  Khorvirab  (N.  of  Little  Ararat)  Abich  t  found 
dark  Carboniferous  Limestone,  "on  the  surface  of 
which  gigantic  individuals  of  Zaphrentis  cylindrical 
ScouL,  stand  out  conspicuously  in  white  calcspar,"  (cp. 
p.  295).  It  overlies  limestones  with  Middle  and  Upper 
Devonian  fossils  (pp.  289,  290). 

The  Produdius  giganteus  zone  was  discovered  still 
further  south  by  Fohlig  f  at  Daniel  Begamish  (on  the 
way  from  Julfa  to  Khoi,  by  way  of  the  Kotur-chai), 
and  also  on  the  shores  of  Lake  Urmi.    The  limestone 

*  VergUichefide  geol.  Grufukuge  etc.  79,  80;  and  his  Ged.  Fonck. 
in  den  kauk,  Ldndem^  i.  za6. 

t  GeoL  Rrsch,  drv.  u.  459. 

X  Verk.  i.'k.  lUkksMst  1884,  aSi. 


Lofwer  Carboniferous  297 

of  the  former  locality  contained  the  following, 
chaiadleristic  Western  European  fossils,  now  in  the 
HaUe  Museum  and  identified  by  Freeh  :* — 

CyathophyUum  Murchisoni^  M-E.;  Syringopora. 
Productus  giganteuSy  Mart.;  Spirifer  strtatus^  Sow., 

S.  triangularis^  Sow.;   S.  (Syringothyris)  cus- 

pidatuSf  Mart. 

Pohlig's  assumption  that  Devonian  fossils  also 
occurred  here  was  erroneous;  his  Stringocephalus 
Burtini  from  Darosham  Evaghlu  was  found  by  Freeh 
{lx^  to  be  a  large  Athyris  of  Carboniferous  habit. 

ft  is  probable  that  the  bituminous  limestone  found 
by  Wagner  t  on  the  west  coast  of  Lake  Urmi,  near 
Guchi  ^ushtchi),  8  hours  from  Urmi,  may  belong  to 
the  same  horizon: — "It  is  a  compacft,  bituminous, 
dark,  limestone,  giving  a  bituminous  smell  at  the 
slightest  rubbing.  It  occurs  also  further  along  the 
coast  towards  Guverchin-kala.'* 

Both  the  zones  of  Spirifer  tornacensis  and  Productus 
giganteus  were  found  by  Tietze  in  the  Demavend 
region  of  the  Alburs  range  and  they  have  also  been 
observed  by  Stahl  and  Tietze  J  further  east,  between 
Asterabad  and  Shahrud.  At  Asterabad  the  Produdius 
zone   occurs    underlying    Permian,    and    contains 

Productus  ovaliSf  Waag.;  Dalmanella  indica,  Waag.: 
Orthothetes  persicus^  ScheUw.§ 

The  total  thickness  of  the  Upper  Devonian  and 
Lower  Carboniferous  in  North  Persia  has  indeed  been 
estimated  at  3400  metres.  These  two  zones  have  also 
been  observedf  N.W.  of  Teheran ;   and  in  the  Kuhrud 

*  Op.  Hi.  305 ;  Ncues  Jahfhichf  1895,  ii.  58;  and  Ltthaa  Gicignast. 
\L  382. 

t  Jieue  nach  Penun^  iL  167. 

%  Lethaa  GtognostUa^  ii.  434. 

{  IHd.  ii.  423. 


298  Geology  of  Armenia 

Mountains^  between  Kashan  and  Isfiahan,  both  tha 
Upper  Mid-Devonian  and  Upper  Carboniferous  are 
present.* 

T.      UFPBR   CAKBOVimOUS. 

The  highest  Carboniferous  beds  of  Armenia  and 
Persia  belong  to  the  Fusulina  zone  (Moscow  stage)  of 
the  lower  beds  of  the  Upper  Carboniferous.  This 
zone  according  to  Abich  t  is  composed  of  dark  grey, 
very  compadt  limestone  with  Fusulina  sphoeroidea^ 
(Abich)  Mollcr.  It  occurs  particularly  in  the  mid-Araxes 
plain,  everywhere  conformably  overlying  the  Lower 
Carboniferous.  It  is  a  zoogenous  rock,  full  of  the 
Fusulina,  with  masses  of  Bryozoa  and  calcareous 
Algae,  and  species  of  Belleropkon.  The  Bellerophon 
shells  always  consist  of  calcspar,  and  stand  out  in 
relief.  The  small  individuals  much  recall  JB.  striatus. 
The  above  species  of  Fusulina  has  a  similar  im- 
portsince  in  distribution  in  Armenia  as  F.  cylindrica 
has  in  the  Carboniferous  Limestone  of. Russia.  A 
variety,  however,  of  the  latter  species  was  found  by 
Abich  (I.e.)  also  in  Armenia,  beneath  the  lavas  of 
Little  Ararat,  but  it  seems  to  be  much  rarer. 

This  horizon  occurs  not  only  north  of  the  Araxes  in 
the  Daralagoz  distriA  (at  Yaiji,J  Yukhary-dansik,§ 
and  Gyneshik  ||),  but  also  on  the  south  of  the  river. 
Abich,^  in  travelling  from  Julfa  to  Tabriz,  ascended 
"over  terraces  of  Devonian  and  Carboniferous  lime- 

*  Zetkaa  GtognosHca^  ii.  383. 

t  Geol.  Forsch.  in  den  hauk.  Laniem^  L  ia6;  and  his  Vergleickendi 
geoL  Grundsiige  €U.  79. 

X  GioL  Catal,  TiJBs  Museum^  93. 

{  Ibid.  zxo. 

II   ViTgieichende  g€ot.  Gnmdw&gibH.  8o. 

\  Bull,  phyu-math.  Ami.  Set.  St.  PHmt.  xvi.  (1858),  3x41  J^S* 


upper  Cafbantferaus  299 

stone  with  Fusulina  sphceroidea  (spherica)^*'  overlain 
by  Cretaceous  beds  at  Gerger;  and  further  south, 
in  the  Asam  valley,  nearer  to  Marand,  this  Carboni- 
ferous limestone  is  much  traversed  by  ores  of  man- 
ganese, in  the  neighbourhood  of  intrusions  of  a  dark 
brown  porphyry. 

This  foraminiferal  limestone  occurs  as  the  highest 
part  of  the  Carboniferous  series  on  the  Urmi  plateau, 
immediately  underlying  the  Miocene  limestone.  It 
was  noticed  by  Abich  *  at  a  spot  nearly  40  kilometres 
from  Urmi,  on  the  road  to  Gavalan.  Here  a  sprini 
rises  out  of  dolomitic  Carboniferous  Limestone  witi 
Fusulina^  BelUrophon  (a  small  species)»  and  corals. 
On  the  shores  and  islands  of  Lake  Urmi  this  horizon 
comes  to  view  as  a  marmoraceous  limestone,  slate- 
coloured  and  with  numerous  calcite* veins ;  its  dark 
colour  stands  in  sharp  contrast  to  the  overlying,  white 
Miocene  limestone.  On  the  island  of  Ispir  Abich  t 
found  it  to  be  characterized  by  Fusulina  and  Bellero- 
)hon.  A  specimen  of  the  limestone  from  the  islet  of 
Ihazalan  fa  .little  north  of  Koyundaghi)  was  colledled 
by  Mr.  R.T.  Giinther  and  was  examined  under  the 
microscope  by  Mr.  R.  Bullea  Newton,  F.G.S.4  who 
found  it  to  be  mainly  composed  of  Foraminifera, 
especially  of  the  following  forms : — 

Endothyra  Bawmanni,  Phillips;    Nodosaria  radi- 
cular Linn.;  Volvulina  hulhides^  Brady. 

^  BuU.  Soc.  Ghi.  FfmUi.  (a),  nd.  317. 
t  Varh.  k.-k.  gM.  Rricksanst  rSy/,  67. 
%  Jaum.  Zdnn,  S^,  xmi^  4St. 


CHAPTER  XVn 

THE  PERMO-TRLA^SIC  (jULFA  BEDS)  OF  ARME^XIA 

A  gap  occurs  in  the  geological  record  of  Armenia 
between  the  lower  beds  of  the  Upper  Carboniferous 
and  the  lower  beds  of  the  Upper  Permian.    A  con- 
siderable amount  of  folding  took  place  during  this 
period,   and  parallel   N.W.-S.E,   ranges   came    into 
being,  from  the  Hercynian  chains  of  Silesia  to  Ar- 
menia and  Persia.    It  was  probably  at  the  same  time 
that  the  Caucasus  became  dry  land,  for  Inostranzeff  ^ 
found  (on  the  north  side  of  the  chain)  a  Catamites 
in  the  Palaeozoic  schists,  which  are  overlain  uncon- 
formably  by  the  somewhat  similar  Liassic  schists; 
here,  as  in  Armenia,  the  moimtain-formation  con- 
tinued during  the  Trias. 

Immediately  subsequent  to  the  Permo-Carboni- 
ferous  earth -movement,  which  raised  a  great  part  of 
Armenia  into  dry  land,  the  Permo-Triassic  beds  of 
Julfa  on  the  Araxes  were  deposited,  probably  in  a 
[ulf  of  a  sea,  which  stretched  across  to  India.  These 
^eds  have  roused  considerable  interest  and  discussion 
among  palaeontologists.  Abich  f  at  first  described 
the  Julfa  fauna  as  Upper  Carboniferous,  but  even 

*  Au  iravers  de  la  chatm  frindpaU  iu  Caucase^  245*    St.  Peten- 
buxg,  1896. 
t  Geal.  Farsch.  m  den  kaidt.  Landem^  i. 


PermO'Triassic  301 

in  his  appendix  he  modified  his  views  so  far  as  to 
consider  it  Permo- Carboniferous.  Mojsisovics^  went 
a  step  further  in  stating  that  the  Triassic  diarader  of 
the  Ammonites  seemed  to  point  to  the  uppermost 
rone  of  the  Permian  with  Lower  Trias,  whilst  the 
brachiopods  and  corals  on  the  other  hand  appeared  to 
indicate  Lower  Permian  or  even  Upper  Carboniferous. 
Freeh  and  Arthaber,t  however,  have  in  recent  years 
been  able  to  carr>'  out  a  thorough  investigation  of  the 
locality  and  its  fauna ;  the  present  chapter  is,  in 
the  main,  a  r^um6  of  their  work. 

At  Karmirvank  the  Araxes  passes  from  the  broad, 
longitudinal  valley  of  Nakhichevan  into  the  ever 
deepening  canyon.  Its  upper  part  is  cut  through 
red  sandstones  and  limestones  of  Eocene  age ;  it  is 
only  at  Negram  that  the  natural  sedtion  of  the 
Permian  limestones  is  laid  bare.  Diredlly  east  of  the 
Negram  fort  a  brownish  or  yellowish,  indistindly 
bedded  quartzite  forms  the  uppermost  part  of  a 
succession  corresponding  to  Upper  Permian  and  Lower 
Trias.  In  contrast  to  the  zone  of  Palaeozoic  folds 
in  the  mid-Araxes  region,  free  from  contemporary 
eruptive  rocks,  these  later  beds  show  only  flexures, 
with  a  general  steep  dip  to  N.E.,  and  with  numerous 
sills  of  no  great  thickness. 

The  whole  succession  in  the  Araxes  gorge  may  be 
tabulated  thus : — 

1.  Middle  Eocene. 

2.  Lower  Trias  (Bunteri :— Q^iartsite  of  Negram,  apparent- 

I7  with  interbeoding  of  Wibi-limestone  with  IVer- 
fen  fossils. 

3.  Uppkr  Permian  (Upper  Zechstein) : — Limestone  wilh  the 

*  Verh.  k.'k,geoL  Reichsamt.  1879,' 17. 

t  Das  Paldozoicum  in  HochamnnUn  6*  Ptrrien^  Beitr.  Pal.  (Estr.- 
Ung.  xii.  161 ;  Neuts  Jahrbuch^  1895,  ii.  58;  and  LUhma  Gtognost  li. 
421,  566. 


302  Geology  of  Armenta 

9:pptBXUiat  of  ff^i/iM-liiiiestone,  and  containing 
Chideru  fosali. 

4.  Upper  Pxrmian  (Lower  Zechstein) :— Julia  limestones 

with  a  rich  fauna,  especially  of  dratUida:    And 
Bnichiopoda.    Stage  of  Otoaras  fulfmu. 

5.  U|V>er   Carboniferous   ujomn  beds): — ^Daric   dolomitic 

limestone  with  Bukrophtm. 

L      UPPEB  PIRMUN  (lOWXR  ZICHSTIUr). 

About  2  miles  east  of  the  Negram  fort  the  louver 
beds  of  the  Upper  Permian  begin  to  be  visible    in 
the  form  of  grey,  unfossiliferous  limestones,  in  thick 
beds,  dipping  25^-35°  N.E.,  with  a  strike  of  N.  30"*  W. 
Between  the  verst-stones  3^  and  4  there  is  a  sill  of 
quartz-porph}^,  only  a  few  metres  thick,  but  in- 
creasing in  thickness  lower  down  the  valley.    No 
megascopic  signs  of  contadl-adion  were  noticeable  in 
the  limestones.    Between  the  verst*stones  4^  and   5 
two  more  sills  on  the  Persian  shore  were  again  seen 
from  a  distance.    A  little  beyond  the  5th  verst-stone 
on  the  road  there  is  a  bed  (2  metres  thick)  of  a  de- 
composed diabase,   which   has   distindUy   metamor- 
phosed the  adjacent  rock. 

After  6  versts  (4  miles),  just  before  reaching  the 
Darosham  fort,  the  Araxes  flows  for  some  distance 
exadly  along  the  N.W.-S.E.  strike  of  the  beds.  On 
the  Persian  bank  the  strata  dip  50^  towards  the 
river,  and  in  consequence  a  landslip  has  occurred. 
The  confluence  of  the  Persian  Kotur  river  (Kisil-chai) 
takes  place  just  opposite  an  extensive  alluvial  fan,  on 
which  the  fort  Daroshinsky  is  built.  Below  this  fort 
the  Julfa  limestones  form  a  fine,  typical  flexure ;  its 
steeper  side  dips  70"^  to  the  river  u.e.,  to  E,N.E.),  and 
is  even  bent  over  westward.  The  fossil  locality 
occurred  in  the  steeper  beds  (see  the  authors'  sedlions). 

At  one  verst  below  the  Daroshinsl^^  forts,  the  fifth 


upper  Permian  [Julfa  Beds]  303 

bed  of  eruptive  rock  (?  diabase)  was  observed  on  the 
Persian  bank,  and  then  3^^  and  5  versts  further  on  two 
more  diabase-sills  were  visible.  The  lowest  rock  is 
everywhere  a  marly  unfossiliferous  Plattenkalk. 

About  halfway  between  the  Darosham  fort  and 
Upper  Julfa  (3I  miles  from  the  first  fort,  4  miles  from 
the  last  fort)  lies  the  locality  whence  Abich  obtained 
his  fossils  (his  section  should  be  reversed) ;  it  occurs 
about  8  minutes  above  the  patrol-path  on  the  Russian 
bank.  The  dip  here  is  60°  to  N.E.,  and  the  strike 
N.  35^  W. 

The  fossiliferous,  marly  limestones  are  8-10  metres 
thick,  and  yield  most  fossils  on  a  small  hill  under  the 
cliff  on  the  Russian  bank.  The  Brachiopods  weather 
out  fre^y,  and  the  rarer  Ammonites  are  equally 
distributed  throughout  the  strata.  There  is  no  petro- 
logical  or  palaeontological  differentiation  in  these  8- 
10  metres  of  limestones.  The  hypothetical  separation 
of  a  deeper  horizon  with  Brachiopods  and  Gastrioceras 
from  a  higher  Otoceras  zone  cannot  be  maintained. 

The  most  frequent  fossils  of  the  Julfa  limestones  are 
Spirigera  protea  and  Proiuctus  intermedium  with  their 
varieties ;  Marginifera  spinosocostatus  and  Amplexiis 
Abichi  are  also  fairly  abundant.  The  following  is 
a  complete  list  by  von  Arthaber  of  the  fauna,  em- 
bodying Abich's  determinations,  many  of  which  have 
undergone  revision;  those  marked  "J"  are  peculiar 
to  Julia : — 

Amplexus  Abichi^  Waagen  St  Wentzel ;  Zaphrentis 
( Pent aphy Hum)  leptoconica^  Abich. 

Favosites  Jabiensis,  Waag.  &  Wentz.;  Michelinia 
Abichi,  Waagen  &  Wentzel. 

Cyathocrinus  cf.  ramosus,  Schlot.;  C.  virgalensis, 
Waagen;  Poterioctinus. 

Polypora. 


304  Geology  of  Armmia 


Produ6ius  intermedius,  Abich  (J),  P.  Abichi,  Waag., 
P.  Waagenij  Rothpl.,  P.  hemisphcerium^  Kut., 
P.  mytiloideSf  Waag.;  Marginifera  spinosocosta- 
ta,  Abich,  (J),  Ibid.  var.  expansa,  Abich,  (J),  Ibid, 
var.  incurva,  Abich,  (J),  /fe/d.  var.  ctKata,  Abich,  (J), 
M.  intermedia  helica,  Abich  (J);  Martinia 
planoconvexa^  Schum.;  Sptriferina  cristata^  Schlth.; 
Reticularia  cf.  pulcherrima^  Gemm.,  R.  indica^ 
Waag.;  jR.  Waageni^  Locsy;  Spirigera  protea^ 
Abich,  (J),  /6id.  var.  quadrilobatd^  Abich,  (J), 
/6td.  var.  multilobata,  Abich,  (J),  /6id.  var.  aZata, 
Abich,  (J)  S.  globulariSf  PhiU.,  S.  subtilita  var. 
armeniaca,  Arth.  (J),  S.  /rfma,  Arth.  (J),  S.  i46tcW, 
Arth.  (J) ;  Uncinulus  Jabiensis^  Waag.,  U.  Wich- 
manniy  RothpL;  Notothyris(Rhynchonella)  jul- 
fensiSf  Abich  (J);  Dalmanella  indica,  Waag.; 
Orthothetes  armeniacus,  Arth.,  0.  eusarcos,  Abich, 
(J),  0.  peregnnus,  Abich  (J). 

Pseudomonotis  sp.  indet. 

Macrocheilus  avellanoides^  de  Kon.  (Buccinum  in- 
cerium  of  Abich). 

Orthoceras  annulatum  var.  crassttwt,  Foord  ;  0.  W- 
cindium,  Abich  (J),  0.  transt^i^rsiim,  Abich  (J)^ 
0.  oblique-annulatum,  Waag.;  Nautilus  comutus^ 
Golov.  {N.  concotms,  Sow.  of  Abich),  AT.  paralU 
ellusy  Abich  (J);  Pleurouautilus  aff.  Wynnei^ 
Waag.,  P.  dorsoarmatus,  AWch  (J),  P.  Pichleri, 
P.  y^rcp,  Arth.  (J),  P.  sp.  indet.  ("AT.  tubercu- 
lariSy  Abich);  Coelonautilus  sp.  indet.  fN. 
armeniacus,  Abich),  C.  dorsoplicatus,  Abich  (J) ; 
Gastnoceros  Abichianumf  MoUer  (J) ;  Hungari- 
tes  Raddei,  Arth.  (J) ;  // .  pessoides,  Abich  Q) ; 
Otoceros  Fedoroffi,  Aith.  (J),  0.  trochoides,  Abich 
f  T),  0.  julfense,  AWch  (J),  0.  tropitum,  Abich  (J). 


up.  Permian  [Chideru  Stage]  &  Low.  Trias  305 

n.    uppnt  FiRiiiAK  (urpsB  iicb8iixN|  CHmnu  tTAOi). 

A  mile  east  of  the  Negram  fort  the  Julfa  series  is 
overlain  by  grey  marl  and  slaty  limntone  (Flatten- 
halk),  whidi  agrees  petiographically  with  the  German 
Wellenkalk^  but  the  following  fossils  indicate  the 
horizon  of  the  Indian^  Chideru  beds,  i.e.^  ot  the 
uppermost  marine  Permian : — 

Lima  Footei,  Waag.;   Aviculopedien  sp.  indet; 

Lithodomina  aUthreviata^  Waag. 
Pleurotwnaria  cf.  punjdhicay  Waag. 

IIL     LOWn   TRIAS. 

The  quartzite  of  Negram  dips  20^  N.E.,  and  shows 
a  strike  of  N.  25**  W.  It  overlies  the  Permian  beds  at 
the  last-named  locality,  apparently  quite  conform- 
ably. The  Werfen  fossils  which  were  determined  by 
Mojsisovics  (l.c.)  probably  came  from  this  quartzite. 
They  comprise  the  following  forms : — 

Pseudomonotis  Clarai. 
Tirolites. 


CHAPTER  XVIII 


THE  JURASSIC   OF  ARMENIA 


I.      OINBRAL  DISTRIBUTION. 


Up  to  the  present  time  beds  of  indisputable  Jurassic 
age  have  only  been  found  on  the  northern  and 
eastern  confines  of  the  Armenian  plateau,  vim.^  in 
the  Tkvibuli  and  Zirula  domes  of  the  Central 
Caucasus,  in  the  border-ranges  to  the  north  and 
east  of  Lake  Gokcha,  in  the  Karabagh  region,  and  in 
Azerbaijan  near  Lake  Urmi.  No  Jurassic  has  yet 
been  discovered  either  in  the  Pontic  Range,  or  in 
the  Taurus  and  Zagros.  The  Armenian  plateau  itself 
is  altogether  free  from  Jurassic  deposits,  yet  it  is 
not  improbable  that  traces  may  eventually  be  dis- 
covered in  the  Dersim  region  since  Lias  has  been 
recorded  further  west,  from  Kessiktash  near  Angora. 

II.      LOWER  JUR^SIC  (lias). 

Rhaetic  beds  seem  to  be  absent  in  the  Armenian 
area,  although  they  are  well  developed  in  the  Alburs 
Mountains  of  North  Persia,  and  in  the  Caucasus  near 
Derbend.  Liassic  beds  occur  in  the  dome-like  masses 
lying  between  the  northern  border-ranges  of  Armenia 
(the  Thrialetic-Imeretic  Mountains)  and  the  Caucasus. 
Both  the  nucleus  of  the  Tkvibuli  dome  and  those 


Lip$  307 

of  the  Khyeiti  and  Vani  domes  (to  the  N.E.)  coosist— 
aocordiog  to  Foumier* — of  '-'schistose  cla>'s  of  the 
Lias,  much  folded,  and  overlain  by  clays,  psammites, 
and  plant-sandstones  of  the  Lower  Oolite.  Lignitic 
l^eds  are  intercalated  in  several  places  in  the  argillo- 
arenaceous  beds  of  the  Lias.  The  arenaceous  parts 
frequently  contain  indeterminable  plant-impressions. 
Limestone  beds  are  also  intercalated  in  tiie  series 
and  contain  the  following  tossils : — 

Terebratula  resupinata^  Sow.,   T.  putuHata^  Sow.. 

T.  subpunctata^  3m. 
Pectettr  ^  *• 

These  fossils  point  to  the  Channouthian  horizon. 
Abichf  and  Neuinayr^  have  cited  Liassic  fossils 
in  the  marly  limestones  and  even  in  the  red  limestones 
of  Ziruli  (E.  of  Kutais)  and  Katzkhi  (S.W.  of  Jiatura). 
But  Foumier  states  (Lc.)  that  ^*  these  marls  are  hot 
only  above  the  Toarcian,  but  alsoabove  a  part  of  the 
argillo-arenaceous  plant-beds.  They  represent  the 
Bajocian,  Bathonian,  Callovian,  and  Oxfordian.V  In 
another  place  §  he  goes  on  to  say  that ''  the  persistence 
of  Liassic  types  in  the  Middle  and  Upper  Jurassic 
of  the  South  Caucasus  has  been  suspe^fled  by  all 
the  geologists  who  have  studied  this  region  ;— Abich, 
Favre,  and  Simonovitch  haye  each  cited  Lia$sic^fprms 
up  to  Oxfordian  limestones;  but  nearly  all  these 
forms  occur  in  the  red  Bajodan  limestones;  if  they 
had  been  derived  from  the  grey,  argillaceous  lime- 
stones or  schistose  clays  of  the  Lias  their  colour  would 
have  made  clear  the  difference  in  the  age  of  the 
fossils.   .   .   In  this  whole  region  the  Bajodao  with 

^  Descripfwn  gial.  dm  Caucam  atUrmJt  94. 
t  Bui/.  And.  SeL  St.  PUmb.  v.  (1M5),  17. 
X  Dekkuhr,  Wwur  Akad.  lix.  94,  95* 
{  Op.cU.  XIX. 


3o8  Geology  of  Armenia 

plant-remains  overlies  the  Toarcian  with  the  follow- 
ing Central  European  fossils : — 

Rhynchonella  tetrahedra^  Sow. 
Posidonotnya  Bronni^  Volu. 
Harpoceras  cf.  boscensef  Rqm.,    H.  touarcense  (//. 
striatulum^  Sow.)/' 

Pompeckj,^  in  his  paper  on  the  Middle  Lias  of 
Kessiktash,  near  Angora,  has,  however,  followed 
Neumayr  and  Uhlig  (I.e.)  in  erroneously  regarding 
as  Middle  Lias  the  Middle  and  part  of  the  Upper 
Jurassic  beds  of  ZiruH  and  Katzkhi ;  consequently 
the  importance  of  his  generalizations  and  land-con- 
stru6lion  of  the  period  is  materially  diminished. 

The  only  known  occurrence  of  Lias  within  the 
Armenian  area  is  in  Azerbaijan,  east  of  Lake  Urmi. 
The  locality  is  Tazehkend,  about  2  miles  east  of 
Maragha;  the  specimens,  which  have  been  much 
crushed,  were  colleAed  by  Rodlerf  and  identified 
by  Weithofer4  They  indicate  a  Toarcian  horizon, 
apparently  slightly  higher  than  that  of  the  Ziruli 
Lias,  viz.: — 

Pecten  cf.  disciformiSf  Schiibler ;  Pleuromya. 

Harpoceras  cf.  radians^  Reinecke,  (typical  of  the 
Lytoceras  jurense  zone),  H.  cf.  Kurrianufn^ 
Oppel,  (marking  the  Middle  Lias  zone  of 
Amaltheus  margaritatus) ;  Belemnites. 

Rodler  also  states  {l.c.)  that  Strauss  afterwards  sent 
him  "Ammonites  from  the  Karangu  valley,  in  the 
district  of  the  Sefi  Rud  [i,e.,  to  the  eastward],  which 
clearly  pointed  to  Middle  and  Upper  Lias." 

*  Zeiischr.  deutsch.  gtri.  Gti.  zliz.  785,  794,  8x3. 

t  SiiMunpber.  k.-Jt.  Aksd.   Wiss.^  Mathem,'nahtrw.   Classic  xcWL 
804. 

t  nni.  xcviii.  756. 


Upper  Lias  309 

According  to  von  dem  Bome  *  the  fossils  of  Tazeh- 
kend  occur  in  ^^concretions  in  a  black,  bituminous, 
micaceous  and  sandy  rock ;"  his  list  differs  but  slight- 
ly from  that  of  Weithofer : — 

Pecten  disciformis^  Schtiblcr;  Mytilus  Matianus^ 
von  dem  Bome ;  Trigonia  Roxane^  von  dem  Bome ; 
Pleuromya  Urmiensis,  von  dem  Bome,  P.  aff. 
exarataj  Bnums ;  Gresslya. 

Pleurotomaria. 

Harpoceras  htropatense,  von  dem  Bome,  H.  Medice^ 
von  dem  Bome,  H.  Kapautense^  von  dem  Bome, 
(all  belonging  to  the  H.  radians  group); 
Belemnites. 

Porapeckjt  remarks  that  "the  Liassic  rock  of 
Tazehkend  vividly  recalls  the  Upper  Lias  of  Dagh- 
estan  and  the  geode-terrain  (Lower  Dogger)  of 
Chirkat  in  Daghestan."  He  is  also  of  opinion  i^ 
that  von  dem  Home's  new  species  are  probably 
merely  new  owing  to  bad  preservation,  and  that 
if  in  good  condition  would  be  found  to  agree  with 
European  forms  like  those  of  the  Lias  in  the  Alburs 
to  the  east  and  west  of  Demavend.  He  accounts 
for  the  admixture  of  Central  European  forms  by 
means  of  a  strait  through  Roumania  and  Transyl- 
vania to  the  Hungarian  sea  ;  and  finally  he  considers 
that  "  the  Caucasus  is  to  be  regarded  as  a  littoral  zone 
between  the  continent  lying  to  the  N.  and  N.E.  with 
its  terrestrial  deposits  and  coal  at  Mangishlak  (pro- 
bably an  inland  lake)  and  the  open  sea,  as  evidenced 
by  the  pelagic  sediments  of  Kessiktash."  The  Liassic 
limestone  of  Azerbaijan  also  indicates  a  more  open 
sea  than  the  Flysch-like  Lias  of  the  Central  Caucasus. 

*  Derjura  am  Ostufer  des  Vrmiasees.     Halle,  1891. 
t  Zeitschr,  deuUch.  geoi.  Ges.  xliz.  794. 
X  Ibid.  818,  830. 


3IO  Gechgy  of  Armenia 

The  Caucasus  at  any  rate  was  a£Fedted  by  a  con- 
siderable depression  during  this  period^  for  the  Lias 
always  lies  m  marked  discordance  to  the  Palaeozoic 
schists  (with  Calamites^  p.  300),  which  show  a  cor- 
roded surface  along  the  plane  of  conta(5l. 

n.     MIDDLB  jumAssia 

During  this  sedlion  of  the  Jurassic  there  was  a 
considerable  amount  of  contemporary  volcanic  activi- 
ty in  Armenia.  The  frequent  eruptions,  occasionally 
submarine,  were  inimical  to  animal  life,  of  which  only 
sparing  traces  occur  in  seams  of  impure  limestones, 
interbedded  in  thick  deposits  of  tu&,  sandstones,  and 
lava-sheets.  Tlje  Armenian  strata  were  clearly  laid 
down  along  a  cbast-line — the  sea  increasing  in  depth 
towards  Persia,  as  in  the  Lias. 

The  Lower  Bajocian  is — according  to  Foumier's 
account* — extensively  developed  in  the  basin  of  the 
Kvirila  and  the  dome  of  Tkvibuli,  immediately  north 
of  the  northern  border- ranges  of  Armenia  (the  Thrial- 
etic-Iraeretic  Mountains).  The  arenaceous  strata, 
with  intercalated  beds  of  lignite,  overlie  the  Lias 
of  this  region.  This  formation  "  is  here  two  or  three 
times  as  thick  as  elsewhere  in  the  Caucasus — the 
beds  of  lignite  are  thicker,  the  sandstones  and 
conglomerates  coarser,  and  the  plant-remains  better 
preserved.  There  is  a  discordance  between  the  Lias 
and  the  Inferior  Oolite  between  Khriseli  and  Ziro- 
vani.  The  deposits  consist  firstly  of  argillaceous 
sandstones,  containing  arenaceous  nodules  together 
with  pebbles  of  the  Liassic  beds  and  of  the  porphy rites 
and  diabases,  which  are  intercalated  in  the  Lias. 

These  are  succeeded  by  arenaceous  schists,  overlain 
by  white,  calcareous  sandstones  with  quartz- grains, 

*  Description  gM.  du  C€mcau  ienirai^  lox,  103. 


Lmver  Bajocian  311 

containing  indeterminable  plant-impressions ;  and  the 
upper  part  of  the  series  consists  of  argillaceous, 
schistose  sandstones  with  interbedded  lignites  of  very 
variable  thickness,  the  total  maximum  being  20 
metres.  .  •  The  lignite  contains  a  considerable 
quantity  of  iron  pyrites.  The  sandstones  separating 
the  beds  of  lignite  contain  the  following ; — 

Silicified    trunks    of    Cycads;     Pterophyllum 

Abichianum^    Goppert   (P.   caucasicum^   Abich); 

Zamites. 
Pecopteris  exilis^  Phill. ;    Tceniopteris  cf.  vittata^ 

Brongn.;    Asplenium    (jEthiopteris)    cf.    whit- 

biensisy  Heer ;    Camptapteris, 

The  series  of  lignites  and  sandstones  is  succeeded  by 
calcareous  conglomerates  with  pebbles  of  quartz 
and  chert,  overlain  by  argillaceous  sandstones  and 
clays." 

In  the  Eastern  Border-Ranges  of  Armenia  (the 
Gokcha  Mountains  or  so-called  Little  Caucasus),  in 
the  Bojan  valley  near  Elisabetpol  Abich  *  found  thick 
greenish  sandstones — "  diabase-psammites  " — contain- 
ing abundant,  well-preserved  impressions  of  cycads 
and  ferns,  similar  to  those  of  the  Scarborough  flora, 
and  agreeing  with  the  forms  in  the  lignites  of 
Tkvibuli  and  Daghestan,!  as  follows : — 

Pterophyllum   (Cycadites)   cf.    comptum^    iJndl. 

&  Hutt. 

Pecopteris  polyda£tyla^  Gopp.,   with  sporangia; 

Tceniopteris  tenuinervis^  Gopp. ;    Sphenopteris ; 

Scleropteris ;    Lomatopteris. 

In  similar  sandstones  of  the  same  horizon,  on  the 

western  slopes  of  the  Elisabetpol  Mountains  towards 

*  Geai.  Bead.  sw.  Kur  6*  Araxu^  58;    and  BuU.  See.  In^.  Nat. 
Moscou.  1874,  L  284. 
t  Giol.  Foruh.  in  ien  kmtk.  LSmiem^  ii.  a6. 


312  Geology  of  Armenia 

Tetter,  Abich*  found  remains  of  silicified  and  car- 
bonized Coniferous  wood,  together  with  casts  of 
marine  molluscs  (not  specified). 

Abicht  was  able  to  trace  this  series  to  the  S.E., 
forming  a  N.W.-S.E.  zone  along  the  East  Karabagh 
Range.  These  greenish,  calcareous  sandstones  and 
slaty  tuffs — "diabase  psammites  and  pelites"  of 
Abich — consist  of  triturated  fragments  of  chloritized 
and  serx)entinized  basalts,  andesites,  and  porphyrites ; 
they  contain  plant-remains  and  thin  seams  of  coal, 
and  were  probably  deposited  in  lagoons,  along  a 
coast-line,  in  the  immediate  neighbourhood  of  adlive 
volcanoes.  Abich  traversed  the  whole  series  whilst 
crossing  the  East  Karabagh  Range  by  the  road  from 
Gerjiisy  to  Shusha.  I  have  condensed  his  obser- 
vations on  the  western  side  of  the  range  down  to 
the  bed  of  the  Akera,  arranging  them  in  downward 
succession : — 

I.  Coarsely  crystalline,  dark  grey  limestone,  with  numerous 
particles  of  chlorite. 

a.  Compact,  crystalline,  dark  greenish  limestone,  with  many 
particles  of  chlorite.  Very  r^ularly  bedded;  cavities  of  Crinoid 
stems  appear  on  the  weathered  surface. 

3.  Fine-grained,  calcareous,  aphanitic  Schaisteiny  similar  to 
No.  6. 

Nos.  I,  a,  &  3  form  an  extensive  limestone-formation  on  tihe 
west  side  of  the  pass.  Here  a  zone  of  intrusive  quartz-porphyries 
occurs  along  the  N.W. — S.E.  strike. 

4.  ^  Fine-grained  *' diabase-psammite,"  with  small,  irregular  in- 
tercalations of  coal. 

5.  Coarse-grained  "  diabase-psammite,**  with  fragments  of  oligo- 
dase  and  diopside,  chlorite-flak^  and  serpentinoos  decomposition- 
products,  and  numerous  carbonized  remains  of  plants. 

6.  Brownish  grey,  fine-f^ained,  calcareous  Schalstein. 

7.  Dark  brown,  porphyritic,  calcareous  "  diabase-psammite," 
full  of  fragments  of  decomposed  augite,  andesine,  chlorite,  mag- 
netite, &c. 

•  Geol.  Beob.  mw.  Kur  6^  Araxes^  57. 

t  Ibid.  4S-S4* 


Ltrwer  Bajocian  313 

Nos.  5,  6,  &  7  are  very  greatly  developed  near  Kabdara  (4812  ft.). 

8.  Two  beds  (probably  flails)  of  ''fine-grained  doleritic  rocksy 
interbedded  in  dark  green,  aphanitic  diabase  schists." 

9.  Dark  green,  coarse-grain^ ''  diabase-psammite/'  with  round- 
ed fragments  of  oligodase  and  green,  decomposed  augite. 

la  (keenish,  fine-grainei  *' diabase-psammite,''  with  rounded 
fragments  of  a  '*^dark  green  ferric  mlicate,  and  a  greenish,  glassy 
felspar,  with  a  little  pyrope." 

II.  Greenish  grey  conglomerate,  with  fragments  of  greenish 
diallage  and  oligodase. 

Nos.  10  &  II  are  traversed  by  dikes  of  a  fine-grained,  reddish 
brown,  or  dark  greenish  grey  angite-porphyrite  (Abich's  "  diabase 
porphyry"),  which  also  forms  the  bed  of  the  Akera  river  (2643  feet). 

Abich,  on  continuing  his  ascent  of  the  range,  after 
traversing  Nos.  1-3,  observed  a  repetition  of  the  lower 
fragmental  beds,  perhaps  as  the  result  of  a  step-fault. 
Here  they  dip  to  N.E.,  and  soon  form  a  syncline. 
They  are  somewhat  disturbed,  and  extend  to  the 
summit  (6827  feet)  of  the  pass  over  this  part  of  the 
Kur-Araxes  watershed,  consisting  of  quartz-porphyry. 
This  is  succeeded  by  a  N.W.-S.E.  zone  of  serpentine 
and  gabbro  before  the  recurrence  of  the  series  of  the 
greenish  sandstones  and  tuffs.  At  the  base  of  the 
Uchdara  valley  they  overlie  an  "old,  greenish  grey, 
very  compadl,  massive  rock,  like  homstone,  with 
ix>rphyritic  quartz  and  felspar,  probably  of  Palaeozoic 
age.  Further  east,  in  the  Dashalti  valley,  Abich 
noticed  a  thick  *^  diabase-psammite,*'  on  the  same 
horizon  as  No.  5  of  the  foregoing  table.  The  series 
extends  still  further  east  to  form  the  base  of  the 
Shusha  plateau.  On  its  N.E.  side,  towards  the  Bazar 
valley,  Abich*  found  them  to  take  a  greater  share 
in  the  orography  than  the  Tithonian  limestones, 
e.g.,  they  form  the  group  of  flat-topped  moimtains 
(4009  feet)  above  Shushakent ;  a  little  further  north, 
in   the  broad,  open   valley  of  Dashkent,   they  are 

♦  Geo/.  Beob.  tw.  Kur  b*  Araxes^  78. 


314  Geology  of  ArtHema 

overlain  bv  calcareous  marls  and  sandstones,  which 
are  probably  of  Kimmeridgian  age  (Coraliian  of 
Abich,  vide  infra).  They  again  appear  at  the  base 
of  the  Khachin  gorge  (N.W.  of  Dashkent),  and 
between  Basarkent  and  Kandsassar  (Vank,  3719  feet) 
these  sandstones  are  similarly  overlain  h^  odcareous 
marls  and  sandstones,  dipping  30''  NJE.^  Abicfat 
found  them  also  in  the  N.W.  part  of  the  East 
Karabagh  Range,  above  TarkhankaLa,  in  the  Daly 
Dagh,  and  in  the  upper  Duchu  valley;  here  they 
evidently  form  the  N.W.  continuation  of  the  zone 
already  tabulated  on  pp.  31a,  313. 

Some  of  these  Lower  najodan  deposits  may  have 
been  lacustrine,  for  (according  to  Chaper*s  account  X) 
the  ore-deposits  of  this  age  of  the  Lialwar  Dagh 
in  Somketia  (N.W.  of  Lake  Gokcha)  must  have  been 
formed  in  a  calm  lake,  iiito  which  mineral  springs 
issued  in  abundance.  Lialwar  at  that  time  was  an 
a<5live  volcano.  He  considers  that  first  of  all  a  thick 
deposit  of  silica  and  silicates  was  formed,  certainly 
exceeding  30  metres.  Subsequently  new  springs  came 
into  operation — ^not  exclusively  siliceous  but  highly 
metalliferous"^— depositing  sulphides  of  copper  and 
iron  in  the  form  oi  lentides  or  ^*  mushrooms, '  capping 
the  ends  of  the  mineraUveins.  Towards  the  end  cS 
this  period  sulphide  of  zinc  was  deposited,  then 
sulphide  of  lead,  associated  with  barium  sulphide, 
the  latter  becoming  transformed  into  sulphate.  Fi- 
nally calcium  sulphide  appealed,  similarly  oxidizing 
into  the  sulphate,  and  associated  with  only  very  little 
silica  and  hardly  any  metals.  The  sulphides  occur  in 
perfect  crystals,  enclosed  in  the  siliceous  matter,  and 

*  GeoL  Beob.  tm.  Xwr  6*  An^ees,  X3S« 

t  INd.  itS,  1 19, 133. 

I  Bu/l  Sac.  GM.  JPhmee,  %6i.  jr^xn.  X03. 


Upper  Bajocian  315 

owing  to  the  impermeability  of  this  matrix  they  have 
imdergone  no  alteration.  The  gradual  deposition 
of  the  silica  and  of  the  metallic  sulphides  does  not 
seem  to  have  been  affedled  by  any  currents. 

On  the  other  hand«  according  to  Conrath,^  who 
was  the  resident  chemist  and  engineer  for.  some  time 
at  the  mines  of  Akhtala  (on  the  S.E.  slopes  of  the 
Lialwar  Dagh),  the  ores  have  arisen  merel]^  as  the 
result  of  the  contadl-a<ftion  which  intrusive  por- 
phyrites  have  exercised  on  the  fundamental  granite. 
The  Jurassic  series  ovei;lies  the  porphyrite  with  its 
associateji  copper-ores,  and  occurs  in  thick  beds, 
nearly  horizontal,  but  with  a  slight  dip  to  E.S.E. 
The  series  shows  the  following  downward  succession : 

I.    Uppie  Bajocum>— 

(i)  Dull  frey  or  grecDiah  sandstoneSi  nnfossiK- 
feroust  containing  a  considerable  amount  cf  triturated 
organic  material  together  with  eruptive  dements. 

(ii)  Arenaceous,  slaty  limestones  alternating  with 
sandstones.    The  numerous  fossils  (vide  infra)  lie  for 
the  most  part  in  a  bed  4  to  5  metres  thick, 
a.    Lower  Bajociak  : — 

f  i^    Conglomerates  of  decomposed  porphyrite. 
[ii)  Dull  green   psammites   or  argillaceous  sand- 
stones, containing  abundant  plant-remains  and  pieces 
of  coniferous  w<kk1,  together  with  enormous  masses 
of  indeterminable  oysters. 

The  highest  beds  have  been  overlain — according  to 
Redlicht — ^and  partly  altered  by  andesite,  in  steep 
cliffs  about  15  metres  thick,  forming  the  peaks  of 
Babeluchan  and  Karagaya  Dagh  (just  S.  of  Akhtala). 
Ih  fadl,  there  are  numerous,  dark  green  or  pink» 
contemporaneous  andesitlfc   lava-flows,   altering  the 

^  Quoted  by  &edlich,  Der  Jura  der  Umgihung  von  Alt-Acktaia^ 
Beitr.  Pal.  CEstr.  Ung.  ix.  55.  Since  Redlich's  work  corrects  many 
of  the  erroneous  determinations  and  conclusions  of  Abich»  Uhlig,^  and 
Chaper,  I  have  followed  his  account,  condensing  it  so  far  as  possible. 

t  Op,  cit. 


31 6  Geology  of  Armenia 

underlying  sediments. 

The  same  fossiliferous  zone  occurs  near  Shamlug^h, 
at  a  distance  of  5^  kilometres  from  Akhtala  as  the 
crow  flies.  The  whole  series  must  have  been  depK>sit* 
ed  near  a  coast-line. 

The  fauna  was  first  of  all  considered  Oxfordian 
in  age  by  Abich  ^ ;    Uhlig  t  subsequently  treated  i  t  as 
Callovian,  but  this  conclusion  was  mainly  based  on 
his   erroneous    determination    of    Waldheimia   pain, 
which  was  so  abundant  as  to  form  a  special,  more 
calcareous    bed.     Redlich,^    however,    re-examined 
Uhlig*s  Shamlugh  fossils,  and  found  that  ''  his  Wald^ 
heimia   pala  are   bad   specimens   and   can   now   be 
identified  as  W.  carinata^  provincialise  Mandelslohi^  and 
Meriani  " — all  belonging  to  the  Lower  Oolite.     Red- 
lich  §  has  also  shown  that  the  fossils  belong  ''  chiefly 
to    the    zones  of   Stepheoceras   Humphriesianum  and 
Parkinsonia  Parkinsont^  and  at  most  to  that  of  Wald- 
heimia  digona.    All  the  fossils  come  from  a  bed  only  a 
few    metres    thick.     It    is    immaterial    therefore   if. 
some  of   the  fossils  are  represented  in  Europe  on 
higher  or  lower  horizons."    Redlich,  however,  points 
out  that  some  of  the  species  belong  to  the  Cornbrash, 
and  that  a  few    are   even   Callovian   forms.    The 
following  analysis,  which  I  have  made  of  his  list 
of  fossils^  arranging  them  in   the  zones   in   which 
they  (or  the  nearest  allies  of  the  new  species)  occur 
in   Europe,    will,    I    think,    demonstrate    the    true 
Upper  Bajocian  age  of  this  comparatively  thin  bed. 
The  admixture  of  Bathonian  and  even  di  Callovian 

^  CM.  Btob.  tw.  Kur  &*  Araxes^  10,  59 ;   and  his  dd,  Fonch.  im 
dm  kauk.  IMntUm^  ii.  26. 

t  Dmkukr.  Wuner  Ak^d.  lix.  97. 

X  op.  eit.  79. 

f  lUd.  78. 


upper  Bajocian  317 

species  may — it  seems  to  me — be  explained  by  sup- 
posing that  these ,  species  might  have  originated  in 
SL  more  easterly  regioi^  and  that  in  spreading  west- 
'wards  they  would  reach  Armenia  at  an  earlier  period 
than  Central  Europe  and  England.  The  facies  of 
this  assemblage  of  fossils  is  entirely  Central  European 
in  character,  and  not  Alpine,  although  some  of 
the  Ammonites  have  somewhat  Alpine  affinities. 
In  this  case  the  question  is  one  of  climate,  for  the 
Caucasus  must  have  been  situated  close  to  the  border 
of  two  geographical  provinces. 

In  the  K)llowing  list  I  have  (on  this  occasion) 
placed  the  whole  series  in  upward  succession,  and 
when  a  species  has  a  greater  vertical  range  it  is 
placed  in  its  lowest  zone  : — 

1.  Zone  of  Ludwigia  Murchtsonce  :-^ 

Terebratula  perovalis^  Sow.,  T.  ventricosa^  Hartm. 
Lima  Roemerif  Brauns;,   Entolium  (Pecten)  de- 

missus  J  Phill.;    Camptone£ies  (P.)  of,  lens^  Sow.; 

Astarte  excavata^  Sow. 

2.  Zone  of  Stephe9ceras  Humphriesianum  : — 

Rhyftchonella  Wrightii^  Dav.,  var.  declivus,  Redlich 
mumerous),  R.  quadriplicata^  Zietcn  (numerousj, 
K.  n.  sp.  cf.  quadriplicatay  Ziet.,  R.  dyptevix, 
RcdKch;  Terebratula  omalogastyr,  Zieten,  T. 
Jamssiiy  BaUy ;  Waldheimia  Meriani,  Oppel ; 
Aulacothyris  carinata,  Lgjn.;  Zeilleria  Waltoni^ 

Davidson. 

Lima  eduda^  Whidbome,  L.  plana^  Redl.;  Pecten 
ambiguuSy  Goldf.;  Avicula  Muensteri^  Biauns, 
A.  akhtalensis^  RedL;  Modiolq  cuneata^  Sow., 
M.  caucasica^  Redlidi;  Pinna  mitis^  Phillips, 
P.  cuneataj  Bean.;  Trigonia  formosa^  Lye, 
T.  IcBvicostatOy  Redlicfa.;  Astarte  detrita^  Groldf., 


.*." 


..•'^-•v 


3i8  Geology  of  Armenia 

A.  d^essa.  Gt.;  Opu  similis^  Sow..;  Pkdadomya 
otmlum^  A(.,  P.  shanUughensis^  Sedlidi.,  P. 
asiatica^  RedKc^i.,  F.  asiatica  w.  incarinata^ 
KedUch.,  P.  pennata^  RedHch.,  P.  cf.  spatiosa^ 
Wludb.,  P.  crassa.  Ag.,  Mactromya  confuse- 
lamellosa^  RedL 

Pleurotomaria  Paktmon^  &OA.;  Natica  cf.  ad- 
ducta^  d'Orb. 

Perisphinctes  Af artitisi7  d'Oib. 

3.  Zone  of  Parkinwnia  Parkinsoni : — 
Rhynchonella  Stuifensis^  Oppd,   R.  cf.  subtetra- 

nedra^  Sow.,   R.  sitbobsoleta,  Dav.;    Terebratula 

flcbata^  Sow.  (numerous),  T.  cf.  decipiens,  Deth., 
'.  orientalis,  Redl.,  (of  Uie  T.  Fhillipsi  group) ; 
Waldheimia  anglica^  Opp.,  H^.  provindalis^  Sow. 
Pecten  vagans^  Sow.,  ilrca  intusplicata^  Terq.  & 
Joiudy,  il.  ^ctfincttlofcids,  Teni.  frjonrdy,  Cucullcea 
elongata^  Sow.,  C.  concinna^  PhiD.;  Pleuromya 
rhenana^  Schl^pe,  P.  Jurassic  Bronsn.;  ilstorte 
^tt/Za,  Romer. 
Lytoceras  polyhelictum^  Bdckh. 

4.  Bathonian  : — 

Tenbratula  maxillata^  Sow.;  T.  subntajaZIatoi 
Morr.  &  Lye.;  ilfiiocotAym  Bentleyi,  Davidson^ 
A.  Bentleyi  w.  losmuscula^  UhUg  (Klaus  beds 
of  Khod  Alagyr  in  the  Caucasus) ;  Zeilleria 
omithocephala,  Sow.,  Watdhdmia  Mandelslohi^ 
0pp.,  (fairly  frequent). 

Lima  Conrathi^  Redtich;  PeAen  pseudotextorius^ 
Redlidi,  P.  cf.  fibrosus^  Sow.,  P.  cf.  doth^atus^ 
Romert  P.  Ryphaeus,  dX)ib. 

Belemnites  of  the  canaliculati  group. 

5.  Callovian  : — 

Terebratula  cforso^licota,  Sum ;   AttlocotXryris  m- 


Upper  Bajocian  319 

pressa^  Sow.,  var.  gravida,  Redlich. 
Pecten  solidus,  Romer. 
Phylloceras  akhtalense,  Redlich  (of  the  P.  tatricum 

group) ;  Belemnites  of  the  hastati  group. 

Among  these  fossils  coUedled  by  Conrath  near 
Akhtala  and  Shamlugh  Redlich  also  noted  spp.  indet. 
of  HinniteSj  Perna^  Area,  Thraciay  Cyprina,  Nerinea, 
and  Perisphinctes.  A  thorough  investigation  of  this 
locality  by  a  palaeontologist  is  greatly  to  be  desired. 

In  the  light  of  Redlich's  re-examination  of  Abich's 
fossils  the  following  lists  of  Abich  and  Uhlig  are 
of  minor  importance.  However,  since  they  seem  to 
indicate  not  only  some  Callovian  but  even  Kim- 
meridgian  forms,  it  is  possible  that  Abich  may  have 
coUedled  his  fossils  from  a  higher  horizon  than 
the  evidently  Bajocian  bed  from  which  Conrath 
obtained  the  fossils  detailed  on  pp.  317,  318.  Abich's 
list  *  is  as  follows : — 

Rhynchonella  lacunosa,  Sow.  (Seq.J;    Terebratula 

bissuffarcinata,   Schloth.    (Seq.) ;    Aulacothyris 

impressay  Sow.  (Call.). 
Lima  figida^  Dcsh.  (Oxf .) ;    Pecten  soliduSy  Romer 

(Call),    P.  subtextoriuSy  Goldf.  (Kimm.) ;    Tri- 

gonia  clavellatay  Park.  (Call.). 
Peris^nites  polygyratus.  Rein.  (Seq.) ;    Aganides 

aganiticuSf  Schlot.  (Seq.). 

Most  of  these  determinations  were  probably  in- 

correcft,  for  Uhlig's  listf  appears  to   be   based  on 

the   same   specimens   from   Shamlugh — "mostly  ill- 
preserved  " — with  some  additions,  as  follows  : — 

Rhynchonella  lacunosa.  Sow.  (Seq.),   R.  moravica, 

•  G^l.  Beob.  zw.  Kur  ^  Araxes^  10,  59 ;   and  his  GeoL  Forsch.  in 
dm  kank,  LMndemy  ii.  26.  . 

t  Denkuhr.  Wiaur  Akad.  Kx.  97. 


320  Geology  of  Armenia 

UhBg  (Seq.),  R.  Wrightii,  Dav.  (Baj.),  Acantho- 

thyris  cf.  spinosa,  Schloth.  (Baj.);   TerAratvla 

cf.  dorsoplicata^  Suess  (Call.) ;  Waldheimia  pala. 

▼on  Bach   [an   erroneous   determination,  see 

p.  316].  ^ 

Lima  semicircularis,  Gf.  (Baj.),  L.  cf.  tenuistnata^ 

L.  cf.  rigida,  Desh.  (Oxf.) ;   Pe&ten  sp.  indet.; 

Hinnites  velatus^  Gf.  (Liassic);    Trigonia  so. 

indet.  (one  of  the  Bajocian  Costatce  group); 

Trichites  (numerous  fragments). 
Nerinea  (small,  weathered). 
Perisphinctes   cf.  funatus^  Oppd  (Callovian  of 

Cutch) ;  Belemnites  cf.  hastatus^  Blainv.,  Call). 

The  fossils  which  Chaper  *  coUeded  at  Akhtala  (cp. 
p.  316)  were  determined  by  Douvill^,  and  (with  the 
exception  of  a  probably  erroneous  determination) 
indicate  the  Upper  Bajocian  age  of  the  fauna,  with 
the  admixture  of  some  species  occurring  in  the 
Bathonian  of  Ceiitral  Europe.  Thus  they  help  to 
support  the  conclusions  stated  on  p.  316 : — 

Rhynchondla  Orbignyi^  0pp.  (Baj.),  R.  Morierei^ 
Dav.  (Bath.) ;  waldheimia  pala^  von  Buch  (pro- 
bably erroneous,  see  p.  316) ;  Zeilleria  tval- 
tom,  Dav.  (Baj.);  Aulacothyris  Bentleyi,  Dar. 
(Bath.). 

Lytoceras  tripartitum,  Katpail  (Baj.). 

Spp.  indet.  of  Pecten,  Plagiostoma^  Pholadomya^ 
Pinnigena. 

To  the  north-west,  in  the  region  of  the  Zirula 
dome  (Meschic  Mountains),  the  Upper  Bajocian  over- 
lies the  Lower  Bajocian  plant-sandstones.  It  is  well 
exposed  in  the  valleys  of  the  Kvirila  and  Cherimela, 

•  jBuU.  Sac  GM.  Frama,  sfr.  3,  xxL  107. 


BathoftSam  331 

and— according  to  Fournicr •^—consists  of  ''bluish 
black  limestones,  calcareous  saiidstones,  arkoses,  and 
red  marmoraceous  limestones";  in  these  red  lime- 
stones the  following  fossils  occur : — 

Acrochordocrinus  cf.  anuUtheiy  Quentt ;    Pent  a- 

crinus  cf.  scalarts^  Qucnst;  Isocrinus  cf.  hasalti- 

formis^  Gf. 
Spirt feprina  cf.  obtusa^  Opp.,  S.  cf.  rostratay  Opp., 

S.  cf.  alpifuiy  Opp.;  Rhynchonella  quinqueplicata^ 

Ziet,  R.  cf.  variabilis^  Schloth.;   lerAratula  cf. 

punctata^  Sow.,    T.  cf.  plicatilis^  Sow.»  (Baj.), 

Aulacothyris  resupinata^  Sow.;  Zeilleria  cf.  sub- 

numisnuUiSf  Dav. 
Lima  proboscidea^  Sow.  (Baj.) ;  Entolium  (Pecten) 

demissus^  Phill.  (Bath.);    Trigonia  costata^  Gf. 

(Baj.),  T.  Meriani^  Ag.  (Bai.). 
Stepheoceras  Bayleyi^  Opp.   (Baj.);     Phylloceras 

cf.  Zetes^  d'Ofb.,   P.cf.  ultramontanum,  Zitt.,   P. 

cf.  mimatense^  d'Orb.,  P.  imeritianum^  Ncum.; 
Amaltheus ;  Mgoceras ;  Nautilus  cf.  truncatus^ 
Sow. 

It  is  noticeable  that  nearly  all  these  fossils  are 
Liassic  forms  with  the  exception  of  those  whose  age  is 
indicated  in  brackets.  The  persistence  of  Liassic 
types  in  the  Bajocian  of  the  South  Caucasus  has 
already  been  commented  upon  (p.  307),  and  Fournier 
ri.c]  states  positively  that  there  can  be  no  confusion 
in  the  age  of  the  strata  owing  to  faulting  or  over- 
folding. 

The  Bathonian  is  represented  in  the  same^  area 
by  similar  red  marmoraceous  limestones,  containing 
—according  to  Fournier  f — the  following  fossils  : — 

t  Md.  III. 


3^2  Geology  of  Armenia 

A^iocrinus  itnpressus^  Qoenst  (Oxf.). 

Rhynchoiulla  decorata^  Qvenst 

Phokuhmya  Murchisoni,  Sow.,  P.  parcicostata,  Ag., 

P.  donadformis^  Gf. 
Pkylloceras  tripartitum,  Set.,  P.  medit^rraneum, 

Neam.  (Oxf.) 

Here  some  Oxfordian  forms  occur  intermingled 
with  the  BathoQian  species  (cp.  pp.  316,  317).  No 
other  instance  of  Bathonian  beds  has  as  yet  been 
recorded  from  the  Armenian  area  or  its  borders. 

ra.    vmsL  JUIAS8IC. 

The  Callovian  (Macrocephalus  and  Anceps  zones)  is 
present  at  several  points  along  the  Eastern  Border- 
Ranges,  from  the  Zirula  dome  into  Azerbaijan  near 
Lake  Urmi.  It  seems  probable  that  the  sea  became 
deeper  in  this  diredUon,  towards  Persia.  On  the  east 
coast  of  Lake  Urmi  thinly  bedded,  marly  limestones 
and  reddish  sandstones  of  this  age  were  found  by 
Pohlig*  and  by  Rodler.j  They  form  a  low  range 
with  N.E.-S.W.  strike,  frequently  concealed  by  the 
lavas  and  tuffs  of  Sahend  or  by  the  Pliocene  marls  of 
Maragha.  In  places,  e.g.,  at  Gushaish,  they  are 
overlain  by  ^feocomia^  beds.  The  fossils  consist 
mainly  of  Ammonites,  and  have  been  described  by 
WeithoferJ  and  by  von  dem  Bome.§  Except  where 
otherwise  specified  they  were  coUedted  at  Gushaish  in 
the  Murdi  valley  on  the  southern  slopes  of  Sahend. 

*  Vkrh.  k,'k.  g€9L  Rekhsamt  1884,  981.  Pohlig,  however,  was  in 
error  in  considering  the  beds  to  represent  only  Lower  Cretaceous. 

t  Sii%umgs^.  k.'k.  AIM.  Wist.,  Mathem.'nat9trw.  Chsu^  xcviL 
904,  and  xcviii.  s8. 

X  lUd*  xcviiL  7S(. 

C  JDtrfitraam  OO^/wim  Urmmtm.    Halle,  1891. 


C<Movh»  593 


A  few  specimens  were  found  At  Alrtahu-dereh  (about 
I  mile  south  of  Gugan  or  Goigan),  and  at  Khanaga 
(Canequa,  Chanaia)  and  Scurgan  on  the  east  coast 
of  Lake  Urmi. 

Weithofer*  was  of  opiniod  that  the  fauna  marked 
a  Kimmeridgian  age,  but  the  fc^;ls  were  very  badly 
preserved  and  crushed,  and  may  perhaps  nrad  some 
revision.    His  list  runs  as  follows : — 

Goniamya  Rodleri^  Weithofer,  (aff.  G.  trapeztna^ 
Buvignier,  Oxfordian  of  N.E.  France),  a  cast. 

Perisphinctes  Lothario  Opp.,  P-  cf.  polyplocus^ 
Reinecke;  Belemnites  persicus,  Wdthofer,  {aff. 
B.  semisulcatus^  Favre,  from  the  Kimmeridgian 
of  the  Jura).  The  Cephalopods  occurred 
also  at  Aktahu-dereh* 

Von  dem  Home's  later  work  on  more  complete 
material  indicates  the  Callovian  age  of  these  deposits. 
He  identified  and  determined  the  following  forms : — 

Serpula  on  the  casts  of  Perisphinctes. 

Spinigera  n.  sp. 

Ludwigia  (He£iicocerqs)  lunula,  Ziet.  (4),  L.  node* 
sa^  Quenst,  L.  krakoviensis^  Neum.  (2  casts), 
L.  gigas^  Quenst.  (2  bad  ex.;  possibly  a  doubt- 
ful determination,  since  this  is  a  Tithonian 
species),  L.  cf.  punctata^  suM,  (a  very'  bad 
ex.);  Stepheoceras  stenostoma^  (Khanaga); 
Macrocephalite^  sp.  indet.  (a  fragment  from 
Khanaga) ;  Perisphinctes  curvicosta^  Oppel, 
P.  panneaticus  Ndtl.,  P.  Cyrus ^  von  dem  Borne, 
P.  Xerxes^  yon  dem  Borne,  P.  poculum^  Leckenby, 
P.  balinensis^  Neum.;  Belemnites  cf.  calUnnensiSy 
Opp.  (2),  B.  persicuSy  Weithofer  (very  numerous 
fragments  from  Aktahu-dereh). 

♦  cy.  rif.  770, 


324  Geology  of  Armenia 

These  fossils  of  Gushaish  point  to  the  presence  of 
the  ReifUckia  anceps  zone  (Upper  Callovian),  while  the 
Stepheoceratidce  of  Khanaga  seem  to  indicate  the 
lower  zone  of  Macrocephalites  macrocepkalus.  Within 
recent  years  Giinther*  brought  back  a  fragment  of 
Perisphinctes  curvicosta^  Oppel,  in  reddish  limestone^ 
from  an  unnamed  locality  near  Maragha — ^possibly 
from  Gushaish.  These  beds  are  not  improbably 
continued  on  the  west  side  of  Lake  Urmi,  opposite 
to  the  Shahi  peninsula,  for  Wagner  t  states  that 
Jurassic  beds  have  *'a  ratbc^r  extensive  development 
here,  rising  up  along  the  shore;  they  are  rich  in 
fossils  such  as  Belemnites^  Fecten^  and  Terebratula.'* 

Nearly  the  same  horizon  has  been  established  by 
the  fossils  which  were  coUedted  by  AbichJ  near 
Dashkessan,  in  the  Gokcha  Moimtains,  N.W.  of 
Elisabetpol;  both  faunas  have  the  species  Ludwigia 
lunula^  Ziet.9  ^-  punctata^  stahl,  and  Belemnites  callovien- 
sts^  Opp.,  in  common.  The  exa(%  locality  is  at  Kabag* 
tappa,  in  the  upper  valley  of  the  Shamkhor  river,  a 
tributary  of  the  Koshkar  river.  Abich  identified  the 
following  species,  which  he  found  "  in  the  uppermost 
beds  of  a  very  thick  formation  of  clastic  diabasic  and 
dioritic  rocks"  [similar  to  No.  5,  p.  312].  The  adtual 
matrix  of  the  fossils  was  a  tuTaceous,  much  decom- 
posed, dark  green  '' diabase-psamnlite  with  Bohn- 
erzenJ**    Abich's  list  §  is  as  follows : — 

Phylloceras  tatricum^  Puich.;  Rhacophyllites  torti- 
sulcatus,  d'Orb.;  Ludwigia  (He^Hcoceras)  hectica^ 
Hartm.;  Ochetoceras  canaliculatumt  Miinst.  (Oxf.); 

^  Jcium.  Linn.  Sac.  xxvii.  418.    Identified  by  Kr.  G.C.  Crick. 

t  Jiiise  nach  PenUn^  ii.  90. 

X  Gtol.  Beob.  WW.  Kur  6*  Araxts^  58. 

f  Here,  as  elsewhere,  I  have  modernised  the  generic  names,  so  as 
10  secure  uniformity. 


Catlovian  325 

Haploceras  Erato,  d'Orb.;  Perisphin£tes  Orianj 
0pp.,  P.  sulci ferus,  Opp.;  Peltoceras  annulare^ 
Remecke,  P.  Athleta,  Phill.;  Belemnites  callo- 
viensiSf  Oppd. 

Abich  ako  noticed   here  indeterminable  casts  of 
bivalves    and    gasteropods.    Subsequently    Neumayr 
and  Uhlig*  worked  on  fuller  material  from  Kabag- 
tappa,  and  identified  the  following  fossils : — 
Lima. 

Pleurotomaria  (one  of  the  omata  group). 
Phylloceras  flahellatum,  Neum.,  (22  ex.),  P.  medi- 
terraneum,  Neum.  (Oxf.;  6  ex.),  P.  Kunthij  Neum., 
Haploceras  Erato,  d'Orb.  (Oxi.) ;  (Ecotraustes 
conjungens,  Mayer;  Ludwigia  (Hedticoceras) 
lunula,  Ziet,  L.  punctata,  stahl ;  Quenstedto- 
ceras  cf.Goliathus,  d'Orb.;  Perisphinctes  promis- 
ctttts,  Buk.  (Oxf.),  P.  c/.  Wartcc.  But  (Oxf.); 
Feltoceras  Athleta,  Phin.;  Belemnites. 

The  majority  of  these  forms  are  Upper  Callovian 
species  of  Central  and  Western  Europe.  The  early 
appearance  of  Oxfordian  types  in  these  Upper  Cal- 
lovian beds  may  perhaps  be  accounted  for  in  the 
same  way  as  for  the  admixture  of  Bathonian  and 
Callovian  forms  in  the  Bajocian  of  Akhtala  (pp.  316, 
317,  and  see  also  p.  322). 

The  Bathonian  beds  of  the  Zirula  dome  (p.  322)  are 
overlain — according  to  Fournieft — ^by  Callovian  with 
the  following  fossils  : — 

Rhynchonella  lacunoso-decorata^  Quenst 

Pecten  fibrosus.  Sow.  (Bath.) ;  Panopea  cf.  elonga- 

ta,  Ag. 
Phylloceras  tatricum^  Pusch.,   P.  mediterraneum, 
Neum.  (Oxf.),  P.  Hommairei,  d'Orb. 

*  Denksehr.  Witmr  Akad.  lix.  96. 

t  Description ghL  du  Caucau  anirtUy  iia. 


3i6  CMogy  of  Armenia 

The  OxFORDULN  (as  restricted  to  the  zones  of  Aspido- 
uras  perarmatum  aad  Peltoceras  transversarium)  has 
not  yet  been  recorded  from  the  Armenian  area,  but 
on  its  northern  borders  this  formation  is  present 
overlying  the  Callovian  of  the  Zirula  dome  (p.  325). 

Here — ^according  to  Fournier* —  it  is  composed  of 
"marly,  grey  or  red  limestones  and  micaceous,  cal- 
careous sandstones/'  containing  the  following  fossils : 

Eugeniacrinus  cf.  Hoferi^  Gf.;  Apiocrinus. 

Rhynchonella  strioplicatay  Quenst.»  R.  cf.  qutnque- 
plicata^  Ziet  (Lias),  R.  multiplicata^  Quenst, 
R.  cf.  variabilis^  Schloth.  (Lias).  R.  sparsico$ta^ 
Quenst.  (Kimm.),  R.  dichotoma^  Quenst  (Seq.), 
R.  lacuKOso-decorata^  Quenst.;  Terebratula  bis- 
suffarcinata,  Schloth.  (Seq.). 

Pecten  cf.  oequatus,  Quenst  (Kimm.) ;  Pema 
mytiioides^  Lam.;  Gervillia  aviculoideSf  Sow. 

Phylloceras  tatricum^  Pusch.  (Call.) ;  Belemnites 
hastatus^  Blainv.  (Call.). 

The  Sequanian  or  C^rallian  (as  restriAed  to  the 
zone  of  Aspidoceras  bimaMmatum)  is  likewise  unrepre- 
sented in  Armenia,  but  ap^ars  to  be  indicated  in  the 
Zirula  dome  overlying  the  Oxfordian  beds.  Here — 
according  to  Fournier  j^'Lc.^— "  pink,  compact  lime- 
stones "  occur  with  the  loUowing  fossils,  which  seem 
to  indicate  the  Sequanian  (Corallian)  horizon  rather 
than  the  Upper  Oxfordian  (Fournier) : — 

Hemicidaris  crenularis^  Lam.  (Seq.j. 
Rhynchonella  lacunosa^  Schloth.  (Seq.) 
Posidonomya  ornata^  Quenst  (Call.)  ;  Astarte. 
Phylloceras ;  Amoeboc^ras  altemans^  y.  Buch  (Kim.). 
Volcanic  adlion  became  strdngly  marked   during 
this  period,  and  Fournier  f  states  ^lat  these  limestones 

*  Disai/Han  gioi.  du  Cattcasi  wUtm/f  1x2. 
t  Md.  113. 


lA^ 


Oxfordian;  Sequanian;  Kimmeridgian       yx^ 

are  overlain  by  ^'  schistose  clays,  psammites,  red  gypsi- 
ferous  clays,  arenaceous  clays  and  conglomerates  with 
breccias  and  tu&  in  the  proximity  of  diabases  and 
melaphyres.  A  great  part  of  these  volcanic  rocks 
were  contemporaneous  with  this  period,  for  whilst 
they  have  altered  the  Lias  and  Lower  Oolite  they 
frequently  occur  as  pebbles  in  these  Upper  Jurassic 
psammites  and  clays/'  The  tuffs  at  Sharopan  con- 
tain AmcetiQceYQ&  altemanSy  v.  Buch,  and  Phylloceras. 

In  this  region,  and  indeed  over  the  whole  of  the 
southern  slope^of  the  Central  Caucasus,  there  is  a  gap 
between  these  Sec^uanian  beds  and  the  Lower  Neo- 
comian  (Valang^nian),  which  rests  discordantly  on 
their  upturned  edges.  On  the  northern  slope  and 
in  Daghestan  thev  are  overlain  by  Tithonian  strata. 
On  the  northern  slope,  however,  and  in  Armenia  some 
Kimmeridgian  is  present  and  only  the  Lower  Ti- 
thonian is  wanting,  but  the  discordance  is  still  great. 
The  volcanic  disturbances  were  apparently  the  pre- 
lude of  the  great  ante-Tithonian  or  Kimmeridgian 
folding  (p.  g),  by  means  of  which  the  Caucasus  and 
a  great  part  of  Armenia  became  raised  above  the  sea. 
The  folds,  both  in  the  Zirula  dome  and  in  the  Central 
Caucasus,  were  observed  by  ^Fournier  ^  to  be  bent 
over  to  the  north. 

The  Kimmeridgian  of  Armenia  shows  the  sponge 
and  coral  facies,  with  forms  for  the  most  part 
identical  with  those  of  the  Nattheim  limestone.  It  is 
developed  in  two  distridis  of  the  Karabagh  region  : — 

(i)  the  West  Karabagh,  between  the  basins  of 
the  Okhchi  and  Bazar  rivers;  here  sponges  and 
echinoderms  predominate. 

(ii)  the   East   Karabagh^  between   the  Terter 

*  DiscrifHoH  ghL  dm  Caucase  centtfd^  33. 


328  Gfhgy  of  Armenia 

and  Araxes  rivei:. ;  here  corals  predominate. 

At  the  bottom  of  the  Okhchi  valley  Abich*  found 
'^a  reddish,  compadl  and  marmoraceous  limestone, 
c*ontaining  Scyphioe^  Cnemtdioe^  and  remains  of  Echino* 
derms,  especially  spines  of  Cidaris.''  Tliis  red  lime- 
stone underlies  a  thick  series  of  sandstone  (derived 
from  volcanic  material),  and  these  are  in  turn  over- 
lain unconformably  by  Turonian  limestone. 

The  sponge  facies  is  well  developed  a  few  miles 
to  the  N.E.,  near  the  mines  of  Gavart.  Here  a  sand- 
stone occurs,  in  which  Tsulukidsef  found  a  number 
of  fossils  (now  in  the  Tiflis  Museum),  as  follows : — 

Cylindrophyma  millepoYOua^   Goldf.;     MeUmdla 

radiata^  Quenst 
Tremadidiyon  reticulatum^   Goldf.;     Craticularia 

clathrata^  Gf.,  C.  cancellata^  Gf.;    Verrucoc(Blia 

verrucosay  Goldf. 
Eudea  perforata^  Quenst;    Eusiphonella  Bronni^ 

Miinst. 

RhcAdophyllia ;  Calamophyllia. 
Ceriocrinus  [?] ;  Apiocrinus  cf.  annulatus^  Quenst.; 
Millericrinus ;   Pentacrinus  cf.  asiraliSy  Quenst. 

In  the  neighbouring  valley  of  the  Semi-chai  he  also 
found  the  following : — 

Cribrocoelia  calopara^  Gf. 
Stellhpongia  glomerata^  Quenst 
Serpuld  socialisy  Gf. 

In  the  valley  of  the  Selin-chai  Tsulukidse  foimd 
the  Asteroid  Sphaerites  scutatus^  Gf.,  and  near  Shaibli, 
oil  the  banks  of  the  Chaundur  river  (the  lower  Okh- 
chi) he  coUedled  spines  of  Cidaris  curvata,  Gf.,  and 
of  C.  caronataf  Gf. 

*  Vergleichendi  geol.  Grundsugi  Cfc.  9a. 
t  (^/.  QUal  Tiflis  Museum^  izo. 


Kimmeridgian 

The  same  Cidaris  limestone  was  observed  by  Abich  ^ 
a  little  further  north,  at  the  bottom  of  the  deep 
Bazar  river,  just  below  :the  monastery  of  Tatev 
jfTatief),  where  it  is  overlain  bv  a  thick  lava*stream 
Irom  the  volcanic  plateau  of  the  Central  Karabaglu 
The  marmoraceous  rock  crops  out  at  the  ccmfluence  of 
the  Sadana  and  Bazar  rivers;  it  is  rich  in  fossils 
in  white  calcspar,  as  follows : — 

^'Clavate   Scyphice   and   cylindrically   rounded 
Cnemidice^'  [cp.  the  sponges  on  p.  328]. 
Apiocrinus ;   Pentacrinus  cingulatus^  Munst. 
Cidaris  coronataj  Gf.;  C.  Blumenbachii^  Miinst. 
Belemnites. 

In  another  part  of  the  valley  Abich  found  a  perfect 
specimen  of  Rhynchonella  lacunosay  SchIoth.»  in  the 
compadl  limestone. 

In  the  East  Karabagh  the  Kimmeridg^an  coral- 
limestone  occurs  on  the  Pyrjamal  plateau,  to  N.E 
of  Shusha.  It  forms  a  rocky  ridge,  with  dazzling 
white  cliffs,  rising  150  feet  above  a  belt  of  doleritic 
rocks  to  the  south-west.  Here  Abich  t  found  a 
number  of  much  altered  corals  in  the  coarsely  oolitic, 
marmoraceous,  white  limestone,  occasionally  reddish 
or  light  yellow.  Here  again  the  fossils  are  mostly 
Nattheim  species,^:  as  follows : — 

Latimaeandra  lotharinga^  Mich.;  Sty  Una  limbataj 
Goldf.,  S.  micrommata^  Quenst.,  S.  castellutnj 
Mich.;  Cyathophara  cavernosa^  Quenst. 

Ai^inaraea  granulata^  Munst. 

Heteropora  tenuissima^  EtaUon ;  Reptomulticava 
(H.)  capilliformiSy  Mich. 

*   VergUichemde  geoL  Grundtugt^.  9a. 
t  GioL  Buob.  tw.  Kur  6*  Araxes^  ^S-iM* 

X  Here,  as  elsewhere,  I  have  mo&Mvuni.  Ite  generic  names,  so  as 
to  secure  unifonnity. 


330  Get^gy  of  Armenia 

Apiocrinus  rosc^ceus^  Schloth. 
Pinnigena  (Trichites). 

The  same  limestone  was  found  by  Valentin  *  at  Kiz 
Kalassi  (a  few  miles  S.E.  of  Pyrjamal),  on  the  border 
of  the  great  Kur  steppe,  and  he  traced  it  as  far  as 
Dagras.    The  reddish  limestone  occurs  in  thick  beds, 
dipping  N.E.,    with  a  steep  escarpment  stretching 
for  6^  miles  to  N.W.,  overlooking  a  belt  of   dark 
red  porphyrite  (dolerite  of  Abich),  which  lies  between 
tliis  limestone  and  the  rugged  Senonian  limestones  to 
the  S.W.    Most  of  the  beds  are  granular,  consisting 
almost  entirely  of  Echinoderm  remains,  whilst  others 
are  white,  fine-grained,  and  saccharoidal.    Red  horn- 
stone    [possibly    indicating   sponge   remains]   occurs 
throughout  the  series  in  slabs  about  4  inches  thick. 
The  fossils  found  by  Valentin  were  very  ill  preser\'ed 
and  consist  of  the  following  forms  : — 

Clavate  spines  of  sea-urchins. 
Rhynchonella ;  Terebratulina ;  Terebratula, 
Exogyra;  Pecten;  Nucula. 

A  hornstone- limestone,  ''similar  to  that  of  Natt- 
heim,"  was  found  by  Abich  t  to  compose  the  summit 
(4707  feet)  of  the  dividing-ridge  between  the  Khachin 
and  Khojali  rivers  (north  of  Shusha).  He  coUedled 
the  following  Nattheim  fossils : — 

Cnemidiastrum  [AchilhumJ  tuberosum^  Zitt 
IsastYoea  confluensj  Gf.;   Stylina  cf.  tubulosa^  Gf.» 

S.  (Pachygira)  cf.  Delude  Dcfr.;    Astrocomia 

cf.  pentagonaliSj  Miinst. 

This  limestone — according  to  Abich  t — overlies  a 
conglomerate,   containing  fragments  01  pale   green 

*  Bericki  Serukenb,  Natutf.  Gts.  Frankftirt  a/M.  1891,  197. 
t  GeoL  JB§0k.  WW,  Kur  6*  Araxes^  109. 
X  lUi.  193. 


Kimtneridgian  331 

dioritic  and  diabasic  rocks,  and  of  a  limestone  with 
£chinoderm  remains.  This  conglomerate  extends 
up  the  Khachin  valley  to  the  East  Karabagh  range  at 
Alagaya  and  Chilg}*as,  and  eastwards  to  the  Podkasa- 
perk  monastery  (4704  feet).  It  appears  to  indicate  a 
marked  transgression  in  the  Middle  Kimmeridgian, 
^vhilst  the  Caucasus  and  northern  Armenia  were 
being  raised  above  the  sea  (p.  327). 

A  similar  conglomerate  occurs  at  the  base  of  the 
limestones  of  the  Shusha  plateau.  Here  Abich* 
found  a  pale  green,  compadt,  fine-grained  sandstone, 
with  fragments  of  opaque  felspar  crystals  and  chlorite 
grains ;  it  forms  part  of  a  "  variegated  clastic  form- 
ation, which  transgresses  over  a  region  of  the  plateau 
where  the  limestones  have  been  much  folded  and 
faulted  out  of  their  original  position." 

In  the  northern  part  of  the  Shusha  plateau,  below 
the  fortress,  the  limestone — according  to  Abich  t — "  is 
in  part  compadl  and  coarsely  crystalline,  composed  of 
shell  and  coral  fragments,  or  else  coarsely  oolitic  and 
looser  in  texture,  with  grains  of  green  earth."  Among 
the  sparing  fossils  from  these  beds,  which  occur 
rather  low  down  in  the  series,  Abich  J  determined  the 
following  forms,  some  of  which  are  badly  preserved 
aifd  need  revision  : — 

Rhynchonella  pinguiSy  Rom.,  {R.  corallina^  Leym.); 
Terebratula  peroualis^  Romer,  [a  doubtful  de- 
termination; this  is  a  variety  of  the  Neo- 
comian  T.  bipltcata,  Defr.,  and  is  not  to  be 
confused  with  the  Bajocian  T.  perovalis^  Sow.]. 

Exogyra ;  Pedlen  aff.  Nillsoni^  d'Orb.,  [a  doubt- 
ful  determination,   for  this  is   a   Senonian 

*  Geo/,  Bead.  nr.  JCur  6*  Araxes,  60. 
t  Jh'd.  70. 
t  Ibid.  12-1$ 


332  Geology  of  Armenia 

fossil] ;  Diceras  (4-5  cm.V 
Ptygtnatis  (Nerinea)  Manaelslohij  Bronn. 

According  to  Valentin*^  "the  beds  of  pale  lime- 
stone alternate  conformably  with  green  or  black  tuffs, 
sheets  of  lava,  and  schistose  sandstones,  which  here 
and  there  contain  carbonized  plant-remains.**"  The 
paucity  of  fossils  is  not  improbably  due  to  occasional 
submarine  eruptions  during  the  deposition  of  the 
limestone.  In  some  cases,  however,  the  igneous  rocks 
were  not  contemporaneous,  for  Valentin  {l.c.)  observed 
on  the  new  post  road  an  intrusive  sill,  which  had 
reddened  and  disturbed  the  limestones  both  above 
and  below. 

The  highest  strata  of  the  Shusha  plateau  are  light, 
fine-grained,  crystalline  limestone,  occasionally  dolo- 
mitic,  rising  to  5068  feet  on  the  S*W.  edge  of  the 
plateau;  here  Abichf  found  the  following  fossils, 
which,  however,  were  badly  preserved : — 

Montlivaultia ;  Trochosmilia ;    Calamopkyllia  cf. 

dichotoma^  Goldf. 
Apiocrinus  cf.  elongatus^  Miller. 

Just  below  this  Kimmeridgian  crinoidal  limestone 
Abichj;  found  the  following  fossils  in  a  compadl, 
smoke- grey  limestone  in  the  upper  part  of  the  town : 

Rhynchonella  Ictcunosa^  Schloth.;  Terebratula  cf. 
selta^  Rom.  [?  =  r.  subsella];  Waldkeimia  hip- 
popoides,  Abich*  (W.  hippopus^  d'Orb.  is  Valan- 
gmianj. 

Valentin  §  in  1890  re-examined  Abich*s  fossiliferous 
localities  in  the  N.  and  S.W.  parts  of  the  Shusha 

*  Beriehi  Sefukind.  Naiutf.  Get.  Fnmkfuit  a/M.  1891,  199. 
t  GtoL  Bioh.  Mw.  Kur  &*  Araxes^  63-65. 
)  lUd.  67-69. 
{  Op.  at.  90U 


I 


Kimmeridgian ;  Tithonian  333 

plateau,  and  found  Rhynchonella  inconstans.  Sow.,  in 
"both,  cases.  Hence  the  age  of  the  whole  series  may  be 
regarded  as  Kimmeridgian  (Tithonian  according  to 
Valentin).  The  Shusha  limestone  was  traced  by 
Valentin  further  east  to  Signakh,  and  to  the  S.E. 
as  far  as  Tug. 

The  same  horizon  of  the  Kimmeridgian  seems  to 
be  present  in  the  West  Karabagh,  between  Sevakiar 
(Sevaker)  and  Seiva,  on  the  Vank  ridge  south  of 
the  Chaisimi  river,  a  right  tributary  of  the  Bergushet. 
Here  Tsulukidse*  found  a  limestone  containing  the 
following  fossils : — 

Rhynchonella  inconstans^  Sow.;    Terebratula  sub- 
sellGj  Leym.;  Zeilleria  pentagonalis^  Br. 

The  Tithonian  seems  to  occur  in  the  West  Kara- 
bagh, only  a  few  miles  west  of  the  last  locality.  On 
the  same  ridge,  in  the  Kiaki  Dagh,  just  above 
Khotanan,  Tsulukidse  |  found  the  following  fossils  in 
a  coral-limestone : — 

Sphcerococcttes  granulatus^  Stemb. 

Microphyllia;    Thecosmtlia;    Styltna  decemradi- 

ata^  Quenst 
Diceras   (Heterodiceras)   Escheriy   dc  Lorioi;    D. 

suprajurensiSj  Thurman. 
Nerinea  Voltzitj  d'Orb.,  N.  Castor^  d'Orb. 

This  limestone-ridge  is  continued  a  few  miles  to  the 
W.N.W.  into  the  Trapassar  Dagh  (S.  of  Tatev). 
Here  AbichJ  found  the  following  silicified  fossils, 
overlain  by  beds  of  Urgonian  age  : — 

*  Geoi.  Catal.  Tiflis  Museum^  1x0. 

t  Ibid.  1x0,  xx6. 

X  GtoL  Beob,  tw.  Kur  &*  Araxes^  X15 ;    GeoL  Fanch.  in  dm  kaiA. 
r,  ii.  a6;  and  his  VergUkhendc  gtoL  Grundziigs^c.  X23. 


334  Geology  of  Armenia 

Calamophyllia  pseudostylina^  M-E.,    C.  Moreausi- 

ana,  Mich. 
Diceras  ( Hoterodiceras)  Escheri^   dt    Loriol,     D. 

Luetic  Favre,  D.  arietinum^  ism. 
Nertnea  stratnbergensis^  Petm»    N.   pupaformis^ 

Abich,  iV.  mutabilis,  Abich,  N.  berpisheti,  Abidi ; 

Ptygmatis  (N.)  Bruntrutanc^  Yhamuui ;  Itieria 

(N.)  pseudO'Moreanc^  Abich. 

This  facies  is  very  similar  to  that  of  the  Tithoiiian 
Stramberg  limestone;  its  nearest  equivalent  is  as 
yet  known  only  on  the  northern  slope  of  the  Central 
Caucasus,  where  it  passes  upwards  insensibly  into  the 
Neocomian. 


^  1 


CHAPTER  XIX 

THE  LOWER  CRETACEOUS  OP  ARMENIA 

I.     NXOGOIOAM. 

The  Lower  Neocomian  or  Valbnginian  has  not  yet 
been  definitely  recorded  from  the  Armenian  area, 
although  it  will  probably  be  found  to  occur  in 
the  Karabagh  region.  Abich^  indeed  states  that 
^^  Neocomian  strata  are  probably  present  on  the 
Trapassar  Dagh,"  overlying  Tithonian  limestone  (p, 
333),  but  he  does  not  mention  any  fossils  of  an  earlier 
age  than  Urgonian  hide  infra). 

The  lower  beds  (Berrias  series)  of  the  Valanginian 
are,  however,  well  developed  on  the  northern  borders 
of  Armenia,  \n  the  Zirula  dome  (Meschic  Mountains). 
Here  in  the  S. W.  part  of  the  synclinal  trough  of 
Lasheh  (S,E.  of  the  Kvirila)  it  consists — ^according 
to  Foumiert — of  marls  and  marly  limestones  with 
the  following  fossils : — 

Terebratula  pseudojurensis^  Leym.;  Wcddheimia 
Moutaniana^  d'Orb.;  Pygi^pe  diphyoideSj  d'Ocb. 

Chama} 

Nerinea  Marcausana^  <POrb.;  Strambus  Sautierij 
Coq.  (Natica  letnatha%  Vkt.). 

*  Geo/.  Bead.  tw.  Kwr  6*  Afoxa^  115. 
t  HfuripHan  gM.  dM  CoMcau  emtral,  199. 


i 


336  Geology  of  Armenia 

The  Caucasus  must  have  existed  at  that  time  as 
a  nairow  isthmus,  for  the  Lower  Neocomian  deposits 
on  the  northern  slope  show  quite  a  different  fades 
from  those  of  the  southern  slope  and  the  Mescfaic 
Mountains.  Whilst  on  the  northern  slope  they  pre- 
sent the  character  of  coral-reefs,  in  complete  con- 
tinuity with  similar  Tithonian  limestones,  they  con- 
sist on  the  southern  slope  of  brachiopod  marls  and 
limestones,  lying  in  marked  discordance  to  the 
Jurassic  strata.  Near  Goreshi,  in  the  Zirula  dome^ 
they  rest  indeed  on  granite,  the  contact  being  marked 
by  breccia  and  conglomerate. 

The  Middle  Nbocomian  or  Hauterivian  has  a  wide 
distribution  on  the  northern,  eastern,  and  north- 
western borders  of  Armenia. 

In  the  north,  it  is  well  developed  in  the  Zirula 
dome  in  the  form  of  marly  sandstones  and  limestones, 
containing — ^according  to  Foumier* — the  following 
fossils,  some  of  which,  however,  are  clearly  Val- 
anginidn : — 

Rkynckonella  cf.  depressa^  d'Orb.,  (Val.),  R. 
multiformis;  ^Orh.;  Terebratula  prcelonga^  Sow., 
T.  Dutempleana^  d'Orb.;  Waldheimia  Mou- 
toniana^  d'Orb.  (Val.),  W.  tamarindus.  Sow. 

Exogyra  Couloni,  d'Orb.;  Alediryonia  rectangularis^ 
Sow.;  Neithea  ataoa^  JUS^m.\  Pema  MtUletif  DtiAi.] 
Astarte  neocomiensiSf  d'Orb.,  A.  cf.  allaudiensiSf 
Matth.;  Cardium  cf.  galloprovinciale^  Matth., 
C.  cf.  peregrinum^  d'Orb.;  Arctica  (Cyprina) 
cordtformis,  d'Orb. 

Nautilus  pseudo-elegans^  d'Orb.,  U.  cf.  Necker- 
ianus^  Pict.;  Olcostephanus  (Astieria)  astierianus, 
d'Orb.;  Hoplites  cf.  cryptoceras^  d'Orb. 

Jb  Cancau  tmiralf  133. 


Hauterivian  337 

By  this  time  the  deposits  on  the  northern  slope 
of  the  Caucasus  show  the  same  characteristics  and 
contain  the  same  fossils  as  those  of  the  southern  slope. 
Neocomian  fossils  were  colleded  by  Tsulukidse  ^=^  in 
i868  in  the  Daralagoz  distrid  east  of  Ararat.  At 
Aznaburt  (N.N.W.  of  Nakhichevan)  he  observed  a 
micaceous  sandstone  with  Rhynchonella  cf.  depressa^ 
d'Orb.,  which  may  perhaps  indicate  the  presence  of  the 
Valanginian  horizon ;  this  sandstone  is  overlain  by 
sandy  limestone  of  Turonian  age.  Both  Neocomian 
and  Turonian  beds  have  been  traversed  here  by  dikes 
of  dioritic  rocks.  On  the  pass  between  Aznaburt  and 
Bielava  he  found  the  following  Hauterivian  fossils  in 
black,  sandy  marls : — 

Olcostephanus  (Astteria)  astierianus^  d'Orb.,  0.  Mer- 

issoni^  Oster.;  Polyptichites  bidichotomus^  Leym. 

Oi  the  banks  of  the  Araxes,  4  miles  S.W.  of  Julfa, 

Tsulukidse   found    Exogyra   Coulont,    dOrb.,    derived 

apparently  from  Neocomian  (Hauterivian)  beds  over- 

lying-Palaeozoic 

In  the  East  Karabagh  Valentin  f  has  recorded 
the  presence  of  Neocomian  marls  with  Exogyra  and 
Lima  at  a  locality  between  Jamiat  and  Karadagly, 
east  of  Shusha.  Further  south,  near  the  Araxes,  the 
same  traveller!  established  the  presence  of  Haute- 
rivian beds.  Here,  about  2  miles  west  of  Bartas,  the 
hills  rise  up  from  the  Araxes  in  terraces,  and  consist 
of  a  smoke-grey  limestone  in  thick  beds,  with  a 
dip  to  S.W.  At  the  base  of  the  hills  Valentin 
discovered  a  fossiliferous  zone  with  the  following : — 

Ofbitolina. 

Echinospatagus  (Toxaster)  complanatus^  Ag. 

Terebratula  sella^  Sow. 

*  CM.  Catal,  Tlfit  Musium^  95. 

t  INd.  no,  135. 

X  Btrieht  Senckent.  MUitff.  Ga.  Fhmkfurt  a/M.  1891,  219. 


338  Geology  of  Armenia 

These  fossiliferous  limestones  are  thinly  bedded, 
and  nearly  vertical,  underlying  the  thick  banen 
limestones.  They  form  an  outcrop  about  200  paces  in 
breadth ;  a  zone  of  intrusive  red  porphyrite  lies 
between  this  Neocomian  outcrop  and  the  alluvial 
plain  of  the  Araxes. 

Further  south,  in  Azerbaijan,  Rodler*  found 
Hauterivian  Ammonites  at  Gushaish,  in  the  hills 
on  the  east  coast  of  Lake  Urmi.  The  beds  overlie  the 
Callovian  limestones  already  described  (pp.  322-324), 
and  appear  to  indicate  the  southerly  extension  of 
the  more  pelagic  facies  occurring  near  Aznaburt  (p. 
337)-    Weithofer  f  identified  the  fossils  as  follow*s : — 

Olcostephanus  Straussi^  Weithofer,  (near  to  O.  Kleini^ 
Neum.  &  Uhl.,  of  the  North  German  Hils),  0. 
cf.  Denkmanni^  Neum.  &  Uhl.,  (of  the  Hils); 
Polyptichites  tetrameres^  Weith.,  (3  bad  casts; 
near  to  P.  multiplicatus^  Rom.,  from  Spilsby,  of 
the  bidichotomus  group) ;  Hoplites  cf.  narbotten- 
SIS,  Pict.,  (a  fragment). 

The  fauna,  fragmentary  as  it  is,  shows  a  striking 
similarity  to  the  Hauterivian  faunas  of  Central 
Europe,  and  of  Simbirsk  in  Central  Russia. 

According  to  Tchihatcheff  J  the  Lower  Cretaceous 
is  altogether  absent  from  Asia  Minor;  my  own 
observations,  however,  indicate  the  presence  of  Neo- 
comian  strata  in  the  basin  of  the  Chorokh,  forming  a 
wide  belt  between  the  Pontic  Range  and  the  valley 
of  the  Frat  (pp.  36-40,  232-238). 

My  traverse  between  the  Chorokh  and  the  Kop 
Dagh   (pp.   36-40)    may  be  briefly  summarized  as 

*  Sitzungsber.  k.-k,  Akad,    fVtss.f  Mathan.-nMiuno.   CUuse^  xcm 
204,  and  xcviii.  88. 

t  /did,  xcviii.  756. 

i  AsU  Mineure^  Ghiogie^  ii.  148. 


n 


I 


Hauterivian  339 

follows,  in  downward  succession : — 

1.  White  or  pale  buff  limestone  of  the  Kop  Dagh,  sometimes 
concretionary,  and  often  with  thick  beds  of  chert.  Th;:  microscope  * 
shows  the  presence  of  numerous  Radiolaria  and  sponge  spicules,  and 
sparing  Foraminifera  (Textularia^  b*c.).  The  only  distinctive  fossil  I 
could  find  was  a  fragment  of  a  Belemnite  (DuvcUia  cf.  polygonalU^ 
Blainv.),  indicating  a  Neocomian  (Hauterivian)  age  for  this  limestone. 

2.  Bluish  grey  or  greyish  buff  limestone,  often  concretionary, 
with  white  veins  of  calcite.  It  contains  similar  organic  remains  t 
to  the  last  rock. 

3.  A  thick  series  of  limestones  and  shales,  in  thin  alternating 
layers.  A  plant-sandstone  is  intercalated  among  the  shales  north 
of  the  Kop  Khan. 

My  second  traverse  of  this  region  lay  nearly  parallel 
to  the  first,  at  a  distance  of  about  14  miles  to  the 
eastward,  between  the  Khoshabpunar  pass  and  the 
Chorokh   (pp.    232-238).     My   observations   may    be 

*  Professor  T.G.  Bonney,  F.R.S.,  has  very  kindly  allowed  me  to 
insert  here  his  descriptions  of  the  limestones  i  &  2,  from  specimens 
collected  by  Mr.  H.F.B.  Lynch  on  a  previous  journey.  No.  1 
''is  a  pale  buff,  very  compact,  sharp-jointed  limestone,  traversed 
by  some  small  cracks  filled  up  with  a  dark  mineral,  which,  under 
the  microscope,  is  found  to  consist  of  two  thin  layers  of  calcite 
with  limonite  between.  The  rock  itself  exhibits  an  exceedingly  minuce 
granular  structure,  and  contains  a  considerable  number  of  Radiolana, 
the  majority  spherical,  some  ovoid.  There  are  also  rather  numerous 
spicules,  some  of  which  are  probably  radiolarian,  but  others  suggest 
sponges.  Besides  these,  I  find  a  Textulaiia^  perhaps  two  or  three 
other  Foraminifera^  and  two  small  fragments  which  may  be  coral ; 
a  few  granules  also  occur  of  a  brownish  isotropic  mineral." 

t  Professor  Bonney's  description  of  this  limestone  is  as  follows  : — 
**  A  compact  limestone  of  a  greyish  buff  colour,  traversed  by  minute 
cracks,  which  apparently  are  filled  with  calcite.  The  groundmiss 
is  similar  to  that  of  the  last  rock,  with  similar  Radiolaria,  Spicules 
occur,  often  too  large  for  Radiolaria^  and  some,  triradiate  or  even 
quadriradiate,  are  from  sponges,  possibly  lithistid.  The  rock  contains 
almost  certainly  a  Globigirina^  and  one  or  two  other  fragments  of 
organisms  not  identifiable.     One  or  two  ferruginous  spots  occur. 

In  these  two  slices  it  is  doubtful  whether  the  actual  test  of  the 
Radiolarian  has  not  been  replaced  by  calcite.  The  hardness  of 
the  former  rock  is  nearly  6 ;  of  the  latter  nearly  5,  and  this 
effervesces  more  briskly  with  HCl.     Obviously  they  are  cherty." 


340  Geology  of  Armenia 

summarized  as  follows : — 

I.  Pale  grey  Radiolarian  limeston^.of  Rhoshabpunar,  so  closely 
similar  to  the  Kop  limestone,  as  described  by  Professor  Bonney,  that 
repetition  is  unnecessary. 

3.  Pale  buff,  concretionary  limestone,  with  calcareous  algs 
(AfuftiertaJ,  and  an  occasional  Miliolitut.  It  is  overlain  near  Uie 
Chorokh  by  a  yellow,  compact  sandstone. 

3.  A  thick  series  of  limestones  and  shales,  in  thin  alternating 
layers.  Near  Akhveran  I  found  a  fragmentary  impression,  which, 
according  to  Mr.  G.  C.  Crick,  may  be  of  a  Crioceras^  an  AmyhKeras^ 
or  possibly  of  Macroscaphites  Ivani^  d'Orb. 

In  the  Chorokh  valley  my  observations  (pp.  236- 
-38)  indicate  a  similar  succession,  but  no  fossils  were 
visible : — 

X.     Massive  grey  marble  of  Baiburt  and  Maden  Khan. 
3.    Yellow  calcareous  sandstone  in  thin  beds. 
3.    A  thick  series  of  limestones  and  shales,  in  thin  altemadsg 
layers,  exactly  similar  to  No.  3  of  the  two  foregoing  successions. 

To  the  south  of  the  Chorokh,  near  Halwa  Maden  (a 
few  miles  west  of  my  Kop  traverse)  Abich*  records 
the  following  series  of  strata,  which  lie  on  the  same 
line  of  strike  as  No.  3  of  my  Kop  and  Khoshabpunar 
traverses : — 

X.  Light  grey  I'mestones,  dipping  steeply  northward,  with  a 
i-lcar  impression  of  a  Scaphites  [unfortunately  not  figured]. 

3.     Grey  limestones  with  many  flints. 

3.  Dark  grey  sandy  limestones  in  thick  beds.  All  these  strata 
have  been  much  folded  and  contorted. 

In  this  connedlion  it  is  interesting  to  note  that 
Ratsevitch  \  found  the  Neocomian  Mcicroscaphites 
Ivani,  d'Orb.,  about  100  miles  to  the  N.E.,  in  the  basin 
of  the  Lower  Chorokh;  it  occurred  in  the  lower 
valley  of  the  Ardanuch  river,  "near  Gorgotakhan, 
in  greenish  grey  limestone,  which  forms  the  base  of 
the  Cretaceous  in  this  region." 

*    Ged.  Forsch.  in  den  kauk.  LMmkm^  ii.  in. 
t  Geol.  Ristarcfus  in  the  Batum  ^  Arivin  Districts^  Mat  Geol. 
Cauc.  Tiflis,  1SS7,  Ser.  a,  i.  131,  155.    [In  Russian]. 


Hauterivian  341 

An  intermediate  observation  has  been  made  by 
Hamilton*  at  a  spot  about  50  miles  N.E.  of  Akh- 
veran,  showing  the  marked  persistence  and  uniformity 
of  the  series  of  limestones  and  shales  over  a  large 
area.  He  remarks  on  the  similarity  of  "the  thinly 
bedded  limestone,  sandstone,  and  calcareous  shale 
between  Liesgaff  and  Tortum"  to  the  equally  con- 
torted series,  on  the  same  line  of  strike,  between 
Baiburt  and  Khoshabpunar. 

As  I  have  already  pointed  out  (p.  237),  Abich  was 
evidently  mistaken  in  his  surmise  (unsupported  by 
fossil  evidence)  that  the  Kop  and  Khoshabpunar 
limestones  are  Turonian  in  age,  whereas  my  fossils, 
fragmentary  though  they  are,  indicate  their  Neo- 
comian  age.  He  merely  states  f  that  "  indistindl  and 
very  sparing  traces  of  fossils  point  to  the  limestone 
of  the  Kop  Dagh  and  Khoshabpunar  system  being 
Hippuritic  Limestone,"  and  again:}:  "the  Upper 
Rudistes  zone  prevails  on  the  heights  between  Bai- 
burt and  Sipikor  [nor^  of  Erzingan  plain],  just  as 
at  Khoshabpunar  and  Baiburt/'  It  is  ther^ore  ob- 
vious that  Abich  considered  the  zone  of  strata 
between  the  Frat  valley  on  the  south  and  the  Kelkid 
and  Chorokh  on  the  north  to  belong  to  one  and  the 
same  geological  formation.  Hence  it  may  be  taken 
as  a  strong  presumption  that  this  entire  belt  is 
Neocomian  in  age.  The  Sipikor  heights  of  limestone — 
"  the  same  as  at  Khoshabpunar  "§ — have  been  altered 
into  snow-white  dolomites  near  the  pass  (8ai6  feet), 
but  they  agree  with  the  limestones  of  itop  and 
Khoshabpunar  in  overlying  shaly  strata. 

*  JUsearckis  in  Asia  Minor^  i.  315. 

t  Ged.  Foruh.  in  den  kauk,  Lindemy  ii.  109. 

X  Ibid,  120. 

§  Ibid.  zx8. 


342  Geology  of  Armenia 

Part,  at  any  rate,  of  the  Muzur  Dagh  is  apparently 
Neocomian,  for  Abich,*  in  visiting  Surp  Oannes  in 
one  of  the  chief  gorges  of  the  Muzur  Dagh,  found 
that  "the  geological  relations  fully  coincided  with 
those  of  the  ranges  to  the  north  of  the  Erzingan 
plain,"    implying   therefore  a  Neocomian  age. 

**  The  limestone  is  of  great  thickness,  and  has  usoally  been  altered 
into  dolomitic  marble.  It  overlies  a  series  of  calcareous  flagstones 
alternating  with  clastic  beds  of  serpentinous  debris.  .  .  The  Frat  coti 
through  the  Muzur  Dagh  in  its  entire  height,  fcmning  the  narrow 
gorge  of  Kemakh  Boghaz,  with  nearly  vertiod  walls.  This  exit  of  the 
river  from  the  plain  is  concealed  [from  the  observer  at  Erzingan] 
by  an  outspor  of  the  Akh  Dagh.  This  plateau-height  consists  of 
well-bedded  limestone,  with  silicified  sponges  in  great  numbecs 
[unfortunately  not  specified].  Where  the  limestone  becomes  porous, 
with  drusy  cavities,  it  is  much  penetrated  by  asphalte  Vkt  the 
Neocomian  limestones  of  the  Val  Travers  and  Perte  du  Rh^ne. 
Where  the  asphalte  is  absent  the  limestone  is  dark  grey,  compact, 
foetid,  and  somewhat  siliceous,  with  many  blade  flints.  .  .  The 
marmoraceous  limestone  [of  Sipik(>r  and  the  Muzur  Dagh]  again 
occurs  on  the  gently  rising  plateau-heights  of  the  Melpert  Dagh, 
which  passes  into  the  Keshish  Dagh  [£.  of  Erzingan]." 

It  is  not  improbable  that  the  easterly  continuation 
of  the  Muzur  Dagh — the  range  termed  Merjan,  Baghir 
and  Girdim — will  be  eventually  found  to  consist  of 
Neocomian  limestones.  In  the  Hach  Dagh  (further 
east  in  the  same  range)  Cenomanian  (q.v.)  is  known 
to  occur. 

The  Neocomian  zone  of  the  Chorokh  appears  to  me 
to  be  continued  westward  in  the  basin  of  the  Upper 
Kharshut  river  (pp.  29-35).  The  grey  marble  of 
Baiburt  sinks  beneath  the  S3aiclinal  Eocene  plain 
of  Varzahan  to  rise  again  on  its  western  border  in  the 
similar  barren  marble  heights  of  Osluk  Khan  (p.  35), 
and  the  high  mountain- downs  of  the  Vavuk  Dagh. 
Owing  to  this  part  of  our  journey  being  necessarily 
hurried  it  was  impossible  for  me -to  make  any  extend- 

*  Geai,  Foruh.  in  din  Jkauk,  Liniem^  ii.  123-125. 


Urgonian  343 

ed  search  for  fossils,  but  Wagner^  in  1852  was 
equally  unsuccessful  after  a  search  of  some  hours 
on  the  Vavuk  Dagh ;  even  the  microscope  revealed  no 
traces  of  organisms  in  the  dark  grey  limestone  of 
the  Vavuk  pass.  My  observations  in  this  area  (pp. 
29-35)  ™^y  ^  summarized  in  the  following  suc- 
cession : — 

1.  Dark  grey  maxble  of  Osluk  Khan  and  the  less  altered 
limestone  of  the  Vavuk  Dagh.  The  massive  grey  and  buff  limestones 
of  Murad  Khan  and  Giimiishkhaneh  (where  they  sink  beneath  £ocene 
strata)  appear  to  lie  on  the  same  horizon. 

2.  A  thick  series  of  shales,  sometimes  with  seams  of  limestone. 

3.  Calcareous,  current-bedded  sandstones  of  Tekke. 

It  was  in  the  massive  limestone  a  few  miles  below 
Murad  Khan,  on  the  Kharshut  river,  that  Hamilton  t 
found  "a  fragment  of  a  large  Ammonite,"  whilst 
to  the  south  of  Giimushkhaneh  de  Chancourtois 
noticed  "several  Ammonites,"  which  he  tried  in 
vain  to  detach,  in  limestones  of  Cretaceous  a^e 
(according  to  Viquesnel  J).  There  is  a  similar  vague- 
ness about  the  locality,  which  is  merely  given  as 
**  between  Giimiishkhaneh  and  the  Frat,  on  the  route 
followed  by  Hommaire  de  Hell ;"  it  was  probably  on 
the  northern  slopes  of  the  Giaur  Dagh,  about  10 
miles  S.W.  of  Giimiishkhaneh.  The  neighbourhood 
would  doubtless  repay  investigation. 

The  Upper  Nbocomian  (Barremian)  is  present  in 
Armenia  only  in  its  Urgonian  facies,  similar  to  that 
of  the  Jura. 

On  the  northern  borders  of  Armenia — in  the  Zirula 

*  JRiiu  nach  Pemen^  L  193. 

t  Ruiorchu  in  Asia  Minor^  i.  170. 

X  Bulk  Soc.  GkoL  France^  i^.  2,  vii.  504.  I  cannot,  however, 
find  that  de  Chancourtois  ever  published,  this  statement;  it  was 
probably  made  verbally  to  Viquesnel. 


344  Geology  of  Armmta 

dome — the  Urgonian  beds  show  evidence  of  a  marked 
transgression,  and  of  closer  proximity  to  land  than 
the  very  similar  deposits  in  the  West  Karabagh. 

Near  Marelisi  on  the  southern  slope  of  the  Zirula 
dome  /^e^utema-limestones  and  marls  rest — according 
to  Foumier** — on  basal  conglomerates  and  breccias 
derived  from  the  underlying  granite,  and  to  the  north 
of  Molita  they  rest  discordantly  on  Upper  Jurassic 
clays,  sandstones  and  tuffs.  In  the  softer  parts  of  the 
white  limestone,  e.gy  on  the  Poni  col,  good  specimens 
of  the  following  fossils  can  be  obtained : — 

Numerous  Foraminifera  (Triloculina). 

Corals  [not  specified]. 

Pygaulus. 

Rhynchonella  lata^  Sow.,  in  the  upper  part. 

Cardium ;  Requienia  ammonia^  Goldf.,  R.  (Toucasia) 

Lonsdaliij  Sow.;  Caprotina  trildxUaj  d'Orb. 
Natica ;  Nerinea. 

The  Urgonian  is  also  well  developed  in  the  West 
Karabagh,  but  has  not  been  recorded  from  any 
intermediate  district.  Here,  at  Alikulikent  on  the 
Bergushet  river,  Abichf  discovered  Urgonian  beds 
(which,  however,  he  regarded  as  Cenomanian).  He 
also  states  that  "a  coral-reef*  occurs  in  the  midst 
of  the  formation,  but  since  all  his  corals  are  Turonian 
(q.v.)  it  seems  probable  that  this  "coral-reef"  is 
merely  a  block  of  the  overlying  Turonian  limestone, 
which  has  slipped  down  out  of  place.  The  following 
succession  can  be  made  out  from  Abich's  description : 

I.  Foraminiferal  limestone,  over  loo  feet  thick, 
dull  white  or  yellow,  containing  the  following  fossils : 

Orbitolina  lenticultna^  d'Orb.,  (the  chief  constituent 
of  the  limestone),  0.  tonica,  d'Arch.,  (much  less 

*  Description  ghL  dm  Caucase  ctntral^  134. 
t  GeoL  Beob.  mw^  Xmt  6*  Araxes,  33. 


Urgomam  345 

abundant);  Cyclolina. 
Echinospatagus  (Toxaster)cf.  Caulonif  Ag. 
Nerinea  gigantea^  d'Hombre  Finnas. 

2.  Dark   grey   and  reddish  brown  marls,  with 
Glauber's  salts  appearing  on  the  weathered  surface. 

3.  Light  grey,  compadt,  granular  limestones  with 
Cytherea. 

t.  A  considerable  thickness  of  compact,  light 
imestones,  much  dislocated,  forming  the  narrow 
gorge  of  the  Bergushet  riven  The  characteristic 
Requienia   amtnaniay    Goldf.,   occurs   throughout   this 


This  limestone  was  also  noticed  by  Abich*  on 
the  Trapassar  Dagh,  a  few  miles  S.,  containing  charac- 
teristic Urgonian  fossils.  In  both  places  Abich  has 
mentioned  some  Cenomanian  forms,  which  appear 
to  indicate  that  the  Urgonian  is  overlain  diredlly 
by  Cenomanian  (g.i».),  without  any  of  the  intervening 
formations.  The  Urgonian  fossils  of  the  Trapassar 
Dagh  are  silicified,  and  stand  out  in  high  relief  on  the 
weathered  surface  of  the  rock;  they  consist  of  the 
following  forms : — 

Corals    "in    great    numbers,    especially  of   the 

caespitose  Asttece" 
Requienia   ammonia^    Goldf.;     Caprotina   trilobatay 

d'Orb.  (Monopleura  urgonensiSy  Math.). 
Nerinea  (several  species) ;    Turritella  (numerous 

small  species). 

Still  further  south  the  Orbitolina  limestone  was 
traced  by  Abich  f  on  the  summit  of  the  Chustup 
Dagh  (10,336  feet),  where  it  has  been  greatly  al- 
tered by  intrusive  "  diabases  and  felsites." 

*  Vergleuhendi  gioL  GtundsugeetCt  123. 
t  Geol,  Btob.  tw.  Kur  6*  Araxts^  41. 


\^6  Geology  of.  ^Armenia 


n.      APTIAN. 


Very  little  is  known  as  yet  regarding  the  presence 
of  this  formation  in  Armenia  except  on  its  northern 
borders — on  the  Zirula  dome  and  the  south  slope 
of  the  Central  Caucasus.  Here  the  Aptian  is  iden- 
tical with  that  of  Provence ;  on  the  Zirula  dome  it 
contains — according  to  Siraonovitch*  and  Foumiert 
— the  following  fossils  in  two  divisions : — 

1.  Marls,  slightly  glauconitic,  with 

Terebratula  sella^  Sow. 

Thetis  major ^  Sow.,  T.  minor ^  Sow. 

Phylloceras  Velledce^  ^ch.;  Desmoceras  Mayorianum^ 
d'Orb.;  Acanthoceras  Steinmanniy  Simonov..  A.  con- 
so6nnttm,  d'Orb.;  Belemnopsis  (Belemnites)  semi- 
canaliculatuSf  Bl.,  B.  minimus^  list. 

2.  Marly  limestones  and  grey  Cephalopod  marls; 
the  fauna  differs  from  the  Aptian  of  the  Tkvibuli 
dome  (to  the  N.W.)  in  the  rarity  of  Cephalopods,  the 
abundance  of  oysters,  and  the  great  numbers  of 
littoral  species,  {e.g.^  Patella) : — 

Turbinolia. 

Echinospatagus  (Toxaster)  Collegniiy  Sism.;  Pseudo- 
diadema, 

Serpula. 

Rhynchonella  decipiens^  d'Orb.,  i?.  lineolata^  Pbill., 
R.  lata^  Sow.,  i?.  Lamxirckianay  d'Orb.;  Terebratula 
cf.  Dutempleanaj  d'Orb.,  T.  semistriata^  Defr.,  T. 
pisum^  Sow.,  T.  cf.  semiglobosa,  Sow. 

Ostrea  socialis^  Simonov.,  0.  flabelliformis^  Nfls.,  0. 
sulcata^  Blum.;  AletHryonia  cf.  macroptera^  Sow., 
A.  cf.  carinata^  Lam.;  Gryphoea  hippopodium^  Nib.* 
G.  cf.  vesicularis^  Lam.,  Exogyra  haliotoideay  Sow., 

*  Afat  Geol  Cam.  Tifli»,  x886,r  i88y^ 
t  Descnftion  gioL  du  Camauamiml^  xjjS. 


Aptian  347 

E.  aquila^  d'Oib.;  Plicatula  placunea^  Lam.,  P. 
infiata^  Sow.;  Neithea  cf.  quinqiiecostata,  Sow.; 
Inoceramus  cf.  problematicuSy  d'Orb.;  Trigonia 
dcedala^  Pgrk.;  Arcttca  (Cyprina)  cf,  rostrata,  Rtt.; 
Venilicardia  ct  cardiformis^  d'Orb.;  Panopea  Pre- 
vostif  Desh.,  P.  plicata^  Sow.;  Cor6fS  covrugatay 
Sow.;  Pholadomya  simplex^  Simonov. 

Pdtella. 

Nautilus  piicatuSj  Fitt.,  ^.  NeckerianuSj  Pict.;  Sca- 
^ftites  c/.  oniatii5,  Romer;  Olcostephanus  Decheniy 
Rom.;  noplites ;  AcantAocexas  cross/castatttmy  d'Orb. 

It  is  noticeable  that  this  fauna  contains  npt  only 
a  few  Neocomian  forms,  but  also  several  species 
ivhich  are  characteristic  of  the  Albian,  and  even  of 
the  higher  divisions  of  the  Upper  Cretaceous  of 
Western  Europe.  This  premature  appearance  may 
perhaps  be  accounted  for  in  the  same  way  as  on 
pages  317,  323  aqd  325. 

The  presence  of  Aptian  in  the  Eastern  Border* 
Ranges  of  Armenia  is  established  by  fossils,  which 
were  coUedled  by  Abich  in  the  Upper  Akstafa  valley, 
above  Delejan,  and  are  now  in  the  museum  of  the 
Vienna  University.  According  to  Anthula,^  they 
consist  of  ",the  Aptian  Exogyra  aquilaj  d'Orb.,  with 
other  Cephalopods  and  Brachiopods,"  which,  however, 
he  omits  to  specify. 

About   eighty   miles  south  of  the   Delejan  pass,'n€4TAi^buYt; 
Tsulukidset  found  the  following  fossils  (indicating 
an  Aptian  horizon)  in  sandy »  glauconitic  limestone, 
underlying  Turonian  limestone,  and  probably  over- 
lying the  Neocomian  of  p.  337  : — 

Camerospongia  (Achilleum)  fungiformis^  Gf.  (Ceno.). 
Pentacrinus  noduhsus^  Rom. 

BelentnQpsis  (Belemnites)  semicanaliculatusy  BI.,  B. 
alpinuSy  Oster. 

*  KreidefossilieH  d.  Kamk.^  Beilr.  Pal.  CEttr.-Ung.  xii,  146. 

t  Gioi.  Catai.  lyUs  Mm$emm^  98. 


£ted  on  the  northern 
^ae  localities  as  the 
i^lirgi  in  continuity.  In 
rai:?@Te  Zirula  dome  and 
{^[Gic^casus,  is  divided — 
"  'ftii^  following  groups : 

fzsu&itctiZits,  Ag.,  Cato- 
^t||^x  os^,  Lun. 
^^I£  DestHocertis  Beudm- 
i^^^W'M'^^^^'  ^^-  ^^^  Ceritkium 

*  '^I^*^  i*'fe^^ SJwfe^'^^^^^  ^°^  Pkylloceras 

•  ^  ^  =*'  f  >  gt|£^H^g|^B|admixture  of  Aptian 
•  -a»="="*^*^^S|D^rforms  a  passage  to 

!^  Js:^E^  no  break  between 

is  well  developed  on 
in  the  Cherimela 
>wing  fossils : — 

t'  "»■  -a  *  '»■  •»•  ■»•  ■»•  •*• 


if^^^^^i 


Albian  349 

Asttea. 

Cidaris ;  Discaidea  Bvhuculus^  Ag.;  Catopygus  can- 
natus^  Ag. 

Terebratula  cf.  sella^  Sow. 

Ostrea  socialts^  Simonov.;  PUcatula  cf.  inflata^  Sow.; 
Pecten  cf.  a$per,  Lam.;  Trigonia  dceaala^  Park.; 
Arctica  (Cyprina)  rostrata^  Fitt.;  Thetis  major. 
Sow.,  T.  minor ^  Sow. 

Solariuin  cf.  alpinum^  Pict.  tt  Roux ;  Trochus  Guyoti- 
anus^  Kct  &  Roux;  Turbo  decussatus^  d*Orb.,  T. 
Saxoneti^  Pict.a:  Roux ;  Nerinea  cf.  gaiiltina,  Pict  & 
Camp.;  Cerithium  trimomlCy  d'Orb.,  C.  excat;atfim, 
d'Orb.,  C.  omatfssf  miitn,  Desh.,  C.  subspinosumy  Desh. 

Nautilus;  Phylloceras  Velledos^  Mich.;  Desmoceras 
Beudantif  Brongn.,  D.  Mayorianum,  d'Orb.,  D. 
Timotheanum,  d'Orb.,  D.  lewesiense,  d'Orb.;  Hc^lites 
splendens,  d'Orb.»  H.  Zatttus,  d'Orb.;  Douvilleiceras 
(Acanthoceras)  m^mmillare,  Schloth.;  A.  Lyelli, 
Schl.,  A.  Hoffw^nni,  Simonov.  &Batsevitch,  -A.  c/. 
tonsobnnum,  d'Orb.,  A.  Milletianum^  d'Oib.;  /fys- 
tatoceras  varicosus,  d'Orb.;  Belemnopsis  (Belem- 
tiites)  minimus^  List. 

Abich^  long  ago  stated  that  beds  of  the  age  of 
Gault  existed  in  the  Upper  Akstafa  valley,  above 
Delejan,  (N.W.  of  Lake  Gokcha),  but  he  did  not 
bring  forward  any  fossil  evidence.  Anthulaf  has, 
however,  examined  Abich's  specimens  (now  in  the 
museum  of  the  Vienna  University)  and  states  that 
at  this  locality  there  occurs  **  the  Albian  Desmoceras 
Mayorianum,  d'Orb.,  in  a  dark  limestone,'*  doubtless 
overlying  the  Aptian  cited  on  p.  347. 

*  Vergleichende  gtol.  Gfundnigt  €k.^  21,  and  his  Gtol.  Faruh.  bi 
Jen  katik.  Landem^  ii.  4. 

t  KriuUfossiliiH  d.  Xauk.,  Bdtr.  Pal.  CEstt.-Ung.  dL  146. 


J50  Geology  of  Amutna 

There  appears  to  be  some  indication  of  Albian 
in  the  Zagros,  so  far  as  can  be  judged  from  the  rather 
meagre  list  of  fossils  which  Loftus*  collec^led.  His 
Ammonites  fdanulatus.  Sow.,  from  Bishiweh  is  a  S3nio- 
nym  (according  to  d'Orbigny)  of  the  Albian  Desmo- 
ceras  Mayortanum,  d'Orb.  sp.  His  other  Cretaceous 
fossils  comprise  Cenomanian,  Turonian,  and  Senonian 
forms. 

Perhaps  Albian  may  also  be  presumed  to  occur 
on  the  south  side  of  the  Taurus  from  the  statement 
by  Ainswortht  that  in  his  "Lower  Chalk  a  pre- 
dominance of  peculiar  forms  of  Cephalopods  occurs 
north  of  Mardin/'  This  distridl,  which  has  been 
much  negledled  by  travellers,  would  well  repay 
investigation  by  a  palaeontologist. 

II.      CENOMANUN. 

Abich  X  has  recorded  beds  of  this  age  in  the  Eastern 
Border-Ranges  of  Armenia,  on  the  southern  slopes 
of  the  Murow  Dagh;  here  he  foimd  greenish  grey 
sandstones,  derived  mainly  from  eruptive  rocks,  with 
small  Caprotince  and  cement-stone  geodes  containing 
Metacanthoplites  rhotomagense^  Dcfr.  Probably  these 
strata  are  continued  southwards  into  the  East  Kara- 
bagh  distridl,  for  Abich  §  assumes  a  Cenomanian 
age  for  the  beds  underlying  the  Turonian  between 
Giillaply  and  Pyrjamal  (N.E.  of  Shusha) ;  .they  were 
found  to  contain  "numerous  species  of  Ceritkium^ 
Nerinea^  and  bivalves  **  [not  specified]. 

At  Alikulikent,  on  the  Bergushet  river,  in  the  West 

*  Qttarf,  /(mm.  Geoi.  S^.  n.  285,  a86. 

t  Rutanhtt  im  Assyria^  269. 

X  GeoL  Beob.  itr.  Kmt^  Ataxu^  197. 


Cenomantan  351 

Karabagh,  the  presence  of  Cenomanian  (overl}dng 
Urgonian,  p.  345)  may  be  inferred  from  the  occurrence  of 
Requienta  cf.  carinata^  d^Orb.,  and  Nerinea  aunisiana, 
d'Orb.,  in  Aoich's  list  of  fossils.**  A  similar  inference 
may  be  made  from  the  presence  of  the  former  species 
among  Abich's  fossils  from  the  neighbouring  heights 
of  the  Trapassar  Dagh.t  In  both  places  therefore 
there  would  appear  to  be  a  gap  in  the  succession 
between  the  Urgonian  and  Cenomanian. 

Cenomanian  beds  may  perhaps  be  present  on  the 
Persian  side  of  the  Araxes  from  Abich's  statement:]: 
that  he  found  ^^Caprotina  limestone  (as  well  as 
Hippuritic  Limestone)  overlying  Carboniferous  Lime- 
stone at  Gerger,"  S.  of  Julfa,  on  the  road  to  Tabriz. 

So  far  as  can  be  judged  from  the  present  scanty 
evidence  at  our  disposal,  the  Cenomanian  of  the 
Karabagh  conforms  to  the  Mediterranean  type,  and 
differs  from  that  of  Northern  Armenia  and  the 
Caucasus,  where  its  facies  shows  a  markedly  Central 
European  character.  Probably  the  two  areas  were 
divided  by  a  barrier  of  dry  land  (upheaved  at  the 
close  of  the  Urgonian),  the  site  of  which  would 
appear  to  be  in  the  Somketian  Mountains,  where 
(according  to  Abich§)  the  lowest  member  of  the 
Cretaceous  consists  of  Turonian  Adieonella-heds^  over- 
lying granite. 

On  the  north  side  of  these  mountains,  at  Bielyi- 
kliuch,  Tsulukidse  ||  has  assumed  a  Cenomanian 
age  for  certain  unfossiliferous  sandstones  and  marls, 

*  Geoi,  Bead.  ww.  Kur  6*  Araxes^  33. 

t  Vergltichmii  geol.  GrundMiige  etc,^  124. 

X  Manganeru  in  TranskoMkasiiH^  Bull.  phys.-tDatb.  Acad.  Sd.  St. 
Petersb.  xvt.  314. 

§   Verh.  k.-k,  geoL  RtUhsanst  1887,  30. 

II  Researchis  in  the  Districts  of  the  Algtt  &*  Khram  Rivers^  Mat 
Geol.  Cauc.  ser  3,  i.  59.    [In  Russian]. 


35^  Gedogy  of  Armenia 

which  are  directly  overlain  by  Senonian.  Abich* 
is  more  explicit  in  stating  that  the  lowest  member 
of  the  Cretaceous  in  the  neighbouring  Khram  valley 
consists  of  'impure,  argillaceous  limestones  with 
Exogyra  columba,  Lam.,*'  whilst  in  another  of  his 
writings!  ^^  states  that  these  beds  consist  in  the 
upper  part  of  the  Khram  valley  of  "dark  calcareous 
sandstones  and  brownish  marls  enclosing  Exogyra 
coluwha^  Lam.»  Inoceramus  concentricus^  Park.,  and  an 
Astarte^'*  indicating  the  iipper  beds  of  the  Ceno- 
manian — a  slightly  higher  horizon  than  that  of 
the  Murow  Dagh  (p.  350). 

To  the  north-east,  on  the  Zirula  dome  (Meschic 
Mountains),  Cenomanian  is  well  developed,  according 
to  Foumier.J  The  lower  part  is  continuous  with 
Albian  (p.  348),  and  consists  of  slightly  glauconitic 
sandstones  with  the  following  fossils : — 

Discoidea  subuculuSy  Ag.;  Catopygus  carinatus,  Ag. 
Inoceramus   Lamarcki^  Rom.,    /.   laibiatuSy  Brongn.; 

Pecten  asper^  Lam. 
Haploceras  djumensey  Simonov.;    Acanthoceras  Man- 

telli^  Sow. 

The  upper  beds  are  thick  sandy  limestones  with 
Cerithtum  cf,  maritimum^  d'Orb. 

On  the  eastern  flank  of  these  mountains,  at  Ali 
and  on  the  Poni  col,  the  Lower  Cenomanian  is  stated 
by  Foumier§  to  consist  of  glauconitic,  sandy  marls 
and  sandstones,  containing  the  following  fossils : — 

Discoidea  subuculus^  Ag. 

Alediryonia  carinata^  Lam.;  Exogyra  haliotoidea^  Rom.; 

*  Jfitt.  aus  dem  Kamk.^  Verh.  k.-k.  geoL  Reichsanst.,  1877,  3^- 

t  Mangantrte  in  Transkattkasien^  Boll.  phy8.*math.  Acad.  Sd.  St 
Petenb.  xvi.  316. 

t  Description  gtoL  du  Caucase  antraif  157. 

{  JNd.  161. 


Cenomamau  353 

Spondylus ;  Pecten  Nilssoniy  Goldf.;  Neithea  qutn- 
quecostata^  Sow. 
Haplaceras  djumense^  SimomT.;    Acanthoceras  Man- 
tellij  Sow.;  Hystrichoceras  varians^  Sow. 

This  facies  of  the  Cenomanian  re-appears  to  the 
S.W.  in  the  Pontic  region.  Here  Batsevitch  *  assumes 
a  Cenomanian  age  for  certain  reddish  and  greenish, 
crystalline,  sandy  limestones  and  calcareous  sand- 
stones with  Cnnoid  stem-joints,  occurring  above 
Gorgotakhan,  in  the  gorge  of  the  Ardanuch  river, 
near  its  confluence  with  the  Lower  Chorokh;  they 
overlie  beds  of  Neocomian  age  (p.  340).  In  the  same 
distridt,  between  Matsalakhet  and  Ardanuch,  Batse- 
vitch t  also  found  dark  grey  marls  with  the  following 
fossils : — 

Discoidea  subuculus^  As.;    Catopygus  cartnatuSy  Ag., 

(both  in  the  form  of  casts). 
Belemnites. 

In  view  of  the  supposed  Cenomanian  age  of  the 
Crinoidal  limestone  of  the  Lower  Chorokh  basin, 
it  is  possible  that  the  same  a^e  should  be  ascribed 
to  the  crystalline  Crinoidal  limestone,  which  was 
observed  by  Hamilton  :|:  2  miles  N.E.  of  Kompor  (east 
of  Ispir),  higher  up  the  Chorokh  valley,  and  on  the 
same  line  of  strike. 

Similarly  the  Neocomian  zone  between  the  Chorokh 
and  the  Frat  is  flanked  on  its  southern  border — 
according  to  Abich§— by  "rather  loose-textured 
limestones,  much  reddened,  with  many  indistinct 
Crinoids  and  Bryozoa,  overlain  by  clastic,  reddish 

•  Gici.  Researches  iVr  the  Batum  6h  Artvin  Districts^  Mat  Geol. 
due.  Tiflis,  1887,  Ser.  a,  i.  131.    [In  Russian], 
t  Ibid.  155. 

X  Reseanhts  in  Asia  Minor^  L  ai6. 
4  Geai,  Fpruh.  «•  dea  hauk,  Liaderay  iL  io6. 

aa 


354  Geology  of  Armenia 

brown  and  dark  green  beds/*  in  theJejen  valley 
north-east  of  Ashkala.  Abich  indeed  postulated  a 
Cretaceous  a^e  for  these  beds,  and  they  are  not  im- 
probably Cenomanian. 

Undoubted  Cenomanian  certainly  occurs  on  the 
south  side  of  th^s  Frat,  almost  due  south  of  this 
locality.  The  Cenomanian  limestone  rises  up  to 
form  the  Hach  Da^h — the  E.-W.  chain  which  divides 
the  distrids  of  Terjan  and  Kighi.  According  to 
Tchihatcheffy^  who  crossed  the  Hach  Dagh  from 
Lichig  to  Pardi  {**  Bardak  *'),  on  the  northern  slope, 

• 

**  the  loftjr  and  rodgr  coimtry  between  these  two  villages  is  composed 
of  white,  dark  grey  or  jellowish  lixnestonesy  compact,  cellular,  or 
crystalling  with  conchoidal  or  saccharoidal  fracture,  associated  here 
and  there  with  marls  of  different  colours.  These  rodu  are  frequently 
interrupted  by  dolerltes,  and  are  much  disturbed,  often  vertical  or 
steeply  inclined,  the  dominant  dips  being  to  N.  and  N.W.  They  are 
generally  unfossiliferous,  excepting  in  the  neighbourhood  of  Bardak, 
where  a  yeUowish  white,  compact  Bmestone  encloses  several  casts  of  a 
C^MiUs  recalling  C.  ianaiiiUa^  Gf.  .  •  According  to  d'Arduact 
this  species  of  tydoliies  recalls  those  of  the  fourth  Cretaceous  stage 
of  Charente  Inf6rieure,  /./.,  the  stage  immediately  below  the  Exegyra 
c^iumba  bed.''    Hence  these  beds  would  be  mid-Cenomaniaa  in  age. 

The  only  indication  at  present  of  Cenomanian  beds 
to  the  south  of  the  Armenian  area  is  the  occurrence  of 
Turrilites  cf.  tuberculatus^  Bosc.,  among  the  Cretaceous 
fossils  collected  by  Loftus  X  at  Bishiweh  in  the  2^gros 
range  (the  S.E.  continuation  of  the  Taurus). 

iir.    nnioMiAK. 

The  Turonian  of  Armenia  continues  to  exhibit  the 
Mediterranean  facies,  which  was  already  well-marked 
in  the  Urgonian  and  Cenomanian;  like  the  latter 
it    forms   a   sharp   contrast    to    the   contemporary 

*  Asii  Mintmre^  Ghkgie^  ii.  ijo,  X4t. 

t  Md.^  P0ihn$9l0gk.  too. 

X  Qiuui.  J$wm.  Getl.  Soc.  n^.  tS$,  st6. 


Turonian  355 

deposits  in  the  Caucasus  and  the  Northern  Border- 
Ranges  of  Armenia,  where  the  northern  type  prevails. 
Both  in  lithological  composition  and  in  the  assem- 
blage of  fossils,  the  Turonian  of  Armenia  shows  a 
striking  similarity  to  the  Gosau  beds  of  the  Eastern 
Alps,  indicating  an  identity  of  conditions. 

It  is  particularly  well  developed  in  the  Karabagh 
distridl,  especially  at  Alikulikent,  in  the  gorge  of 
the  Bergushet  river,  where  nearly  the  entire  Creta- 
ceous is  laid  bare  (pp.  344,  351).  Here  Abich^  found 
a  considerable  thickness  of  Turonian  strata,  much 
dislocated,  dipping  northwards,  and  traversed  by 
dikes  of  amygdaloidal  dolerite.  The  succession  is 
as  follows : — 

r.  Yellowish  grey,  landy  marls,  and  grey,  marly  limestones 
with  Gervillia  soUnoiies^  Defr.,  [a  possibly  erroneous  determination, 
for  this  is  an  Aptian  form],  large  Radiolites  (70  mm.  in  diameter  with 
a  broad  upper  valve),  and  various,  ill-preserved  Htppurites, 

3.  Greenish  yellow  sandstones,  looser  in  texture,  and  with  dark 
grey  aigillo-calcareous  concietions,  large  fragments  of  Radiolites^ 
especially  of  R.  Martoni^  Mantell,  AcUtanella  cawasicOf  Zek.,  and 
frequent  fragmentary,  cylindrical  casts,  5-6  mm.  in  diameter,  probably 
of  Ae  Scaphopod  /Ksea  tubuiifera^  Serr. 

In  the  midst  of  the  Urgonian  foraminiferal  lime- 
stone of  this  locality  (p,  344)  Abich  erroneously 
included  a  *'  white  saccharoidal  limestone  like  dolo- 
mite, very  similar  to  the  coral-limestone  of  the  White 
Jura.'*  As  I  have  already  suggested  (p.  344),  this 
"coral-reef"  is  probably  of  the  nature  of  a  block, 
which  has  slipped  down  out  of  place,  for  it  contains 
the  following  Turonian  fossils : — 

Siphonia ;  Vertidllites  incrassata^  d'Orb. 
Astrocoenia    decaphylla^  m-E.  &  H.;     Phyllocoenia 

pediculata^    SIe.;     Thamnastrcea   multiradiata^ 

Reosi,  T.  agaricites^  Gf . 

*  CM.  Biti.  wm.  Kur  6*  Aruxes^  33. 


J56  Geology  of  Armmiia 

Spondylus ;  Arctica  (Cyprina)  ligeriensis^  d'Oib. 
Turritella  cf.  Icetduscula^  Sow.;  Strombus. 

It  was  perhaps  from  this  neighbourhood  that 
Abich  coUeSed  the  following  fossils ;  they  are  now  in 
the  museum  of  the  Vienna  University,  and  have 
recently  bec^  named  by  Anthula,*  who  merely  states 
that  they  come  from  ''  several  localities  on  the  Kara- 
bagh  plateau  " : — 

Radiolites  MartoHi^  ICant;  Corbis  (Fimbria)  armeni- 
aca^  Eichw. 

Turritella  nodosa^  Rodim  T.  (2^ria)  multistriata, 
Reust;  Glauamia  ornata^  Dmch.,  G.  cf.  Refer- 
steini^  Gf.;  Nerinea  pulchella,  d'Orb.,  M  Buchii^ 
Keferst.,  N.  cf.  forojuliensis,  Piraut,  N.  Pailletheana. 
d'Orb.;  Itieria  abbreviata^  Phill.;  Cerithium  Abichi^ 

AnthulA. 

A^aeonella  Renauxiana^  d'Chb.,  A.  cylindraceOy 
Stol.;  Volvulina  laevis^  Sow.,  V.  armeniaca^  AnthnUu 

This  fauna  has  a  distinctly  Eastern  Alpine  charadter 
and  belongs  to  the  Upper  Turonian  horizon.  The 
numerous  uasteropods  indicate  a  shallow  sea  no  less 
than  the  absence  of  Cephalopods. 

An  assemblage  of  fossils  (so  similar  to  the  above 
as  almost  to  suggest  that  by  some  mis-labelling  they 
might  be  the  same  specimens  with  revised  nomen* 
clature)  was  found  by  Abich  f  in  the  Daralagoz,  west 
of  the  Karabagh  region.  Here  the  Actceanella  and 
Nerinea  limestones  are  particularly  well  exposed  in 
the  Vedi  valley,  30  miles  S.E.  of  Erivan  and  about 
25  miles  from  the  Araxes  plain ;  here  the  river  has 
cut  a  narrow  gorge  through  a  fossiliferous  zone  of 
argillaceous  marls,   rich  in  iron  oxide,   alternating 

*  Kreiiefossiiien  d.  jKMu/k.f  Beitr.  Pal.  GEstr.-Ung.  xii*  143. 

t  VergUicheHdi  gtoi.  GnmiMuge  etc.^  133,  124;  and  his  dot.  Btob. 
sw.  Kur  (r  Araxu^  40. 


357 

vrith  brownish  red,  impure,  argillaceous  limestones. 
This  fossiliferous  zone  is  intercalated  in  a  considerable 
series  of  unfossiliferous  limestone  ahd  sandstones. 
The  marls  in  particular  contain  ^^an*  incredible 
number  of  large  Gasteropods,  mostly 'of  Gosau  and 
Uchaux  species.**  Amongst  them  Abich  (I.e.)  was 
able  to  identify  the  following : — 

Cardium. 

Turritella  difficilh^  Zek.;  Glauconia  Kefersteint^  Gf., 

G.  contca^  Zek.,  G.  Coquandiana^  Zckf;  Chemnitzia ; 

Nerinea  Buchii^  Kefent.»   N.  nobilis^  Zek.,   N.  in- 

cavata,    Br.,    N.   moutlifera^   d'Orb*;    Cerithium 

depressum^  Zek.;  Natica. 
Actceonella   rotundatc^   Zek.,     A.    caucasica^    Zek., 

A.  valuta^  Zek.,  A.  obtusa^  Zek. 

The  same  horizon  was  found  by  Abich*  diredUy 
overlying  Palaeozoic  limestones  on  the  Chaganly 
Dagh,  a  northern  parallel-chain  to  the  C^ynlerly 
Dagh,  south  of  the  Vedi  River.  Abich  traced  it  also 
on  the  north  of  the  river,  forming  the  northern 
edge  of  the  Araxes  plain  between  the  Vedi  and  Garni 
Rivers ;  here  the  limestones  have  been  broken  through 
by  the  trachyte  of  the  Ag  Dagh. 

To  the  S.E.,  at  Aznaburt  (N^N.W.  of  Nakhichevan) 
the  Turonian  rises  up  again  trom  beneath  Nummulitic 
Limestone ;  it  forms  cliffs  of  concretionary  marble,  in 
which  Abich  j  found  the  following : — 

Astrea. 

Hibpurites  cf.  cornU'Vaccinum^  Goldf.;  Radiolites. 

Amjeanella. 

At    this   same    locality    Tsiilukidse^    found    the 

*  VifgteMifide  liol.  Grumbuge  Oe..  its,  123. 
%  dial,  Caiai.  Tijtti  Muaum^  97. 


358  Geology  of  Armenia 

following  fossils  in  grey,  sandy  limestones,  overlying 
Aptian  sandstones  (p.  347) : — 

Coscinopora  infundibuliformis^  Goldf. 

Reptomulticava  flabellum^  d'Orb. 

Exogyra conica^ d'OA. {Qenoxxk.) ;  Spondylus;  Lima; 
Entolium  (Pecten)  membranaceus^  Nils.;  Pholado- 
mya  Esmarkii^  Pusch.,  P.  ligeriensis^  d'Orb. 

Nautilus  radiatuSf  Sow.  (Cenom.). 

Although  he  charadierizes  these  limestones  as 
Turanian  yet  the  appearance  of  some  Cenomanian 
forms  may  perhaps  indicate  the  presence  of  that 
formation.  There  appears  to  be  a  similar  admixture 
of  earlier  forms  in  the  Turonian  sandstones  which 
Tsulukidse*  found  to  the  north  of  Aznaburt  in  the 
same  Kasanyaila  mountain-system,  at  a  point  between 
Khachik  and  Belev.  It  is  essentially  a  Pelecypod 
fauna,  and  probably  indicates  a  littoral  deposit.  The 
fossiliferous  sandstones  crop  out  in  a  narrow  valley 
below  vertical  cliffs  of  white  Senonian  limestones. 
The  fossils  are  as  follows  : — 

Prionastraea. 

Botriopygus ;  Echinocorys  (Ananchytes)  ovata^  Leske, 
var.  arnica^  Ag. 

Gryphcea  vesicularis.  Lam.;  Exogyra  auricularis,  Gf.; 
Lima  Pickleri  d'Orb.  (Gosau),  L.  pulchella^  d'Orb., 
L.  Hoeruesii^  Zitt.;  Neithea  quadricostata.  Saw. 
fCenom.),  N.  cf.  septempiicata^  zitt.  (Gosau), 
Modiola  flagellifera^  Forbes,  Arcaligeriensis^ dfOib., 
A.  carinata^  d'Orb.;  Trigonia'  limbata^  d'Oib.; 
Myoconcka  cretacea,  Desh.;  Crassatellites  arcacea. 
C.  regularis,  d'Orb.;  Acropagea  radiata,  d'Oib.; 
Protocardia  HillanUm,  Sow.  (Cenom.);  Isocardia 
prcelonea,  DeOu;  Arctica  (Cyprina)  quadrata^ 
d'Orb.  (Alb.) ;  Venus  plana,  Sow.,  V.  faba,  Sow.; 
«  C7^.  QUai.  ItJBs  Jfmsium,  97. 


Turaniam  359 

Panopea;  Lucina. 
Pteroceras ;  Fusus  lineatus^  Zek.  (Gosau). 
Cinulia  [RinginellaJ  cf.  inflata^  &Otb. 

The  Turonian  is  well  developed  in  the  South  Kara- 
bagh,  south  of  Alikulikent  (p.  335).  Abich^  found 
Plagtoptychus  (Caprina)  Aguilloni,  d'Orb.,  and  large 
Radiolites  on  the  Trapassar  Dagh,  evidently  overlying 
Cenomanian  (p.  3^1),  and  the  outcrop  receives  a  still 
wider  extension  from  Abich's  statement  {Ix.)  that 
^'Turonian  limestones  form  the  alpine  heights  be- 
tween the  Bergushety  Kapan  and  Chundur  rivers." 

Further  south,  in  the  Araxes  gorge  between  Ordu- 
bad  and  Migri,  the  Adiceonella  beds  re-appear  as  dark 
grey,  much  metamorphosed  limestones,  alternating 
with  black,  bituminous,  compact  slates.  The  whole 
series  rests  on  granite  and  syenite,  or  on  Devonian 
and  Carboniferous  limestones,  which  the  Turonian 
limestones  much  resemble.  They  are  traversed  by 
dikes  of  basalt,  and  are  overlain  by  Nummulitic 
sandstone,  also  much  disturbed.  Abich  f  found  them 
to  contain  the  following  fossils : — 

Cardiumy  n.  sp.,  thick-shelled  and  very  numerous, 
near  to  C.  xnndenense^  d'Orb.,  C.  mqutonianum, 
d'Orb.,  C.  conniacunif  d'Orb.,  of  the  Gosau  beds. 

Nerinea. 

A£teonella  gigantea^  Sow.,  A.  obtusa^  Zek.,  A.  cf. 
crassa^  d'Orb. 

It  was  probably  from  the  same  locality  thai 
Tsulukidsef  found  the  following  fossils,  in  "dark, 
argillaceous  sandstones,  forming  heights  along  the 
banks  of  the  Araxes,  near  Kilita,  between  Ketam  and 
Korchevan  "  (a  few  miles  below  Ordubad) : — 

*   VergUichende  geoL  GruHdnigi  eU.f  124. 

t  IHd.^  122,  123 ;  and  his  GeoL  Becb.  tw.  Kur  <y  Araxes^  40. 

X  Geai.  CaUU.  TiflU  JiiuseuM^  96,  97. 


360  Geology  of  Armonia 


amtca^  d'Ardi. 
Spatangus. 
GrypkcM  vesicularis^  Lam.;    Fema  cf.  acuminata^ 

Zittd;    Modiola   cf.   capitata^   Sow.;    Cardium; 

Plagioitychus  (Caprina)  AguilUmi,  <f Oib. 
Natica  noemesiana^  Zek.;    lurritella;    Glaucoma 

aff.  Kefersteinif  Gf.«  G.  turgida^  Zek.,  G.  bicincta^ 

Abich. 

AAceonella  chtusa^  Zek.,   A.  cf.  conica^  Zdc.,   A.  cau- 
casica^  Zdc.,  A.  glandiformts^  Zek. 

The  Tiiioniap  of  the  Karabagh  evidently  crosses 
the  Araxes  into  Persia  along  the  same  N.W-S.E. 
strike,  for  Abich*  found  Radiolites  socialise  d'Oib., 
on  the  northern  slope  of  the  Daradys  (Kharzan)  Dagh 
above  Gerger  (between  Julfa  and  Marand);  on  its 
southern  slope  there  also  occurred  a  white,  coarse- 
grained marble  with  Nerinea^  fragments  of  Hippurites, 
and  Caprotina  (p.  351),  overlying  Carboniferous  Lime- 
stone. 

On  the  north  side  of  the  Bergushet  river  the 
Turonian  of  Alikulikent  (p.  355)  disappears  beneath 
the  vast  pile  of  lavas  and  tuffs  of  tbe  Central  Kasa- 
bagh  plateau ;  but  on  the  N.E.  side  of  the  axis  of  this 
volcanic  depression  (p.  1 1)  a  few  blocks  of  Turonian 
limestone  still  rise  up  like  nunatakr  out  of  the  en- 
veloping lavas.  Several  of  these  block^  have  been 
marked  by  Abicht  on  his  map,  and  Valentin^  has 
added  to  their  number  by  discovering  another  isolated 
block  at  Alagol  of  dazzling  white  limestone,  about  \ 
square  kilometre  ih  extent,  with  remains  of  Rudistes. 
At  the  side  of  this  limestone  Valentin  observed  a  grey 
^o   red,    marly    rock   of   less   thickness,   containing 

*  VergUickefidi  geoL  Gfunigugt  de.<,  123 ;  and  bis  Mtngamnt  in 
Tnmskaukasien^  Bull.  phy8.-math.  Acad.  So.  St  Petenb.  xvL  314. 
t  GtoL  Fi^nch.  in  den  kauk.  Uuutem^  Adas. 
t  Berichi  Saukenb.  Naiuff.  Gis.  Fxankfurt  a/M.  1891,  113. 


Turonian  361 

numerous  individuals  of  an  Orbitoltna. 

In  Hie  East '  Kanibagh  the  A£tasonella  beds  seem 
also  to  be  well  developed  According  to  Abich  ^ 
this  horizon  occurs  in  calcareous  sandstones  near 
Engidja  (N^E.  of  Shusha  and  W.  of  Kishisfakent),  and 
can  be  traced  more  distindly  between  Gullapfy  and 
Nakhichevanik ;  here  the  Turonian  rises  n^  from 
under  Senonian,  dipping  to  the  S.W.t  Among  the 
numerous  fossils  species  of  AAeonella  predominate,  e.g.y 
A.  caucasica^  2A^^  [no  others  specified].  On  nearing 
Nakhiehevanik  the  beds  become  more  and  more 
dislocated  and  traveised  by  zeolitic,  amygdaloidal 
rocks,  which  have  risen  up  alon^^  an  anticlinal  zone. 
At  Nakhichevanik  the  Kimmendgian  limestone  (p. 
330)  rises  up  in  a  fault-sckrp,  overtopping  the 
Cretaceous  at  its  foot. j; 

At  Dashkent  and  Shushakent  ([east  of  Shusha) 
Turonian  beds  overlie  Kimmeridgian  limestone  in 
marked  transgression.  They  reach  a  considerable 
thickness,  and  consist  of  variegated,  calcareous  marls, 
together  with  a  very  charadleristic  stage  of  reddish, 
brecciated  limestones,  rich  in  Bryozoa.§ 

To  the  N.E.,  there  appears  to  be  an  indication 
of  Turonian  in  the  basin  of  the  Upper  Terter  (North 
Karabagh).  In  the  valley  of  the  Bashtubel  stream  (a 
west  tributary  of  the  Du^u,  which  flows  into  the 
Upper  Terter)  Abich  ||  fojund  frequent  blocks  of  a  dark 
brown,  coarse-grained,  i^armoraceous  limestone,  with 
numerous,  large,  ill-preserved  shells  of  the  following : 

Trigania  scdbrdf  Lam.,  (Lower  Turonian) ;  Cardium 
cj.  Neckerianum^  Pict,  (Aptian). 
The  blocks  had  evidently  proceeded  from  the  lower 
slopes  of  the  Daly  Dagh. 

t  jud.  93.     X  nu.  95.      {  nu.  78.      1  nu.  148, 149* 


36a  Geology  of  Armenia 

A  somewhat  interrupted  zone  of  Turonian  lime- 
stones extends  to  the  N.E.  from  the  North  Karabagh 
through  th6  Gokcha  Ranees  (the  so-called  Little 
Caucasus^  to  the  Somketian  Mountains.  Throughout 
this  distndfc  Abich  ^  Cound  the  same  thick  limestones, 
with  Rudistes^  Norinea^  and  Actcsonella^  overlying  old 
crystalline  rocks.  To  the  east  of  Lake  Gokcha  Abich  t 
found  Turoaian  Umestcmes  with  Radiolites  and.  Neri^ 
H$a  on  the  eastern  heights  of  the  great  ^*  cauldron- 
vidley'*  of  Shamkhor  between  Zaglik  (Seglik)  and 
Dashkessan.  Down  in  the  valley,  at  Zaglik  ^where 
alunite  occuis),  basal  conglomerates  crop  out,  dipping 
to  NX.,  and  composed  mainly  of  eruptive  rocks.  At 
Dashkessan,  where  cobalt  is  still  mined  with  success, 
the  cobalt  ores,  according  to  Abich^  **  are  restrided 
to  a  narrow  zone  between  the  massive  magnetic  iron 
ore  and  the  gangue-rock,  which  consists  of  dark^ 
metamorphic  pelites  (at-times  hard  and  quartzdse,  or 
else  softer  and  like  ser^ntine,  but  always  weU 
bedded),  associated  with  IZudtstes-limestone  changed 
into  white,  coarsely  crystalline  marble,  and  also  with 
a  iine«grained  ganiet-rock  in  compadl  bedded  masses/' 

At  the  bottom  of  the  Akstaila  valley  (evidently 
near  Delejan)  Abich  §  found  the  same  Rudisfes-lime- 
stones  with  larffe  species  of  Actasonella  [not  specified] 
overlying  conglomerates  probably  of  Cenomanian 
a^e.  He  traced  them  to  the  N.W.  in  the  Fambak 
(Uorchal)  valley  near  Syzismadani,||  where  they  have 
been  partly  marmorizea  by  intrusive  porphyrites,  and 

t  nUL  117. 

X  BmU.  Soe.  GM.  /'imwf,  tfr.  t,  nL  217,  and  also  in  Smii.  Soc. 
imp.  Nmt  JHflncm^  1S74,  L  978. 

{  VifgliUkitdt gkt.  ^nmigSgt^  44r«,  lai,  itt. 

I  MM.  67,  M. 


Turanian  563 

contain  copper  ores  near  the  contadl-zoiie. 

Further  north,  in  the  Somketian  Mountains,  the 
Turonian — ^aoconling  to  Abich  ^ — ^is  conformably 
overlain  b^  Senonian,  and  rests  in  the  Lialwar  Dagh 
on  Jurassic  clastic  deposits  (Bajocian,  p.  315).  In 
its  lower  part  it  has  been  much  intruded  by  felsitic 
rocks. 

The^  most  northerly  point  in  Arlneaia  where  the 
Turonian  exhibits  its  Mediterranean  fades  is  in  the 
valley  of  the  Botnis  river,  a  tributary  of  the  Masha- 
vera.  Here  Abich  f  found  the  limcsiooe  to  contain 
Rcfdiolites,  HippuriteSf  Nerinea^  and  Actaamella.  It 
overlies  ancient  crystalline  rocks,  and  it  is  utilized 
as  a  flux  in  the  iron  works. 

Only  a  little  further  north,  at  Bielyikliuch,  the 
Turonian  is  absent,  and  Senonian  beds  rest  directly 
on  Cenomanian.  Turonian  is  not  met  with  again 
until  the  Meschic  Mountains,  where  it  presents  the 
totally  difiEetent  fades  of  Central  Europe.  Here,  on 
the  Zirula  dome  at  Poni,  the  Turonian — ^according  to 
Foumierj; — is  a  marly  chalk  with  the  following 
fossils : — 

Discaidea  infera^  Ag. 

Terebratula  semigbbosa^  Sow. 

Ostrea   proboscidea^  d'Azch.;    Inouramus    labiatus^ 
BroDgn.,  /.  Cuinerif  d'Orb.,  /.  calcinatus,  d'Orb. 

Bad  impressions  of  Ammonites. 

This  sudden  change  in  charadier  can  only  have 
l^en  caused  by  a  land-barrier  on  the  site  of  the 
present  Thrialetic-Imeretic  Mountains —  the  Northern 
Border- Ranges  of  Armenia. 

*  Geoi.  Beob.  sw.  Kur  6r*  Araxis^  9;  Aad'bb  VergliuhimU  gtoi, 
Gtundniigi  Hc,^  65. 

t  G€ol.  Btob.  ifcf.  Kur  6*  Aruxis^  9;  and  hb  MiU.  out  dim  Kamk,^ 
Verb.  k.-k.  geol.  Reichsanst,  1877,  30. 

X  Dtscriptum  gjtoL  dm  Caucau  UfUrul^  159,  :i6j. 


j64  Geoiogy  of  Armenia 

The  recent  volcanic  accumulations  of  the  Armenian 
plateau  have  concealed  the  westward  continuation 
of  the  Turonian  of  the  Daralagoz-districft  (p.  357),  but 
it  can  still  be  traced  in  the  lolty  Aghri  or  dhatin 
Oagh — "  the  spine  of  Armenia.*'  According  to  Abich  ^ 
this  range  consists  essentially  of  ^  an  axis  of  syenite 
and  gabbro,  flanked  on  each  side  by  a  zone  of 
Hippuritic  Limestone/*  These  rocks  rise  up  first  of 
ail  in  the  Akarak  valley  at  Takjaltu  from  beneath 
the  series  of  great  volcanoes  which  extend  westwards 
from  Ararat  to  the  Perli  Dagh.  At  Tandurek  and  on 
the  Takjaltu  Dagh  (south  of  Kulpi)  the  Turonian 
consists  of  **  Hippuritic  marmoraceous  limestone  and 
greenish  slates,  with  W.N.W.  strike,"  intruded  by 
andesites,  and  overlain  by  Eocene  sandstones.^  It 
was  probably  from  these  greenish  slaty  rocks  that 
Abich  found  the  ^^Inaceramus  with  numerous  fucoids 
in  Flysch-like  beds  from  the  Araxes  valley  near 
Kulpi."  The  specimen  is  now  in  the  museum  of 
the  Vienna  University. J 

It  was  probably  this  zone  of  Turonian  limestone 
which  Koch§  noticed  further  west,  whilst  crossing 
the  Aghri  Dagh  on  his  way  from  Toprak-kala  to 
Kagizman.  He  states  that  '^the  whole  base  of  the 
mountain,  here  called  Chashlu  Gedik,  consists  of  a 
greyish  yellow  limestone,  similar  to  the  Jura  lime- 
stone," and  further  on,  in  the  Karajehennem  valley, 
he  observed  ^*  a  gre3dsh  green,  fine*grained  sandstone," 
which  may  be  a  continuation  of  the  fucoidal  sand- 
stone of  Takjaltu. 

Further  west  Abich  |I  found  that  numbers  of  pebbles 

*  Gefii.  Farsch.  im  den  kauk.  Zatukm,  ii.  159. 

t  JMd.  154,  i6i« 

t  Anthula,  KfrnirfostUim  d.  Kauk.,  Beitr.  PaL  (Estr.-U&g.  m.  143. 

{  Reisi  im^mttbdUn  Gebirge  ^v,  ii.  449. 

Jl  Qiol.  EtndL  m  den  kauk.  Zandem,  ii.  59,  69. 


Turonian  365 

^ 

brought  down  frcHii  this  ranj^e  into  the  Pasin  plain 
by  the  Araxes  and  its  tributaries  consisted  of  a  "  grey 
limestone  with  traces  of  Rudistes.*'  Still  further  west 
along  this  range  Abich^  has  recorded  the  presence 
of  siucified  limestones  of  the  upper  Rudistes  zone  in 
the  great  Palandoken  cirque,  south  of  Erzerum,  but 
his  statement  is  perhaps  open  to  doubt  (p.  50). 

A  somewhat  more  definite  occurrence  has  been 
recorded  by  Tchihatcheff  *  lower  down  the  Frat  basin 
at  Elmali  (Almaly),  between  Mamakhatun  and  Er- 
zingan.  Here  he  found  Trochosmilia  cf.  inconstans,  de 
From.,  (TuronianV  underljang  Eocene  strata. 

The  only  indication  of  Turonian  in  the  Pontic 
region  ckxrurs  near  the  confluence  of  the  Ardanuch 
river  with  the  Lower  Chorokh.  Here  BatsevitchJ 
found  a  very  thick  series  of  limestones,  crystalline 
in  places,  overlying  Cenomanian  and  underlying 
Senonian  strata ;  although  he  did  not  discover  any 
fossils  iie  assumed  a  Turonian  age  for  the  series. 

Turning  to  the  Southern  Border-Ranges  of  Armenia 
the  references  to  Turonian  beds  are  somewhat  vague 
and  disconnected.  L6ftus§  has  described  a  con- 
tinuous zone  of  Hippuritic  Limestone  through  the 
21agros  on  the  qast  side  of  the  long  outcrop  of  Num- 
mulitic  Limestone.  In  the  series  he  found  several 
bands  of  ironstone  and  also  of  red  chert. ||  He 
considered  the  succession  in  general  to  be  as  follows : 

z.  Blue  marls,  sometimes  bituimnous,  and  cream-coloured 
limestones  with  jFkd^  passing  imperceptibly  ihto  the  conformably 
overlying  Nummulitic  Limestone.    [SenonianJ.  • 

*  Geci.  Foruh,  in  den  kauk,  Landem^  ii.  106. 

t  AsU  Mituurt^  Ghhgie^  ii.  130. 

X  GtoL  Rtuarchis  in  tlu  Batum  ^  Artuin  Districts^  Mat.  Geo). 
Cauc.  Tiflis,  1885,  Ser.  3»  i.  iji,  155.    [In  Russian]^ 

{  Quart,  raum.  GeoL  Sac,  xL  285. 


)66  Geology  of  Armsma 

f.    Hitfd,    liglit   grey,    indociled   limestone  with   SpharuUitu 
(Turonian]. 

Both  Turonian  and  Senonian  are  represented  by 
the  following  fossils  which  Loftus*  found  in  the 
Cretaceous  belt  of  the  Zagros,  without,  however, 
specifying  the  localities : — 

Corals. 

Serpula;  worm-casts. 

Rhynchonella ;  Terebratula  cf.  cornea^  Sow.  (Senon.). 

Ostrea  (small,   plicated);    Gryphcea;    Exogyra; 

Pecten;  Radiolites;  Venus;  Tellina. 
Voluta. 
Ammonites  {2^  feet  in  diameter) ;  Belemnites. 

In  the  centre  of  a  mass  of  bright,  lustrous  bitumen 
from  the  Zagros  [no  locality]  Loftusf  found  "a 
brown,  striated,  rounded  Carpolite,  a  small  Nucula^ 
and  Terebratula  cf.  carnea^  Sow.** 

At  Imam  Mir  Akhmet  Loftusf  found  the  following 
succession : — 

X.  Hard,  yellow,  compact,  crystalline  limestone,  probably 
Nummulitic.  In  its  lowest  beds  it  becomes  aeam-coloixred,  passing 
into 

3.  Hard,  reddish,  grey  or  cream-coloored,  Iitho|;raphic  lime- 
stone, very  compact,  but  splitting  into  thin  layers,  with  abundant 
specimens  of  Ammomtis  (crushed),  Grypkam^  Stfpula^  TMima^  and 
Voluta.  Interstratified  are  numerous,  tabular  layers  of  black, 
siliceous  substance,  with  a  strong  bituminous  odour  when  strudi:* 
[Upper  Cretaceous]. 

3.  Highly  crystalline,  blue,  foetid  limestcmes.  [Pkobably  Car- 
boniferous]. 

Indications  of  Upper  Cretaceous  beds  in  the  Taurus 
are  even  less  definite.  They  appear  to  flank  the 
metamorphic  axis  (p.  279)  of  tiie  Kuriki  Dagh 
(between  the  valleys  of  the  Khabur  and  Great  Zab 

*  Quart.  Jcmm.  GeoL  Soc.  iL  tSS. 

t  Ibid.  287.  X  aU.  S89. 


Turonian  367 

r 

rivers).  On  the  southern  flank— according  to  Ains- 
worth* — ^**the  iron  mines  of  Tura  VDuri  (south  of 
Lisan)  are  worked  on  the  surface  in  beds  of  iron 
oxide,  disposed  parallel  to  the  strata  of  a  fissile, 
yellow  limestone,  dipping  26^  W.  and  belonging 
to  the  Upper  Chalk  formation.  •  .  The  rock  of  the 
Kuriki  Dagh  [flanking  its  schistose  axis,  7652  feet]  is 
a  slaty,  argillaceous  limestone,  dipping  parallel  with 
the  slope  to  Taraspino.  A  mine  on  the  side  of  the 
hill  showed  a  promising  vein  of  galena,  the  veinstone 
being  barytes.  Madreporites  abounded  in  this  lime- 
stone." On  the  northern  flank  of  the  Kuriki  Dagh 
Ainsworth  also  noticed  ^*a  lead  mine  in  the  Kiyau 
valley  in  a  slaty,  yellow  limestone  of  the  Upper 
Chalk."  This  limestone  is  continued  to  the  granitic 
and  gneissic  area  between  Eslaya  and  Julamerk. 

Further  west,  to  the  north  of  Mardin,  Ainsworth  j 
found  that  after  crossing  the  first  line  of  hills  north  of 
Mardin  **  the  Upper  Chalk  occurs — ^a  friable,  laminar 
rock,  buff-yellow,  remarkablv  redolent  with  botry- 
oidal  hematite,  frequently  hollow  with  cavities  filled 
with  calcspar,  and  sometimes  forming  beds.  At 
Khan  Kajurin  [25  miles  from  Mardin  on  the  way 
to  Diarbekr]  the  yellow,  fissile  Upper  Chalk  comes  in 
contact  with  white,  cretaceous  Lower  Chalk,  which 
contains  flints  and  Echinids  [not  specified],  and 
occupies  the  country  to  the  north  to  a  few  miles 
of  Khan  Aghpur.  ,  The  Lower  Chalk  presents  less 
lofty  hills  than  the  Upper  Chalk."   . 

It  was  probably  the  easterly  extension  of  these 
Cretaceous  strata  which  de  Chqncourtois  J  traversed 
just  south  of  the  Tigris,  whilst  travelling  from  Jezireh 

*  Travels  6*  Researches  in  Asia  Minor^  etc,j  ii.  315. 

t  Researchis  in  Assyria^  etc.^  350. 

X  Coff^tes-rendm  AeaJL  Sa.^  xviii.  829. 


368  Geology  of  Armenia 

to  Diarbekr— "a  siliceous  limestone  rising  up  from 
beneath  basalt  to  form  mountainous  country/^  In 
the  opposite  direction,  W^.W.  of  the  basaltic  Karaja 
Oagh,  the  hills  near  Urfa— according  to  Viscoimt 
PoUington* — ^are  composed  of  "chalk  mixed  with 
flints/'  and  perhaps  form  the  westerly  continuation  of 
this  series. 

Ainsworthf  is  of  opinion  that  "all  the  coimtiy 
between  the  Euphrates  and  Urfai  from  Bir  to  Samsat, 
including  the  Nimrud  chain  of  hills,  is  composed 
of  the  Upper  and  Lower  Chalk  formation,"  that  "  the 
Chalk  formation  becomes  in  the  hills  of  Nimrud  a 
compadl  limestone,"  and  that  the  Sham-chai  (west  of 
the  Karaja  Dagh)  "flows  through  a  deep  ravine  in 
Chalk/'J  It  is  more  than  probable,  however,  that 
part  of  Ainsworth's  "  Chalk  "  may,  on  future  investi- 
gation, be  found  to  include  Eocene  and  even  Miocene 
strata. 

Vr.       BBKOMIAN. 

At  the  close  of  the  Turonian  the  Thrialetic  land- 
barrier,  which  had  separated  the'rArmenian  from  th? 
Caucasian  .'waters  (p.  363),  sank  down  and  permitte|9 
an  invasion  of  the  northern  fauna  into  the  southeim 
area.  It  is,  however,  probable  that  land  had  jji^ 
at  the  same  time  to  the  west  and  south  of  a  line 
running  roughly  from  Batum  to  A^rat.  Numerous 
submarine  eruptions  took  place  during  the  Senonian, 
probably  as  a  result  of  these  earth-movements. 

The  Senonian  of  Armenia  is  well  developed  in  the 
Karabagh  region.  In  the  West  Karabag^,  at  Alikuli- 
kent  in  the  Bergushet  valley,  the  Senonian— according 

*  /imrm.  Ruy.  Gtogr.  Soe.^  x.  449. 

t  Riuamhis  in  Astyfrn,  4te.^  23. 

t  Ihitfeb  6*  JUsianka  in  Asia  Minor^  /ft:,  L  S70. 


Senanian  369 

to  Abich* — conformably  overlies  the  Turonian  (p. 
355)  J  it  shows  a  northerly  dip,  and  is  strongly  dis- 
located and  traversed  by  dikes  of  amygdaloidal 
dolerite.  It  attains  altogether  a  thickness  of  several 
hundred  feet.  The  lower  part  consists  of  thick  beds 
of  white,  fissile,  chalky  limestone,  containing 

Echinocorys  (Ananchytes)  avata^  Leske,  (crushed). 
Inoceramus  Cuvieri^  d'Orb.  (fragments),   /.  Crippsi^ 

Mant  (casts) ;   and  a  few  other  indeterminable 

bivalves. 
Ammonites  (large  casts). 

The  upper  beds  are  perhaps  of  Danian  age  and 
consist  of  calcareous,  sandy  marls  with  gypsum,  over- 
lying light  grey  argillaceous  marls.  They  indicate 
a  gradual  shallowing  of  the  sea,  with  the  formation 
of  lagoons. 

Further  south,  in  the  Araxes  gorge  below  Ordubad, 
Tsulukidse  f  found  Inoceramus  CrUfpsi^  Mant.,  in  sand- 
stones overlying  Turonian  (p.  359).  In  the  northern 
part  of  the  West  Karabagh  Range  the  Senonian  is 
represented  by  deposits  of  a  littoral  character:  on 
either  side  of  the  Bechinakh  pass  (5307  feet),  between 
Arakhly  and  the  Salvarty  Dagh  (10,422  feet)  Abich  % 
observed  pelites  and  sandstones  (derived  mainly  from 
felsite-porphyry)  alternating/  with  Senonian  lime- 
stones— the  whole  series  being  "traversed  by  veins 
of  eurite-porphyries  and  fissile  trachyte-porphyries 
like  those  of  Ponza."  Probably  a  series  of  adlive 
trachytic  volcanoes  studded  this  ridge,  separating  the 
deeper,  clearer  waters  of  the  East  Karabagh  area  {vide 
infra)  from  the  Daralagoz  gulf  on  the  west. 

In  the  Kasanyaila  and  Gyneshik  distridb   of  the 

*  G€ol.  Biob.  WW.  Kur  6*  Araxes^  30. 

t  GtoL  CataL  Tiflis  Museum^  97. 

i  Gtol.  Biob.  WW.  Kur  6*  Araxes^  a;,  98. 

28 


370  Geology  of  Armenia 

Daralagoz  region  the  Senonian  is  well  represented 
by  marls  and  limestone  overlying  Turonian  (p.  358). 
Abich*  found  good  sedlions  in  the  valleys  of  the 
Poyas  and  fiielav  (Belev)  rivers,  tributaries  to  the 
eastern  Arpa-chai.  Among  other  fossils  [not  si>ecified] 
Abich  found  Austinocrinus  Erckerti,  Dames,  and  Ino- 
ceramus  Crippsi^  Mant.  The  specimens  are  now  in 
the  museum  of  the  Vienna  University.! 

Senonian  fossils  (now  in  the  Tinis  Museum)  were 
found  by  Tsulukidse^  in  the  following  localities 
in  the  Kasanyaila  distridl,  between  the  Arpa-chai  and 
Jagry-chai : — 

In  white  limestones  below  Akhgach,  in  the  valley 
of  the  Jagry-chai : — 

Inoceramus  Cuvieri^  d'Orb*,  /.  annulatus^  Goldf. 

In  compact,  white  limestones  at  Belev : — 

Micraster  cor-anguinum^  Lam. 
Inoceramus  cordtformis^  Sow.,   /.  tnvolutus^  Sow.,   /• 
latus^  Mant.,  /.  Brongniartiy  Sow.,  /.  striatus^  Mant 

At  Pais,  in  the  Chalkhan  Dagh : — 

Terebratula  semiglobosa^  Sow. 
Ostrea  gigantea,  Bran.;    Pecten  (Entolium)  mem- 
branaceus,  Nils. 

In  compaxft,  grey  marl,  overlying  compact,  white, 
limestones  of  the  same  age,  between  the  Chalkhan 
Dagh  and  Kairagash,  west  of  Aznaburt : — 

Inoceramus  Crippsi^  Mant. 

In  glauconitic,  sandy  limestone  at  Terkesh : — 

Terebratula  carnea^  Sow. 

Between  Arpa  and  Omaga-dansik : — Terebratula. 

*   VergieidieHde  geoi.  Grundsuge  tU.  xaa. 

t  Anthula,  KmSefossUum  d.  Kauk.^  Beitr.  FaL  CEstr.*Ung.  ziL  X4S* 

X  GeoL  Catal,  TiJKs  Museum,  99,  1x3. 


Senonian  371 

In  grey  marl  2  miles  N.W.  of  Terkesh : — 

Echinocorys  (Ananchytes)  (xoata^  Leske,  together  with 
the  following  Cenomanian  forms  which  may  either 
indicate  the  presence  of  that  stage,  or  may  be  derived 
specimens : — 

Catopygus  carinatuSy  Ag.,  C.  columbartus^  d'Arch.; 
C.  ovulunty  Lam.;  Py gurus ;  Holaster  trecensis^ 
Lcym.;  Off  aster  in  flatus,  Desh. 

In  the  East  Karabagh  the  Senonian  is  extensively 
developed  in  the  form  of  fissile  lithographic  lime- 
stone (Plattenkalk).  At  Merikent  (N.E.  of  Shusha) 
this  limestone — according  to  Abich*^ — occurs  in  beds 
of  varying  thickness,  alternating  with  light  grey, 
shaly  marls.  This  lithographic  limestone  contains 
12.24-46.43%  of  calcium  silicate  uniformly  dissem- 
inated through  the  rock,  which  is  extremely  hard, 
compaA  and  marmoraceous ;  it  splits  into  large  slabs, 
yielding  a  sonorous  sound  when  struck.  Fossils  are 
scarce  and  consist  of  Annelid  casts,  much  twisted  and 
of  considerable  length,  4-5  mm.  in  breadth,  occurring 
in  relief  on  the  upper  surface  of  the  limestone  beds. 
Abich  (Lc.)  considered  that  these  casts  belong  to 
the  same  species  of  Scolica  which  is  found  in  the  shaly 
beds  of  St.  Sebastian  Bay  in  the  Cretaceous  of  the 
Pyrenees;  the  lateral  margins  are  strongly  ciliated, 
almost  denticulate  and  very  sharply  defined. 

Between  Merikent  and  Giillaply  these  strata  have 
a  southerly  dip  down  the  valley  to  the  foot-hills, 
and  present  steep,  bare  escarpments  to  the  north.f 

According  to  Abich  J  the  white,  marly  limestone 
range  between  Merikent  and  Dagdagan  is  clearly 
Senonian,  overlying  Turonian  (p.  361).  The  strata 
have  a  gentle  northerly  dip,  and  are  often  reddened 

*  Ged.  Btob.  ww*  Kur  6*  Araxes,  83. 
t  Ibid.  93.  t  I^id.  78. 


37^  Geology  of  Armenia 

by  doleritic  basalts. 

Valentin^  found  that  the  low  watershed  at  Akh- 
bulakh  between  the  Kelus  aad  Kendalen  (Kjundeljan) 
livers  is  formed  of  similar,  fissile,  snow-white  litho- 
graphic limestone  (Plattenkalh).  ^He  traced  it  from 
Akhbulakh  through  Akhkent  (E.N.E.  of  Shushaj 
to  Dagraz  in  the  north.  To  the  south  he  found  the 
same  limestone  near  Jebrail  in  tne  Araxes  valley, 
where  it  forms  the  last  flattened  spurs  towards  the 
plain,  and  a  little  further  west  at  Kubatlu  on  the  left 
bank  of  the  Lower  Bergushet  riven  It  is  always 
snow-white,  with  conchoidal,  ^intery  fradhure ;  small 
white  splinters  occur  continually  in  the  vegetation 
of  the  rounded  heights;  the  only  fossils  of  this 
limestone  are  the  following : — 

Annelid  casts  (Scolica^  p.  371). 
Echinocorys  (Ananchytes)  wata^  Leske. 
Inoceramus  fill  preserved). 
Amtnonites  (large  casts). 

Valentin  {l.c.)  noticed  that  both  Akhkent  and 
Karakent  are  built  of  this  ^^  Plattenkalk*'  and  at 
Sus  (a  little  to  the  south)  this  zone  of  Senonian 
limestone  is  succeeded  (perhaps  by  a  fault)  by  an 
outcrop  of  Kimmeridgian  limestone.  At  Akhkent 
Valentin  t  found  Inoceramus  Cuvieri,  d'Orb.»  in  the 
limestone,  and  at  Khachmach  (east  of  Shusha)  he 
coUedled  this  fossil  together  with  other  bivalves. 
Ammonites,  and  sea-urchins  [not  specified]. 

At  Nakhichevanik  (a  few  miles  north  of  Dagraz) 
Abich  it:  found  the  reddened  Senonian  **  PlattenlMlk  ** 

to   conformably   overlie   Turonian   (p.    361).     The 

i 

*  Btruki  Snukmb.  Natutf.  Ges.  Fxmnkfurta/IL  1891, 196, 197,  S09. 

t  G€al.  CataL  TiJIis  Museum,  124. 

i  Gud,  Biob.  9W.  Kur  6*  Araxes^  103  it  sff. 


Senonian  373 

Cretaceous  range  is  4500  feet  high  in  the  Ketuk  Dagh 
south  of  Nakhichevanik.  The  Senonian  beds  are  very 
greatly  disturbed  and  altered  by  igneous  rocks  in 
IJie  Cretaceous  heights  between  Nakhichevanik  aiid 
the  Barsar  (Kargar)  river.  In  proceeding  eastwards 
along  the  left  bank  of  this  river  to  the  broad  Creta- 
ceous outspurs,  Abich  found  in  the  valleys  near 
Kishishkent  Senonian  ^^  Plattenkalk  "  similar  to  that 
of  Merikent,  but  without  any  intercalations  of  litho- 
graphic limestones.  Here  they  underly  psammitic 
limestones  containing  Bryozoa  and  fragments  of 
Inoceramus.  On  the  western  slope  of  the  ridge  (4260 
feet)  above  Kishishkent  Abich*  also  found  thick, 
compadl  limestones  containing : — 

Echinocorys  (Ananchytes)  ovata,  Leske. 
Inoceramus  Cuvieri^  d'Orb. 

In  general^  the  Senonian  of  the  Karabagh  seems 
to  show  indications  of  deeper  water  to  the  N.E.  The 
formation  is  doubtless  continued  beneatti  the  great 
depression  of  the  Kur  (p.  10),  for  not  only  does  it  form 
the  foothills  of  the  Eastern  Border-Ranges,  where 
they  sink  down  to  the  wide  Kur  plain,!  but  similar 
limestones  with  the  same  fossils  rise  up  again  on 
the  opposite  side  in  the  Caucasus,  and  in  Daghestan 
reach  the  thickness — according  to  Sjogren  J — of  over 
1000  metres. 

The  Senonian  of  the  Karabagh  is  continued  to 
N.W.  thrpugh  the  Gokcha  Ranges.  Near  Elizabetpol 
the  only  recorded  Senonian  fossils  are  Foraminifera. 
Mittermayer§  worked  out  and  identified  the  following 
forms  from  saponite  occurring  at  Elenendorf  (marked 

*  Gtol,  Beob.  »w,  Kur  &*  Araxes^  io7»  108. 

t  Abich,  VergUichende  geoL  Grundtugg  etc,  is 3. 

X  Jahrb,  k.-k.  geoL  Reichsanst^  xxxix  (1889),  425. 

§  Mikro/auna  d,  oberen  Knidischichtenv.  Transkauk.  Erlangen,  1896. 


J74  Geology  of  Armenia 

K)  and  a  little  further  west  at  Shamkhor  (marked  S) : 

Nodosaria  subconstridiaf  Mittennayer  (£),  N.  sp.  (S) ; 
Glandulina  panicea.  Mitt.  (K) ;  Textularia  glob- 
ularia^  Ehrenb.  (E  &  S) ;  Bolivina  (S) ;  Anomalina 
biumbilicata^  Mitt  (S);  Globigerina  cretacea, 
tfOrb.  (E  &  S) ;  Orbulina  (S) ;  Discorbina  Struvei, 
Mitt.  (E) ;  Rotalia  (S). 

In  the  valley  of  the  Zaglik  river  (a  tributary  of 
the  Shamkhor  river)  a  block  of  coniferous  wood 
was  found  by  Arzruni  at  Pekhthor  Arwak.  It  is  now 
in  the  Breslau  Museum,  and  has  been  described  by 
Gurich*  under  the  name  Araucarioxylon  armeniacum. 
It  is  uncertain,  however,  whether  this  specimen 
was  found  in  Senonian  strata,  or  in  the  imderlying 
Turonian  of  this  locality  (p.  362). 

Senonian  limestones — ^according  to  Abichf — preach 
their  greatest  height  ip  the  region  between  the 
Debeda,  Pambak,  and  Akstafa  rivers.  To  the  west 
and  north  the  littoral  facies  predominates,  especially 
in  the  lower  part  of  the  formation.  It  occurs  in  many 
localities  in  the  Pambak  and  Gokcha  Ranges*  over- 
lying the  granites,  which  until  this  period  had 
apparently  remained  dry  land.  Felsitic  tuflFs  and 
conglomerates  are  particularly  noticeable  in  this 
series  ;  they  are  mostly  yellow  in  colour,  and  regular- 
ly bedded.  These  rocks  form  the  eastern  border  of 
the  Somketian  Mountains,  and — ^according  to  Abich  J 
— are  extremely  siliceous,  forming  excellent  building- 
stone  ;  for  this  purpose  **  the  finest  varieties  of  these 
compadl,  felsitic  pelites  are  to  be  found  in  the  small 
cirque- mountains  of  Ardevi  and  Golurt,  near  Uzun- 
lyar"   on   the   Debeda   (Borchal)   river.    The  same 

*  2^atsch,  deuisch.  geol,  Ges,  xxxvii.  433-440.  • 
t  Ver^eUhefuU  gtU.  Grumbiige  etc.^  68. 
X  GioL  Biobn  MW.  Kur  6*  Araxes^  xo. 


Senonian  375 

rocks  occur  in  the  Lialwar  Dagh,  overlying  Jurassic 
(p-  315)  and  Turonian  beds  (p.  363);  Siey  pass 
upwards  into  limestones  with  Ihoceramus.  Beds  of 
the  same  age  were  found  by  Abich*  to  occur  in 

''the  lowor  vaUeys  of  the  Debeda  and  Akstafa  rivers  (near  Kazakh), 
and  near  Pepis  and  Gotkent  in  the  Inja  valley:  they  consist  of 
stratified  and  unstratified  felsite-porphyries  and  eruptive  pyroxenic 
rocks  with  oligoclase.'^ 

Clastic  rocks  of  the  same  felsitic  character  and 
Senonian  age  are  well  exposed  further  west,  in  the 
heights  above  Alexandropol  (Giimri),  where  the  Pam- 
baJ^  Ranges  disappear  beneath  the  recent  lavas  and 
tuffs  of  the  Armenian  plateau. 

The  pass  from  the  Pambak  valley  to  Alexandropol 
— according  to  Abich  t — 

'*  rises  steeply  over  siliceous  Senonian  limestones,  alternating  with 
felsitic  tuffs  and  banded,  jaspery  strata,  at  a  height  of  6967  feet, 
not  far  from  Aghbulakh.  .  .  These  clastic  and  metamorphosed 
members  of  the  Senonian  are  overlain  by  siliceous  Nummulitic 
limestone,  whilst  to  the  north  of  this  pass  this  series  consists  of 
green  tuf&  and  quartzose  peHtes,  alternating  with  green  felsite- 
porphyries,  and  associated  with  reddish  sandstones  and  argillaceous 
marls-— the  whole  series  overlying  the  Hippuritic  Limestone,  ser- 
pentine, &c.,  of  the  Aglagan  chain/' 

Abich,t    in    a   second    passage,    states    that 

*'on  approaching  this  pass  of  Aghkent  from  the  upper  valley  of 
Amamly  .  .  .  there  is  a  considerable  outcrop  of  silicified  white  chalk 
(with  Inaceramus)  interbedded  with  clastic  beds,  in  whidi  fragments 
of  a  chrysoprase-green  felsite-porphyry  are  the  most  abundant 
elements.  Overlying  these  are  pelites,  metamorphosed  into  green 
and  brown  jaspery  rocks  as  hard  as  hornstone,  with  intercalary 
beds  of  reddish,  argillaceous  limestones  crowded  with  shells  of 
Inaceramus,  This  thick  chalk  formation  is  continued  westwards  in 
the  low  Kamshut  and  Rieti  heights,  which  form  the  northern 
boundary  of  the  plateau  of  Alexandropol/' 

*  Aperfu  de  mes  vayagis  en  TranscaucasU  en  1864^  Bull.  Acad.  Sci. 
St  Petersb.,  v.  17. 

t  Geoi.  Foruh,  in  den  kauk.  Ldndem^  ii.  20. 

X  Ibid.  40,  41. 


376  Geology  of  Aptnenia 

The  same  clastic  beds  come  to  light  even  further 
west  in  the  volcanic  plateau,  along  the  southern  base 
of  Modatappa,  just  beneath  the  lava-sheets.  Here 
Abich*  observed  dark  brown  sandstones  and  pelites, 
derived  from  constituents  of  felsite-porphyry  and 
diabase,  occupying  large  stretches  of  country,  par- 
ticularly in  the  marshy  region  between  Modatappa 
and  the  Okys  Dagh. 

An  outcrop  of  ^'Senonian  limestone  rises  up  in 
white  cliffs  m  th^  upper  Karabulakh  valley  near 
Akha,**  t  just  north  of  the  Shindiyar  Dagh,  between 
the  upper  valleys  of  the  Mashavera  and  Khram  rivers, 
in  the  western  part  of  the  Somketian  Mountains. 

To  the  north  of  these  mountains  Senonian  (and 
Danian)  strata  are  well  developed  in  the  dome  of 
Bielyikliuch,  overlying  sandstones  and  marls,  which 
are  regarded  by  Tsulukidse  %  as  Cenomanian  (p.  351), 
The  Senonian  limestones  are  very  compact  here,  and 
pass  into  true  lithographic  limestones  like  those  of 
the  Karabagh.  They  contain  manganese  ore,  yield- 
ing 87  %  of  manganese  dioxide. 

Among  the  fossils  from  Bielyikliuch  which  Abidi 
eolle(5led  (now  in  the  museum  of  the  Vienna  Uni* 
versity)  Anthula  §  has  identified  the  following : — 

Echinoconus  globulus^  Dcsor  (Senonian);   Stegaster 
caucMtcuSf  Le  Dm.  (Danian). 

Abich  II  had  previously  recorded  from  Bielyikliudi 
the  following  Senonian  fossils : — 

*  Geoi.  Foruh.  in  den  kauk.  L&ndem^  iii.  a6. 

t  Abich,  GtoL  Beob,  tw.  Kur  &*  Araxts^  5. 

X  Researches  in  the  Districts  of  the  Alget  6*  Khram  Rwert,  Mat 
Geol.  Cauc.  ser.  a,  i.  9.    [In  Russian]. 

(  Kmde/bssilien  d.  Kauk.,  Beitr.  Pal.  CEstr.-Ung.  m  146. 

II  Jlianganeru  in  Transhauhasien^  Boll.  phyi.omatlL  Acad.  ScL  Sl 
Petersb.  xvi.  317. 


Senonian  377 

Echinocorys  (Ananchytes)  ovata^  Leske  ;  Echinoconus. 
Inoceramus  Cuvieri^  d'Orb. 

Here  also  Tsulukidse  *  colledled  the  following  forms 
.(now  in  the  Tiflis  Museum)  :— 

Inoceramus  Cuvteri^  d'Orb.>  /.  Crippsiy  Mant 

According  to  Abich  f  the  Senonian  of  Bielyikliuch 
indicates  great  contemporaneous  volcanic  adlivity. 
The  upper  beds  of  the  thick  Senonian  series  alternate 

**  with  friction-conglomerates  of  a  very  quartzose,  litfaoidal  quartz- 
trachyte,  which  cannot  be  distinguished  in  band-specimens  from 
old  felsite-porphyriesy  when  it  occurs  in  dikes  or  sheets.  These 
clastic  deposits  are  confined  to  the  south  side  of  the  Thrialetic 
Mountains,  and  are  absolutely  wanting  on  the  norUi  side;  they 
are  traversed  by  dikes  of  quartz-trachytes." 

In  another  place  Abich  i^  describes  these  rocks  as 

<<  siliceous  light-coloured  pelites,  and  sheets  of  quartz-porphyries 
together  with  their  firagmenUl  sandstones,  underlying  the  white,  nuuiy 
Upper  Chalk.'' 

Senonian  limestones — according  to  Abich  § — com- 
pose the  greater  part  of  the  Arjevan  and  Thrialetic 
chains.     Their    framework    consists    essentially    of 

*^  limestone  altered  into  marbles,  interbedded  with  hornblende-  and 
augite-andesites,  melaphyres  and  amygdaloidal,  diabasic  lavas,  ^th 
conglomerates  derived  from  these  lavas,"  which  were  evidently 
submarine  and  contemporaneous. 

The  chief  summits  of  the  ThrijJetic  Mountains,  e.g., 
the  flat-topped  Sanislo  (9351  feet),  frequently  consist 
of  marble,  which  has  shown  a  greater  resistance  to 
weathering  than  the  clitic  and  eruptive  rocks.  The 
whole  series  has  been  traversed  by  dikes  of  Tertiarj^ 

"^  Geol.  CaiaL  Tiflis  Mustum^  itZ. 

t  iff//,  aus  dim  KatdLy  Verb.  k.-k.  ge<d.  Reichsanst,  1877,  3s  ; 
and  his  Gtol,  Fmrsek.  in  dm  kauk.  Limitm^  Atlas,  PL  3,  Sections. 

X  Mofigamru  in  Trasukaukasien^  Bull.  phys.-math.  Acad.  Sd.  St. 
P^rsb.  xvi.  316. 

{  Giol.  Fortck.  in  dm  kauk.  Landem,  uL  50-52,  55. 


37^  Geology  of  Armenia 

lavas.  It  is  evident  from  Abich's  descriptions  ^  that 
the  Thrialetic  chain  sinks  down  by  a  succession  of 
great  step-faults  to  the  south,  through  the  Shubaret 
plateau  to  the  volcanic  plateau  of  Armenia.  The 
small  Shubaret  plateau  (6753  ^^^)  ^^  composed  of 
'^Senonian  limestones  altered  to  marble,  associated 
with  brecciated  marbles  and  reddish,  banded  lime- 
stones." 

Still  further  north  a  gradually  sinking  land-area 
must  have  existed  in  Senonian  times  on  the  Zirula 
dome  (Meschic  Mountains).  The  gneiss  and  granite 
of  this  dome  are  diredlly  overlain  by  granitic  breccias 
and  conglomerates  formed'!^  the  encroachment  of 
the  Senonian  sea.  According  to  Simonovitcfaf  and 
Foumier  ^  the  calcareous  beds  intercalated  in  this 
detrital  series  contain  shore-molluscs  such  as  limpets 
(Helcion);  fossils  are  veiy  numerous  at  Kharaguli, 
Bielogore  and  Satchkheri  and  comprise  the  following 
forms : — 

Siphonia;  Coscinopora. 

Pentacrinus  Bronni^  Hag. 

Echinoconus    albogalerus^    Lamk. ;      Echinobrissus ; 

Echinocorys  (Ananchytes)  ovata^  Leske,  E.  artatuSf 

Simonov.,    E.   conoideus^  Ag.;    Micraster  car-an- 

guinuniy  Lam. ;  Hetniaster. 
Reticulipora  obliqua^  Reuss. 
Rhynchonella  difformiSy  Lam*;    Terebratula  canea^ 

Sow.,  T.  carneaytLt.  semiglobosa^  Sow.,  7.  obeso.  Sow. 
Ostrea  santonensisy  d'Orb.,    0.  dzewensis.  Simooo^.; 

Exogyra   decussata   (comu-arietis),   Coq.;    Ino* 

ceramus  Crippsiy  Mant. 

*  Gfol.  Foruh.  in  dm  kauk.  Ldndemy  iii.  X7»  57,  59. 

t  Geology  of  th$  District  of  Shofopon^  Mat  GeoL  Cane  Tiflii, 
1886.     [In  Russian]. 

i  Description  glol,  du  Caucau  central^  159. 


Senonian  379 

Helcion. 

Ammonites;  Belemnitella  mucronata^  ScUoth.; 
A6iinocamax  quadratus^  Bl. 

The  upper  part  yields  fossils  of  a  Danian  facies,  viz.: 

Coraster  Munieri^  Seunes;  Stegaster;  Echinocarys 
(Ananchytes)  cf.  Douvilleiy  Seunes,  E,  cf.  conoidea, 
Ag. 

Inoceramus  Crippsif  Mant. 

Nautilus  cf.  danicus,  Schloth. 

The  same  beds  were  traced  by  Fburnier**  on  the 
eastern  flanks  of  the  Meschic  Mountains ;  near  Suram 
the  Senonian  contains  abundant  fossils,  as  follows  : — 

Crmoid  remains. 

Echinobrissus ;  Echinocorys  (Ananchytes)  ovata^  Leske, 

E.  semiglchuSf  Lam.,  E.  Beaumonti^  Lam. 
Terebratula  carnea^  Sow.,  T.  camea  var.  semigUhosa^ 

Sow. 
Exogyra  decussata  [cornu-arietis] ,  Coq.;  Inoceramus 

Cuvieri,  d'Orb.,  I.  Crippsi,  Mant. 
Ammonites ;  Belemnitella  mucronata,  Schloth. 

Here  again  the  upper  part  appears  to  show  a 
Danian  character;  it  is  a  very  soft  chalk,  and 
contains  the  following  fossils  a  little  south  of  Poni : — 

Coraster  Munieriy  Seunes ;   Echinocorys  cf.  Douvillei, 

Seunes,    £.  ovata^  Leske,  E.  cf.  semiglchus^  Lam. 
Nautilus  danicus^  Schloth. 

The  Cretaceous  in  the  eastern  part  of  the  Thrialetic 
Range,  between  Gori  and  Mtskhet,  is — according 
to  Foumier  if/.c.) —  represented  only  by  Senonian  with 
Inoceramus  Cuvieri  d'Orh. 

The  Senonian^  which  has  sunk  down  from  the 
Thrialetic  Range  to  disappear  beneath  the  Tertiarv- 
basin  of  Akhalzik  and  the  volcanic  plateau  of  Akhal- 

*  Description  ghl.  dm  Cauease  central^  162. 


380  Geology  of  Armenia 

kalaki  and  Dokhus  Punar,  re-appears  on  the  inner 
side  of  the  Pontic  Range,  in  the  Ardanudi  valley. 
In  the  intervening  district  it  was  only  exceptionally 
that  Abich^  was  able  to  observe  ^*  clastic  beds  of 
Senonian  age"  at  the  bottom  of  the  deep  gorges v 
which  the  Kur  (and  its  tributaries)  have  eroded  to 
a  depth  of  several  thousand  feet  in  the  volcanic 
plateau  of  Dokhus  Punar. 

According  to  Abich  f  the  whole  valley  of  Ardanuch 
has  been  carved  out  of  Senonian  limestones  overljring 
a  "dioritic  conglomerate  with  psammitic  layers!^* 
On  the  right  bank  the  beds  consist  of  a  thick  series  of 

*'  weU-bedded,  white  chalk,  with  many^  horixontal  layers  of  flat  flints. 
This  is  divided  into  many  deep  gullies  towards  the  upper  part  of 
the  Panaskert  valley,  all  leading  up  to  the  chalk  plateau  whidi 
culminates  in  the  Agh  Dagh  and  the  lofty  Tigma  uagjtiJ' 

On  the  talakheti  ridge  (W.S.W.  of  Ardanuch), 
on  the  way  from  Tsria  to  Gurjany,  Batsevitcti  J  found 
a  grey  limestone,  containing  the  following  fossils : — 

Echinocorys  (Ananchytes)  ouata^  Leske>  and  casts  of 

other  sea*urchins. 
Inoceramus  Cuvievi^  d'Orb. 

On  the  right  bank  of  the  Ardanuch  river,  between 
Matsalakhet  and  Ardanuch,  Batsevitch  |  also  observed 
a  series  ot  sandy  and  argillaceous,  yellow  and  grey, 
limestones,  altemat;ing  with  dark  grey  marls,  and 
containing  the  following  fossils : — 

Miaragter  cor-anguinum^  Lam.;    Echinocorys  (Anan- 
chytes) ovata^  Leske. 

These  beds  conformably  overlie  Turonian   lime* 

^  GeoL  Fanch,  in  den  kauk.  Landem^  iii.  76. 

laid.  VL  182. 
X  Ged.  Researches  in  the  Batum  6*  Artvin  DistriOt^  Mat  GeoL 
Cauc  Tiflis,  1885,  ^*  ^»  ^*  <34*    \^^  Russian], 
(  lUd.  X31,  135;  and  GioL  C^tal.  Tijlis  Museum^  lax. 


Senanian  381 

stones  (p.  365).  Abich  *  had  previously  noticed  that 
yellow,  Senonian  limestones,  overlying  clastic  beds, 
occur  right  down  the  Ardanuch  valley  to  its  con- 
fluence with  the  Chorokh  valley,  and  thence  to 
Artvin,  where  the  strata  become  increasingly  disturbed 
and  dislocated.  At  Dampal  (below  Artvin)  this 
much  dislocated  clastic  formation  resumes  its  original 
regular  bedding,  giving  a  gentler  charadter  to  the 
Chorokh  valley. 

At  Norbert  (further  south,  in  the  valley  of  the  Olti 
river)  Abich  f  again  found  Senonian  limestones,  over- 
lying a  series  of  regularly  bedded,  coarse  and  fine 
conglomerates  and  sandstone,  derived  from  dioritic 
porphyrites  as  well  as  from  hornblende*  and  pyroxene^ 
andesites.  These  clastic  beds  rest  on  granite  and 
other  crystalline  rocks,  and  probably  indicate  the 
proximity  of  a  coast-line  (p.  368). 

Reference  has  already  (p.  366)  been  made  to  an 
indication  of  Senonian  with  Terebratula  of.  camea. 
Sow.,  in  the  Cretaceous  belt  of  the  Zagros,  south 
of  Armenia. 

*  GeoL  Fonek.  in  den  Huk.  Landem^  ii.  181,  187. 
t  JUd.  181. 


ADDENDA. 

P^e  36a.  DoQvOU  {BuU.  Soc.  GM.  Francis  wbt.  4,  1901,  441) 
has  ezunined  some  fragmeiits  of  BSppuriUs  from  the  summit  oT  the 
range  N.E.  of  Delejan,  5  or  6  kilom.  from  the  Haghanm  monastery, 
and  considers  that  tbqr  belong  to  a  species  recalling  IT.  pmtOaUus 
and  JK  CkalmQsi* 

Page  367.    DouTin^  {!.€.)  has  also  recorded  a  group  of  two  Fadif* 
Afitf,  recalling  J?.  4f9tgmdis,  from  the  right  bank  of  the  Euphrates 
near  Keban,  in  the  immediate  neighbourhood  of  Hakim  Khi^. 


CHAPTER  XXI 


THB  EOCBNE  OF  ARMENIA 


I*     LOWBE  SOCnOL 


Beds  of  this  age  art  sometimes  difficult  to  separate 
from  the  uppennost  Cretaceous,  owing  to  similarity 
of  sediments  and  a  very  general  abs^oe  of  fossils, 
but  there  is  usually  a  marked  unconformity.  There 
was  a  considerable  amount  of  contemporaneous  vol- 
canic adUvity,  much  of  which  was  submarine  and 
consequently  inimical  to  the  development  of  organic 
life.  The  Lower  Eocene  is  well  represented  in  the 
Northern  Border-Ranees  of  Armenia  and  along  the 
southern  foot  of  the  Taurus.  It  is  apparently  absent 
in  the  mid-Araxes  region,  where  there  is  a  very 
marked  break  between  the  Upper  Cretaceous  and 
the  Middle  Eocene.  Probably  a  great  part  of  the 
Armenian  plateau  was  dry  land  at  this  time,  and 
the  Lower  Eocene  merely  marks  the  beginning  dl 
a  transgression  which  became  more  extensive  in  the 
Middle  Eocene. 

In  the  north  of  Armenia  and  in  thie  Caucasus  the 
Lower  Eocene  presents  the  fades  of  the  Alpine  Flysch. 
It  consists — according  to  Simonovitch  *  and  Foumier  f 

*  Gmkgjf  cf  Oi  Ihimfy  AMaink  JOatm^  Mat.  GeoL  Cauc.  TiBis, 
1886.    [In  Rtmsian]. 
t  IkscripHam  gt$l.  im  (Umeasi  atUrot^  it6,  171,  181. 


r 


.  Lower  Eocene  383 

&  thick  series  of  dark,  argillaceous,  fissile  sand- 
stones and  marls ;  harder  beds  of  variegated  shal^ 
and  laminated  limestones  with  flints  are  intercalated 
in  the  series,  but  the  only  fossils  are  obscure  im- 
pressions of  thick-stalked  Fucoids  (Chondrites).  Tuffs 
and  volcanic  breccias  frequently  occur,  together  with 
contemporary  diabases,  andesites,  dacites  and  tra- 
chytes. They  are  well  developed  on  the  north  side  of 
the  Akhalzik  basin,  near  Borjom  and  Mikhailov. 
According  to  Foumier  ^  the  same  beck  are  continued 
northwards  into  the  basins  of  the  Kvirila  and  Lpwer 
Rion.  They  occur  also  to  the  eastward  ;|  on  the  N. 
slope  of  the  Thrialetic  Mountains  the  Lower  Eocene 

'*  is  folded  be&eath  the  Qetaceous,  and  at  the  same  time  overliet 
the  Upper  Eocene^  which  is  discordantlj  overstepped  by  Sarmatiao. 
The  folds  of  the  Thiialetic  Mountains  are  overfolded  to  the  north, 
in  the  contrary  direction  to  those  of  the  Caucasus." 

Abich  X  had  long  ago  found  that  the  Lower  Eocene 

**  darky  eruptive  fragmental  beds,  with  Ckondrites^  of  the  Thrialetic 
ranges  rest  transgressively  on  Soionian  in  the  Alget  valley,  west  of 
Tiflis,  and  pass  insensibly  upwards  into  mid-Eocene  sandstones, 
€.g.f  at  Bidyikliuch." 

Although  the  Lower  Eocene  appears  in  general 
to  be  absent  in  the  Eastern  Border-Ranges  of  Ar- 
menia,  Abich  §  found  in  the  Duchu  valley  (North 
Karabagh)  a  series  of  dark  greenish  grey  sandstones, 
with  intercalary  argillaceous  beds,  which — ^apparently 
on  purely  stratigraphical  grounds — he  considered 
to  be  of  Lower  Eocene  age. 

To  the  south  of  Armenia,  Loftus  |(  found  the  Lower 

*  DescriptioH  f^,  dm  Caucast  central^  i68. 
t  Ibid.  174,  175. 

%  Vergkickindi  giol.  Grundtugi  eic^  14s ;    and  his  Jfiit.  out  dm 
JSjntJk,^  Verh.  k.-L  geol.  Reichsanst,  1877,  3a. 
{  doL  Biob.  fw.  Kur  6*  Arexes^  147. 
H  Quart,  /aiaru.  GtoL  Soc.^  xL  275,  285, 


^ 


386  Geology  of  Armema 

of  Jezireh.  This  range  is  '*  a  steep  wall  of  limestone^ 
nearly  horizontal,"  with  ^  sandstones  and  limestone* 
<x>ngIomerates  with  coarse  marls  "  according  to  Ains- 
worth;*  the  associated  carbonaceous  beds  of  these 
Lower  Eocene  beds  evidently  reach  a  fuller  develop- 
ment in  this  region^  for  Major  Maunsell  has  kindly 
informed  me  that  coal  occurs  in  shales  north  of  the 
iudi  Dagh,  at  Segirik^  just  below  Shemakh;  and 
also;that  ''good,  worlotble  coal  occurs  at  Herbol,** 
evidently  on  the  same  line  of  strike,  in  the  valley 
of  the  Hazil  Su,  at  the  east  end  of  the  Judi  Da^h. 
Still  further  S.fi.,  probably  in  the  same  beds,  Major 
Maunsell  observed  ''good,  bituminous  coal-seams  in 
vertical,  shaly  strata  N.  of  Begil  [on  the  N.  side 
of  the  Akra  Dagh  and  Begil  Dagh,  between  Akra  and 
the  Zab].  The  seams  crop  out  on  the  surfece  for 
200  yards." 

Whilst  the  foregoing  observations  point  to  the 
Lower  Eocene  age  of  tiiese  coal  beds  in  the  Tauric 
region,  the  same  age  may  be  tentatively  assumed 
at  present  with  regard  to  the  coal  occurring  between 
Van  and  the  Persian  frontier.  Major  Maunsell^  is 
again  my  authority  for  stating  that  coal  exists 
between  Hazara  and  Rahmin,  on  the  N.  slope  of  the 
Akhta  Dagh — the  range  between  the  Mermid  and 
Mehmedik  rivers — ^associated  with  red  and  altered 
marls ;  coal  is  also  present  on  the  N.  slope  of  the 
Nermit  Dagh,  just  S.  of  Seivan.  Almost  due  £.  of 
this  locality  Blauf  observed  "fine  coal  between 
Gemavig  and  the  valley  of  the  Kotur  river,"  /.e., 
between  Kotur  and  the  frontier,  and  again  "thick 
beds  of  coal  nearer  Kotur."  Still  further  E.,  near 
Kamzakendi  (Hamzakend),  in  the  E.-W.  range  N. 

*  Travels  ^  Rts$anhis  in  Asia  Minar^  dt^  fi.  359. 

t  Fom  Urmia-Sie  naek  dm  FoM^See^  PeCenmuui's  MittlL  bu  1863, 
toz,  207. 


Lower  Eocene  387 

of  Dilman,  he  noticed  ''fragments  of  coal  in  a  dry 
stream-bed,  associated  with  the  same  pebbles  as  at 
Khoi."  Now  Abich*  has  given  a  sedtion  of  this 
range  between  Khoi  and  Dilman,  ascribing  an  Eocene 
age  to  the  ''clastic  beds  recalling  Alberese  and 
Maciginp,*'  which  flank  the  central  mass  of  gabbrj 
and  serpentine  of  the  Gadykh-tapa  (5287  feet). 
According   to   Loftusf  these    beds    consist    of 

I.    Thick  beds  of  calcareous  mark  in  alternate  blue  and  white 


a.    Deqp  red»  indurated  calcareous  marl. 

This  series  rests  on  Palaeozoic  and  crystalline  rocks. 
Near  Kotur,  close  to  the  coal  recorded  b^  Blau, 
Loftus}   found   a  similar   series   of   unfossiliferous 

'*rod,  aharad  sandstones,  oontainiog  thick  beds  of  gravd-con- 
g^omeiate^  and  resting  on  a  hvd  chert  rock,  a  pinkish  grey,  frequcntljr 
traversed  by  veins  of  a  beautiful  red  felspar  with  talc." 

Hommaire.de  Hell  §  also  noticed  this  series  both  to 
£«  and  W.  of  Kotur,  forming  ''masses  of  altered 
and  disturbed  conglomerate,  green  or  more  generally 
a  red-brown,  inclined  in  all  directions/' 

It  is  probable  that  a  Lower  Eocene  age  should 
be  ascribed  to  the  serpentinous  sandstones  of  the 
Upper  Frat ;  at  Mamakhatun  Abich  ||  found  them 
to  be  discordantly  overlain  by  Miocene  limestones. 
Proceeding  eastwards  from  this  place  Abich  (Ix.} 
found  at  Karajuren  (6360  feet)  a  thick  series  of 

**  sandstooes  and  oooglomeialei,  accompanied  by  highly  fei^uginottii^ 
chqrsy  interbedded  witii  grey  limestone  (presumably  Eocene),  as  at 
*'or  Ogly.  •  •  The  Miriam  Dagh  is  essentially  composed  of  this  series 

*  Giffl.  Fonek.  in  den  ktmk.  Landirm^  ii  274,  275. 

t  Qiuurt.  Jourm.  CM.  S^c^  xi  308. 

%  lUd.  311. 

f  Vayagi  m  TWyacar  ^^.,  ir.  GMcgitf  103. 

I  GiffL  JFItnch.  im  dm  Mauk.  LSniem^  iL  xs8-i3s. 


388  Geology  of  Armtnia 

of  daik  wl  ftrmta  (with  S.W.-K.E.  strike  and  steep  dip  to  S.E.), 
traversed  by  sones  of  grass  green  serpentine.  •  .  From  Gok-dara 
(63 IS  feet)  to  the  pass  (7093  feet)  brown  and  greenish  grqr,  softer 
beds  penist  (with  S.W.-N.E.  strike),  continuing  as  far  as  Kiikurthi 
[Kirkut,  6443  fee^  The  serpentinous  conglomerate  recurs  here, 
;dtemating  (as  at  Karajuren)  with  smokegrejr,  ttnfossiliferoas»  frag- 
mentel  limestones,  dipping  60*  S,  Thence,  as  far  as  Chisheyan, 
a  considerable  succession  of  friaUe  pelites  (with  S.W.  strike)  is  visible^ 
t^ndosing  a  seam  of  lignite.  Chishevan  itself  Ues  on  variegated 
marls  [Upper  Miocene],  with  thick  beds^of  gypsum.'' 

Abich*  indeed  has  stated  in  general  terms  that 

'*a  U^te  like  true  coal  occurs  in  many  places  in  Armenia  and 
Geoigia  in  yellowish  sandstones  as  well  as  in  dark,  coarse  and 
Ane,  fngmental  beds  iriiidi  endose  NummmtiUs^  although  sparini^y." 

The  littoral  charadter  of  the  deposits  of  the  Miriam 
Dagh  and  the  fadt  of  their  containing  lignite  renders 
it  not  improbable  that  the  beds  noticed  by  Taylor  f 
in  the  Merjan  Dagh  (to  S.W.)  are  also  of  the  same 
date.  In  crossing  this  range  from  Erzingan  by  the 
Merjan  Boghaz  (10,000  feet),  *'  past  barren,  crumbling 
cliffs/*  he  observed  "  in  two  places  indications  of  cosu 
.  near  Vank  (Surp  Lusavoritch)  on  the  S.  side  of  the 
range. 

Indications  of  coal  occur  lower  down  the  Frat 
valley  at  Keumur  (=coal  in  Turkish),  a  few  miles 
N.  of  Kemakh.  llie  coal  occurs  above  the  village — 
according  to  Wunsch^ — and  is  apparently  worked 
to  a  slight  extent.§  It  cannot  be  rar  from  this  place 
that  Siller  is  situated — **  4  hoi^irs  from  the  Kuru-chai  ** 
— where,  so  Taylor  Q  was  told,  there  is  a  rich  deposit 
of  very  good  coal." 

Now  the  succession  in  the  valley  of  the  Kuru-chai, 

*  VifgUcknuli  gfot.   ChmtukUff  tfflr.  139. 
t  /mrm,  Jt^y.  Gitjgr.  JW.  xncniL  535. 
X  Mittk.  A.4.  gi^gr.  Git.  ncrii.  (1S84)  904. 

§  Arsruni,  £t»n0$nu  f0siiim  9f  Uf€  Anmmtmt.    Tiflis,  1894.    pn 

Armenian]. 


Middle  &  Upper  Eocene  389 

above  Chiftlik  (35.  miles  W.S.W.  of  Keumur,  and 
on  the  same  line  of  strike)  consists — ^according  to 
Hommaife  de^Hell  ♦ — of 

I.    Nmnmiditic  Ximcttone. 

s.    Coane  ludttoiie. 

$.    Congloiiierate  at  the  bottom  of  the  TaUey.  ^ 

It  would  seem  therefore  not  improbable  that  the 
coal-beds  at  Keumur  are  also  oi  Lower  Eocene  age. 

n.    im>DLS  ft  vvpEE  Bocim. 


Whilst  the  Numn^ulitic  facies  is  entirely  abs^t 
on  the  northern  slope  of  the  Caucasus  it  is  present 
on  its  southern  slope,  showing  a  littoral  charadter, 
and  in  Armenia  and  A^ia  Minor  the  charaAeristic 
Nummulitic  Limestone  is  extensively  developed. 

In  Northern  Armenia  a  rich  fauna  has  oeen  ob- 
tained from  the  Akhalzik  basin.  Here — according  to 
Foumiert  the  formation  consists  of  argillaceous 
sandstones  with  intercalary  calcareous  b^s,  from 
which  Simonovitch  and  Sorokin^  coUeded  the  fol- 
lowing fossils,  chieflv  in  the  Postkoff  valley,  at  Suflis, 
and  /Uchalkalaki,  above  a  basement-bed  with  (ktrea 
bellovatifuij  Lam.: —    . 

Nummulites  Icevigatus^  Brag. 

Lima ;  Meretrix  incrassata,  Dedu,  Af .  nitidula,Sow. 

Ampullina  sigaretina^  Sow.;   Diastatna  GrateUmpi^ 

d'Orb.;  Turritella  edita^  Sow.;  Rostellaria;  Cassis 

striataf  Sow.;    Vqluta. 

Abich§  had  previously  investigated  the  fauna  of 

^  ViBfyagg  m  Turfmu  ^c.^  it.  Ghl^^  73. 

t  Descripiiam  gt^l.  dm  Cattaue  €miral^  168. 

X  G^l(iy  of  iki  TtrHaiy  AkhaUik  Basin,  Mat  Ged.  Cane  Tlilis, 
1S86.    [In  Ruatian]. 

I  VtrgteUhinii  ge^.  Gnmdti^  «fr.,  143 ;  and  hit  Gml.  J^^nck.  in 
den  kemk:  Ldndam^  vL  984-391. 


390  Geology  of  Armenia 

the  Akhalzik  basin.  Whilst  Oppenheim  ^  has  pointed 
out  its  similarity  to  that  of  the  Priabona  beds,  the 
presence  of  Nummulites  lomgatus^  Brag.,  and  of  many 
species  of  the  "Calcaire  gross ier''  and  Kressenberg 
beds  render  it  probable  that  the  Lutetian  stage  is 
i^presented  here  as  well  as  the  later  Bartonian  (Pria- 
bona horizon).  It  passes  upwards  into  Qligocene. 
Abich  coUeded  his  specimens  not  only  from  Akhalzik, 
hut  also  from  the  neighbouring  villages  of-Ab  and 
Grel.  Oppenheim  (l.c.)  was  able  to  examine  Abich's 
specimens,  some  of  which  he  corredled ;  the  species  in 
the  following  list  of  Abich's  fossils,  about  which 
he  expressed  a  doubt,  are  marked  [?] : — 

Nummulites  Icevigatus,  Brug.  (in  the  basement- 
bed),  N.  planulatus,  d'Orb.  (between  Akhalkala- 
ki  and  Akhalzik) ;    Operculina  canalifera^  d'Aidi. 

Turbinolia  subelliptica^  Abich  (cast  in  a  calcareo- 
siliceous  tu£F  close  to  Akhalzik). 

Serpula  keptagona,  Abich. 

Ostrea  gigantea^  Brander,  0.  latissima^  Deth.,  &  O, 
bellavacina.  Lam.  var.  (all  in  the  basement-bed) ; 
Crassatellites  tumida^  Lam.  (forming  beds  near 
the  bridge  between  Akhalzik  and  Suflis),  C.  cf. 
sulcata^  Braadcr  (with  the  last);  Meretrist  in- 
crassata^  Dtsh.,  (in  great  numbers  at  Ab,  ih 
calcareous  concretionary  beds),  M.  Lamarcki^ 
Desh.;  Panopea  cf.  intermedia,  Sow.;  Ludna 
(Miltha)  Argus,  {?],  MelL  (in  great  numbers 
at  Ab),  L.  suhargus,  Abich  (Ab). 

Capulus  chamoides,  Abich;  Ampullina  sigaretina^ 
Sow.;  Diastoma  Grateloupi,  d'Orb.  (Melania  cos- 
tellata.  Lam.*  of  Abich ;  at  Ab),  D.  bicinAum^ 
Abich,  (at  Ab,  with  the  last;  not  frequent); 
Turritdla  edita,  Sow.  (often  at  Ab  and  Giel)» 

*  Du  PfUbma-'sMMm.    Fabeontographica,  ilni.  30$, 


Middle  &  Upper  Eocene  391 

T.  imbricataria  [?],  Lam.  (Ab),  T.  Renevieri,  d'Arch., 
T.  cariniferay  Dcsh.  (frequent  and  in  different 
sizes);  Rostellaria  tricarinata^  Abich  (at  Ab; 
not  frequent) ;  Cassis  striata,  Sow.  (at  Ab ;  not 
very  frequent) ;  Valuta  suturalis,  Nyst.  (at  Ab ; 
fairly  frequent). 
Nautilus  sp.  indet.  (145  mm.  in  diameter). 

Bayem  also  colledled  numerous  fossils  from  the 
Eocene  beds  of  the  Akhalzik  basin.  They  were 
identified  by  Abich,  and  are  now  in  the  Tiflis 
Museum.  From  "  a  quarry  at  the  Cossack  post  near 
AMialzik  '*  Bayem  ♦  coUeAed  the  following  forms : — 

Orbitolites  discus^  Riitim. 

Ostrea;  Lima;  Pinna;  Modiola;  Cardita;   Car- 

dium;    Arctica    transversa,    d'Arch.;     Meretrix 

incrassata,  DtaLf    M.  cbliqua,  Desh.;    Carbula; 

Panopea intermedia^  Sow.;  teredo;  Lucina;  Cla- 

vagella. 
Trockus;     Xenophora    confusa,    Dcsh.;    Ampullina 

sigaretina,  Sov.,  A.  epighttina,  Lam.;    Turritella 

imbricataria  [?],  Lam.,  T.  incerta,  Oesh.;  Cerithium 

interruptum^  Lam.;    Aporrhais  Sowerbyi,  Mant; 

Fusus    adculatus,    Lam.;    Pyrula    nexilis,   Sol.; 

Voluta;    Mitra;    Conus;    Fleurotoma  Voyseyi, 

d'Aich.,  P.  angulosa^  Desh. 
Adtceon  alligata^  Desh. 
Dentalium. 
Nautilus  lingulatus,  r.  Bach  (a  Kressenberg  form) ; 

Belemnitella. 
On  the  way  between  Akhalzik  and  Suflis  Bayem  t 
found  the  following  fossils  of  the  same  horizon : — 
Orbitolites ;  Nummulites  planulatus^  d'Oth.;    Oper- 

culina  canalifera,  d'Aicfa. 

*  CM.  Caial.  TSJKs  Ifutnm^  Si-^S* 
t  Ibid.  55-66. 


399  Geology  of  Armenia 

Serpula  kettagona,  Abidi. 

Ostrea  heltovacina^  Lam.;  Spondylus  rarispina^ 
Pinna;  Modiola;  Area;  Limopsis  granulata. 
Lam.;  Cardita ;  Venericardia ;  Crassatellites  stU- 
cata,  Bnnder,  C.  tumida,  Laiil,  C.  sinuasd^  Desli.; 
Cardium;  Meretrix  (Tivelina)  deltoidea^  Lam.; 
Corbula  striata^  Lam.,  C.  faba^  Desh.;  Panopea 
intermedia^ Sow.;  Teredo;  Ludna (Milt ha) pseud- 
argus^  d'Arch.,  L.  (Dentiludna)  concentrita^  Imm^l 
Pholadomya ;  Clavagella. 

Emarginula ;  Calyptroea  laevigata^  Desh.,  C.  lamell- 
osa^  Desh.;  Trochus ;  Turbo ;  Xenophora  confusa^ 
Desh.;  Ampullina  patula.  Lam.;  Scala  [Crtspo^ 
scalaj  crispay  Lam.;  Turritella  imbricatarta  [?}, 
Lam.;  Certthium;  Rostellaria  rimosa^  Sow.,  !?• 
cras5t7a6ruifi,  Desh.;  Rimella  fissurella^  Linn.; 
Terebellum ;  Streptochetus  intortus^  Lam.;  Volnta 
(Eopsephia)  torulosa,  Desh.;  FoZiitf'ZitAes  mutate^ 
Desh.;  Hemiconus  scabriculus^  Sol.;  Flettrotoma 
(SurciiZa)  catenata^  Lam. 

i4(^ae(m  in  flat  a^  Per.;  j5tt//a  Ugnaria^  Lam. 

Nautilus  lingulatuSj  v.  Buch  (a  Kressenberg  form). 

The  same  fauna  was  found  by  Bayem*  on  the 
Ariiche  plateau,  4  miles  west  of  Akhalzik,  beyood 
Suilis.  He  gathered  the  following  fossils  in  a  '*  cal- 
careous, sandy  conglomerate  ": — 

Miliola. 

Area  biangula^  Lam.;  Limopsis  granulata^  Lam.; 
Plagiocardium  granulosum^  Lam.;  Solenocurtus 
strigilatuSy  Linn.;  Corbula  striata^  Lam.;  Panopea; 
Teredo;  Lucina  (Dentilucina)  cohcentricay  Lam.* 
L.    (D.)    saxorUm,    Lam.,     L.    antiquata.    Sow.; 

*  (%^/.  Catal.  lyUs  Jtfusium,  56-66. 


Middle  &  Upper  Eocene  393 

Trockus;  Ampullina  stgaretina^  Lam.;  Diastama 
Grateloupi,  d'Orb.  (Melania  costellata.  Lam.  of 
Abich) ;  Littorina ;  Turritella  tmbricataria  [?], 
Lam«,  T.  incerta^  Desh.;  Rimella  fssureUa^  Lmn.; 
Buccinum;  Fusus;  Mitra;  Valuta;  Canm; 
Pleufotoma  coUm,  Sow. 

AdicBan  inflata^  Fer.;  Atys  wulala.  Lam. 

Dentalium. 

In  the  neighbouring  ravine  of  Arikhe  Bayem^ 
found  a  similar  assemblage  of  fossils : — 
Tufbinolia. 

Periaster  obesus^  Desor. 
Ostrea  mtdticostata^  Desh.»  0.  (jEtastreon)  UUissima, 

DtA.;   Anomia;  Pecten;  Mytilus;  Limopsi^  an- 

gusticastata^  Lam.;  Nucula ;   Plagiocardium  gra- 

nulosum^  Lam.;  Carbula  tn^anaUs^  Sow.;  PaHopea 

intermedia^  Sow.;  Teredo;  Lucina. 
Patella ;   Calyptraea  lamellosa^  Dcsh. ;    Xenophora 

agglutinans^  Lam.;  Nerita ;  Natica  ;  Turritella ; 

Aporrhais  Sowerbyi^  Mant;    Cassts;    /2afie/Za; 

Buccinum ;  Fmus  ;  Pyrula  nexilis^  Sol.;  Cofi«s. 
Bulla  lignaria^  Lam. 

At  Khero  (Okhero).  a  few  miles  east  of  Akhalzik, 
the  same  stage  is  repres^ted  by  *' sandstone  and 
shell-conglomerate/*    in    which    bayemt    found 

Ostrea ;    Pecten. 

Scala  decussata^  Lam.;  Fusus. 

At  Yenikow  (to  the  S.E.,  but  still  W.  of  the  Kur) 
Bayem's  fossils  ^  indicate  a  lower  horizon  pTpresian). 
The  fossils  occurred  in  a  '*  calcareous  t^ummulitic 
sandstone/'  and  comprise  ^  ^ 

Nummulites  planulatus^  cPOrb. 

•  CM.  Ckiet.  ItJttiMwsnm^  60-66. 

t  Mi.  66|  }6.  %  Mi.  64,65. 


394  Geology  of  Armema 

Ostrea ;  Pecton  comeuo^  Sov.;  Area ;  Cardium. 
Nerita  conoidea^  de  Kmqr  (a  Cuise  fonn) ;  Natica ; 
Cassis  \  Fusus;  Voluta;  Conus. 

On  the  N.  side  of  the  Akhalzik  basin,  at  Kaiataban, 
Hayem  ^  found  the  following  fossils  of  this  stage  in 
*'  sandstone  and  shell-conglomerate*': — 

Operculina  canaltfera^  d'Aidu 

Sorpula. 

Ostrea  cariosa^  Desh.,  0.  (Pycnodonta)  DejroMcd^ 
Deth.;  Pecten ;  Pinna ;  Area ;  Linu>psis  /rans- 
latOf  Lam.;  Nucula  margaritacea^  Lan.;  Plagkh 
cardium  granulosum^  Lam.;  Corbula  comfiamota^ 
Dtfh. ;  C.  rugosa^  Lam. 

Emarginula  costata^  Lam.;  Calyptrcea  tamMosa, 
Desh.;  Neritina  DeskiisUllii^  Deth.;  Ltttofifia; 
Vermetfis;  Bayania  la£tea^  Lam.;  Melamopsis 
ancillaroides^  Dtsh.;  Tntoninm;  Fnsns;  Plevfv* 
toma. 

Nummulitic  limestone  and  sandstone  are  recorded 
by  Bayem  f  at  Shokka,  a  few  miles  S.W.  of  Abas* 
tu^nan,  and  at  Abastuman  itself  he. found  a  **  crystal- 
Une  limestone'*  containing  the  following  fossils: — 

Cardita;  Cardium;  Ludna. 
Neritina ;  Pyrula  nexilis^  Sol. 

In  all  the  above  mentioned  localities  of  the  Akhal- 
zik basin  these  fossiliferous  Eocene  beds  are  overlam 
by  the  Flvsch  facies  of  the  Upper  Eocene,  and  thism 
turn  by  Oligocene  (^.t;.).  According  to  Simonovitch  X 
and  Foumier§  this  facies  is  represented  by  majriy 

*  Geol.  Catal.  Tf/Us  Musum^  66^  67. 

t  iM&  «S. 

t  G^fiU^  ^  iki  Taiiafy  AUabik  JSasMt  Mat  GeoL  Cauc  THEs, 
1886.    [In  Rintian]. 

I  Descr^iitm  ght:  dm  Omeau  cmtral^  z68. 


Middle  &  Upper  Eocene  395 

cfiilcareous  beds,  much  laminated,  with  fucoidal 
z^emains,  fish-scales  and  (at  Akhalzik)  impressions 
jaokd  teeth  of  the  following  fishes : — 

Lamna  elegans^  Ag.  (teeth) ;  Otodus  (teeth). 

MeUtta  sardiniteSy  Ag.;  Zeus  colchicus^  Simo&ov. 

These  laminated  limestones  with  fish-remains  occur 
throughout  the  Caucasus  at  the  top  of  the  Eocene. 

The  Middle  and  Upper  Eocene  of  the  Lower  Rion 
and  Kvirila  basin  show  the  same  fades  as  that  of  the 
Akhalzik  ( basin.  The  Nummulitic  facies  is  indeed 
rather  more  marked  i;  here  it  consists — ^according  to 
Foumier*^ — of  clays,  fissile  marls,  and  fossiliferous 
calcareous  beds  wi& 

Ofbitoliies  discus,  Rudm.,  0.  Fortisii,  4'Arch.«  0. 
sella^  d'Arch.|  0.  papyracea,  d'Arch.,  0.  discus, 
Hiitiiii.,  0.  patellar  is,  RutiniM  0.  parmula,  Rudm.. 
0.  stellata,  d'Arch.;  Cristellaria ;  Nummulites 
intermedius^  d'Arch.,  N.  biarritzensis,  d'Arch., 
N.  Murchisoni,  Bmnner,  N.  Ramondi^  Defr. 
Sckizaster  cf.  riniosus,  Desor. 
Serpula. 

Cancer  Desmaresti,  Mumt.  (a  Kressenberg  form). 
TerAratula. 

S^and^lus  subspinosus,  d'Arch.;  Cardium ;  Cyprina. 
Nautilus. 
These  fossils   indicate   the  presence   of  both  the 
Kressenberg  and  Priabona  horizons. 

The  highest  beds  of  the  Eocene  in  this  reig^ion  show 
the  following  succession  according  to  Foumier  (I.e.)  : 

I.    Maagancet-beds  in  white,  arenaceous  limestones,  frequently 
•  wick  iron  pgrntes,  and  teeth  of  Lwnna  and  Otodus. 

%*    AigillaoeoHMrenaoeous  slates,   slatjr  clays,  and  very  fissile 
marls  witk  uie  following  Teleostean  fishes  :— 

Ly¥9iipU  mwasiois,  Romanowski ;   Osmemdes  tewisieHtis, 
Ag.l?]i  Mgletiasard$miu,Ag.;   Ciadotycizs 
' .'  .  Geinits;  Zeus  coUhitus,  Simonfovitcfa. 

^  DeUf^MmflM.  dm  Caucau  aninUf  i8s. 


396  GmUgf  •/  Armemia 

On  the  other  handt  in  the  Kur  basin  (above  TilEid 
the  Middle  Eocene  is  altogether  wantk^  on  the  ML 
and  W.  borders^  although  toe  whole  series  is  complete 
and  concordant  in  the  Thrialetic  Range  on  the  SL 
Thus — acoQrding  to  Foumier  ^ — ^tbere  is  a  veqr  sliaip 
discordance  on  the  S.E.  side  of  the  Zimla  dome  (the 
Mfschic  MountainsX  to  S.E.  of  the  Suram  tunnd. 

On  the  N.  side  of  the  Thrialetic  Mountains^  and 
to  the  &  of  Gori  and  Khidistav,  Foumier  t  fom^  the 
following  fossils  in  yellow,  argillaceo<alcareous 
stones : — 

Amfki$t0gina  nummulincefomiis^  Abkh;   Nt 
UUs  km^atMf  Br^.,  iV.  intermedins^  d'Aich. 

But  in  the  rest  of  thiitKur  basin  both  Middle  and 

Upper  Eocene  are  absent  except  in  the  distridk  W. 

of  Tiflis.    Near  Manglis-^according  to  Tsululddse  ^-*- 

the  Middle  Eocene  appears  in  the  form  of  Uiidk  beds 

of  sandstone.    At  Eiisabethinskdi  black  argillaceous 

slates   occur   of    Upper    Eocene    date,    containing 

DtMtoma  €o%tAlata^  Lus.  (probably  D.  Gratelanfi^ 

d'Orb.i  cp.  p.  390);   Pyrula  condita^  BnN«s.  \J\ 

a  Helvetian  form]. 

The  same  horizon  occurs  at  Bogvi  and  Assureti 
in  the  Ganzileef  Moui^tains.  In  the  Assureti  yaUey 
AbichS  found  a  very  fossiliferous^  brownish  gtey» 
sandy  limestone,  underlying  Oligocene  c6nglomerates« 
and  containing  the  following  fossils : — 

Sp(mdylns  subs^nosns^  d'Ar^.;    Limopsis  angnsttr 

oostato,  Lam.;  Cardinm;  Venns;  Lmcina. 
Trochns ;    Natica ;    Diastama  Gratelwpi^  d'Ofb.  (==? 

*  Daerf^Hm it$L  in  CtmemtnOrmt^  171. 
t  nu.  175. 

t  Xmenktt  At  ikt  DiUruti  ^  ike  Algti  6*  Kkmm  Riun^  Hat 
G«ol.  Csttc.  aer.  a,  L  9.    (In  lUntiaii]. 


Middle  &  Upper  Eocene  397 

Melania  costellata^  Lam.  of  Abich) ;    Turritella  ; 
Pyrula  condita^  Brongn.  [?]  ;  Volutd, 

In  the  Alget  region  Abich*  divides  the  Middle 
Eocene  into 

u  Calcareous  sandstones,  passing  up  into  thick  beds  of  lime- 
ftone-fla^  separated  from  each  other  by  argillaceous  shales. 

2.  Yellow  sandstones,  occasionally  g&uco'nitic,  either  thick  and 
compact  with  sparing  JVammuiiL's  planu/atus,  d'Orb.,  and  M  Ramandi^ 
Defir.,  or  else  shaly  with  abundant  fringed  worm-tracks  and  casts 
of  thick*8tiaked  fucoids  (CkoHdritis). 

3.  light  grey,  argillaceous  shales. 

An  Upper  Eocene  age  has  been  attributed  by 
Tsulukidse  f  to  the  sandstones  of  Aklioti  with  Serpula 
and  SpondyluSf  and  also  to  the  hard  coral-limestoiies 
with  Astrcea  of  the  Alget  river. 

Just  as  in  the  Akhalzik  and  Rion  basins  (p.  395) 
the  highest  beds  of  the  Upper  Eocene  of  the  Kur 
basin  consist — ^according  to  Fournier  J— of  argillaceo- 
arenaceous  shales  with  numerous  fucoids.  Calcareous 
fissile  beds,  with  Chondrites  and  indeterminable  fish- 
impressions,  are  intercalated  in  the  series.  ^  Abich, § 
however,  found  in  them  occasional  small  bfummulites 
[not  specified]  and  only  seldom  a  well-preserved 
Meletia  sardinites^  Ag.;  th6  series  is  charadlerized  by 
the  contemporary  eruption  of  basic  hornblende- 
and  augite-andesites,  followed  by  quartz-trachytes 
and  true  trachytes. 

Eocene  strata  seem  to  be  entirely  absent  in  the 
Somketian  Mountains,  and  along  the  outer  zone  of 
the  Eastern  Border- Ranges.  On  the  inner  (western) 
side,  however,  from  the  Pambak  Mountains  along  the 

*  JfitfA.  aus  dem  Xauk,^  Verh.  k.-k.  geoL  Retchsanst,  1877,  33. 

t  Xiuarcka  in  the  Dhirkts  rf  iki  Aigei  6^  Khram  Jthetf^  Mat. 
Geol.  Cauc.  ser  s,  i.  9.     [In  Russian]. 

i  JkuripHoH  gtcl,  iu  Cauiou  anifui^  178. 

{  Op.  di.  33, 


398  Geology  of  Armenia 

N.E.  coast  of  Lake  Gokcha  down  to  the  valley  of 
the  Upper  Tester,  and  also  along  the  W..  border 
of  the  West  Rarabagfa  to  Ordubad,  and  across  the 
Araxes  into  Persia,  Eocene  beds  form  a  wide  belt, 
sometimes  partially  eoocealed  bv  lavas. 

in  the  Pambak  Mountains  Abich  ^  found  the  lo^er 
beds  of  the  Middle  Eocene,  overlying  Senonian  (p. 
375).    They  consist  .of  siliceous  limestones,  containing 

Orbitoliie^;     Nummulites   planulatus^    dXMK,    N. 
Ramondi^  Defr.»  N.  granulosus.  d'Aidu;  OperciiHna. 

These  Nummulitic  beds  ^appear  not  only  in  the 
upper  valley  of  ^^  Pambajk  river,  but  Aibicht  also 
found  them  in  the  upper  regions  of  the  Akstafa 
and  Zanga  rivers.  The  same  horizon  occurs  in 
the  Duchu  and  iTertegr  valle}^,  and  the  two  areas 
were  evidently  in  complete  continuity  through  the 
N.W.-S.E.  Eocene  zone,  which — according  to  Abich*s 
map— borders  Lake  Gokcha  on  the  N.  and  E.  On 
the  left  side  of  the  Duchu  valley  Abich  (Lc.)  found 
that  **  Nummulitic  limestone  of  considerable  thickness 
forms  the  summit  of  the  mountain-ridge  (7500  feet)/' 

In  passing  from  the  Du<5hu  valley  to  Bashtiibel 
Abich  X  found  Nummulitic  limestone  forming  a  high 
cliff,  about  halfway  to  Davachukhur  (the  first  village 
in  the  Bashtubel  valley).    According  to  his  account 

"  this  cliff  is  a  disconnected  fiagment  of  a  once  conttnnous  fonaatioii 
of  Eocene  limestone,  orerlymg  clastic  feUtic  strata  [Senonisn].- 
It  dips  slightly  to  N.K,  wiA  it.W.— S.E.  strike.  The  senes  is  30-40 
metres  thick,  and  consists  of  nmneions  thin  beds  of  fiejr  limestone. 
Its  rough  snrface  reveals  mainr  sumII,  yeOow-brown  or  was-ydloir 
inclusions,  irregular  in  form  and  with  an  eroded  appearance,  togetfier 
numerous,  smooth,  roimded  indnsions,  TarTsag  in  siao  fton 


^  GmI  Fm€k.  im  ien  katk.  Limdem.  ii.  36. 
t  <7m/.  Bsok.  wm.  Kur  6^  ArmxUy  145. 
t  INd.  137-14S. 


Middle  &  Upper  Eocene  399 

%  bean  to  a  pigoon's  eg^;  a  good  doal.of  iroD  pyrites  ooQuti  in 
the  interior  of  tncie  indnaons.  Closer  examination  of  the  weathered 
•orfsoe  disclosed  nnmarous  spathic  tests  of  Numm^Uus  standing 
out  in  relief,  from  the  sixe  of  a  grain  of  millet  up  to  io-i5  mm. 
in  diameter*  When  fractured  the  rock  is  a  dark  smoke-grey,  with 
lighter  m^ottUng,  presenting  somewhat  the  appearaiice  of  A  brecciated 
limestone.  SmaD,  white  spathic  porfiona  resemble  the  stem- joints 
of  Csfamds.'' 

In  this  limestone  Abich  *  found  numerous  fragments 
of  shells  and  the  following  Foraminifera : — 

Nummulites  planulatuSf  d'Orb.,  N.  Ramondi^  Defr., 
N.  granumus.  d'Arch.,  iV.  cf.  scaber^  Lam.;  Oper- 
cuUna  cf.  Baissyi^  d'Aich.,  0.  atnmonea^  Leym. 

Abich  (l.c.)  continues  to  state^'that  these  limestones 

^*  at  a  few  versts'  distance  from  Davachnkhur  overlie  crystalline  t^da 
and  rise  up  to  a  b^dit  of  900—1000  feet  above  the  lii^,  which  cuts 
r  Arovi^  them.  •  T1ie]|r  show  many  sifpis  of  reddenins  and  meta- 
moffpnism  owing  to  the  intrusion  of  felsite-porphyiy  and  serpentine. 
The  rode  has  becouMT  altered  into  an  amorphous,  yellowish,  porous 
limestone  with  dendritic  manganese." 

The  ^est  seAion  of  the  Nummulitic  beds  in  the 
mid-Anizes  r^on  is  near  the  mona,stery  of  Karmir- 
vank|  at  the  entry  of  the  gorge  leading  to  Julfa. 
Abich  t  described  this  sedUon  in  1839,  but  recently 
Fredit  re-examined  it  Whilst  the  two  descriptions 
are  somewhat  difficult  to  collate  in  detail  yet  they 
show  a  broad  general  ag[reement,  and  I  have  therefore 
ananged  them  side  by  side. 

The  uppermost  red  sandstones  are  overlain  dis* 
cordiamtly  by  Sarmatian  saliferous  marls. 

*  G^.  Biot.  wm^  JBur  6*  Arexa^  139. 

t  Des  SUmuate  dv.,  M^m.  Acad.  Sd.  St  Pfteisb.,  ntt.  6,  Sd. 
madL  ft  phys.  yi.  6$. 

t  Hes  PeUm^Uum  im  MMMrmmm  6*  PenUn^  Beilr.  Pd.  Obtr.- 


400  Gmtlogy  of  A 


THE  KARMIRVANK  SECTION  ACOOftDING  TO  FRSCH. 

X.  200  metres.  Red,  ooii-cftlcarBous  sandstopes, 
ThesCi  togetiier  with  IX  (a),  may  possibly  repnaeut 
the  Upper  Eocene. 

IX.    Conglomerate  and  saAdsiooe : — 

(a)  20  m.  Cpnfflomentte,  dipping  gently  to 
the  north.  Karmiryank  monastery  lies  on  this  bed. 
Pebbles  chiefly  of  Palaeozoic  dolomites^  Devonian 
limestones  (with  Emmphalus  Waklenbergi^  Orthoceras\ 
schists  and  various  eruptive  rocks.  A  bed  ct  tia* 
vertine  (j-i^  metres  thick)  is  interbedded  in  tfiis 
conglomerate. 

(b)  3  m.    Fine-grained,  calcareous  sandstone. 

(c)  8  m.    G>nglomerate. 

VIII.    Upper  Limestone  CongIom»ates : — 
(a)  6-7  m.    Limestone  with  large  oysters ;  dip 
17^  N.,  strike  N.  85^  W. 


(b)  4  m.    Conglomerate. 
II. 


VII.    Lower  Limestone  Conglomerates : — 
(a)  a    m.    Crinoid    Limestone    with    Echinid 
spines. 


(b)  ^  m.  Conglomerate.  Pebbles  chiefly  of 
Devonian  hmestones  with  Euompholu^  Wahlenbergi.^ 

VI.  [Thickness  not  stated].  Thickly  bedded, 
fine-grained  sandstone,  rarely  witfi  large  individuals 
of  Nautilus. 

V.(a)  5-20  m'.  Conglomerate  of  varying  thick- 
ness. 

(b)  2  m.  Rhixocorallium  Sandstone,  reddish 
and  compadt,  with  branching,  cylindrical  Rkim* 
cortUlia, 

IV.    4  m.    Sandstone,  light  brown,  no  fossib. 


Middle  &  Upper  Eocene  401 

THB  KASMIRVANX  SECTION  ACCORDING  TO  ABICH. 

L.  Fine-grained  sandstone*  pasdng  downwards 
through  beds  only  slightly  calcareous  into 

K.  Calcareous^  congloiberatic  sandstone  with 
iM>  fossila 

!•    Red   conglomerates,    many   fathom^   thick. 

Pebbles  often  large,  but  always  rounded,  consisting 
.  of  Palaeozoic  dolomites  and  limestones  with  Devonian 
foesilsy  and  also  of  yellowish  sandstones  and  marls, 
and  of  felsite*  and  quartz-porphyries. 


H.  A  considerable  series  of  yellowish,  very 
compaift  limestones,  marmoraoeous  in  the  upper'part, 
with  the  same  fossils  as  in  D.  NummuUtee  Rammdi^ 
Dcfr«»  occurs  only  sparingly. 

G.  Yellow,  crystalune  limestone,  almost  entirely 
ipomposed  of  Nummulites  Ramondi^  ha. 

P.  faV Yellowish  dolomite. 

(b)  Thick,   compact,   yellowish    grey%    hemi- 
crystallme  limestone,  with  Nummulites  and  Oysters. 

£.    Conglomerate ;  an  extremely  compadt  Nagel- 

fiuh^  consisting  of  pebbles  of  Palaeozoic  limest<»es, 
several  inches  in  diameter,  in  a  pale  red,  dolomitic 
groundmass.    It  contains  much  quartz  sand  and  flint. 


•  D 


4011  Geology  of  Armmia 

THB  XARMIKVANK  SBCTION  ilCCORDIIfG  TO  FRECH  (oMtlQ. 

IIL    Nummulitic  Limestone : — 

(a)  5  UL    PuR^  light  reddish  limestones  with 

massive  Nummulites ;  the  Oysters  and  Corals  of  (b) 

still  persist.    The  fossils  oomprise  the  following : — 

Nummulites  Luca$a$ms,  Defr.  (the  most  frequent), 
N.  pmrf^i^us^  ifOA.;  OfldPna$s  pdfyracea^  Bovb6ei 

Tfockocyathus ;  Pattaldph^llia  submfiata.  ifAdUudL 
Rhabdophyllia  of.  craastramosa^  Tdbu 

Ostrea  ranlameUa^  m^»  0.  cyathula^  Lmb.  (01igoc>; 
Gtyphcea  Braugmanti^  Dtdi.,  Sp(mdylus  cf.  rm- 
sfnna,  Dedu»  S.  cf.  multisiriatus^  Dtth. 

PUurotamaria  ef.  Karrin-KwienMis^  dTAich. 
Lithatkamnum. 


(b)  5   m.    Marly   limestones   with  numerous 

S'gantic  oystersi  long  sea-urdiin  mines  [cp.  Abich's 
' ,  and  corals.    Nuqumulites  scaicely^  present 
Trom   this   horizon   Tsulukldse*   found 
Stylina  Perexiana^  Mich. 
Ostrea  Umgiroitris^  Lam.,  0.  s^orocmsfs,  Defr.]. 
II.    20  m.    Sandstones,  light  brown  and  reddish. 
In  thick  beds,  alternating  in  tte  knrerpart  with  dark 
brown,  fine-nained  conglomemtes.    Large  botilders 
and  amygdaloidal  fragments  occur  along  the  plane 
gf  contact  With  I. 

I.    7  m.    Green»  much  weathered  diabasic  rock 

emitted  at  a  period  antecedent  to  U. 


W^W«    ^fW^Bwi   ^CHMv  JBWp^V^W,    XwSi 


Middle  &  Upper  Eocene  403 

THB  KARMIRVANK  SECTION  ACCORDING  TO  ABICH  {cofltd). 

D.  Limestone,  brownish,  impure,  very  fine- 
grai^d  Upper  beds  more  compad  and  yellow,  with 
numerous  remains  of  Cidaris  acicularis^  d'Aich.  Corals, 
e.g.,  Trochocyathus  multistriatuSf  m?B.  ft  H.»  and  Stylo- 
phora  tkyrsiformis,  M-E.  [?;  a  Crosara  species],  chiefly 
occur  in  the  middle  beds.  The  whole  thick  series 
contains  numerous  fossils,  as  follows : — 

Nummulites  intermedins^  d* Aich.  [?]. 

Ostrea    latissima^    Desh.   (very    large);     Pecten; 

Peetuncnlns. 

[The  following  fossils  were  coUeded  by  Tsulukidse  * 

evidently  from  this  bed,  *^a  compad  limestone,  lying 

between  sandstone  and  a  limestone  with  Ostrea^  3 

versts  S.E.  of  Karmirvank  monastery  '^-7- 

Cidaris  striatogranosa^  d*AiA.;  Linthia  biarritMensis. 
Cptt.  ^ 

Spondylns  rarispina^  DtdL,    S.  grannloeius^  Dctb.; 
Cytherea  suberycinoides^  Dcdi.] 

C.  Reddish  brown  a.renaceous  limestone,  passing 
down  into  the  underlying  congiomemte.  Abundant 
fossils,  but  only  casts  (Natica).  Distindl  traces  ci 
Nummulites  and  Operculina. 


B.  Sandy  conglomerate,  reddish  brown,  very 
fine-grained.  Pebbles  from  size  of  millet  to  a  pea,  of 
qiiiartz  grains.  Palaeozoic  limestones,  decomposed 
felspar-rocks,  felsites,  &c. 


A.  Very  fine-grained,  clastic^  hemicryst^lline 
rock.  The  violet-brown,  pelitic  groundmass,  with 
silky  lustre,  encloses  angular  fragments  of  a  much 
altered  igneous  rock. 

*  GmI.  Caial.  ZyUt  Mmsemm^  101. 


4^4  (Mogy  «/  Annitdm 

At  Darosham  in  the  Araxes  fforgep  a  few  versts 
abovi  jul&,  aaother  fine  cliff-section  of  the  Eocene 
beds  is  exposed,  and — according  to  Abich^ — the 
succession  is  as  follows :— . 

I.     Red  sanditonei^wtth  tradiTtic  fragments  (X  of  Kjimunrank). 

a.  Blood-red  conglomenite  of  Najpljimk  chancter.  Pebbles 
partly  of  Pabeoioic  Umetconett  ptitly  of  tracfaytic  rock  (IX  of 
( iCarminrmiik). 

3.  Red»  calctrcotts  tandiriwif.  No  Nummulttes  in  the  vpper 
l^eds ;  the jr  gradually  increase  in  i^amber  downwuids  (IV — ^VIll  of 
Karmirvaak). 

4.  CoarNi  Naples-yeUow  sandstone  of  consideiaMe  thickness^ 
with  intereal^  beds  of  sandy  limestone  (III  of  Kanmrvank).  Tbe 
limestone  contains  the  fdlowing  foesQs : — 

Boiibee,  N.  Jlmm^ndi,  Defr.»  M  flmmiiHus,  dt>rb.,  N.  om^ 

phmUus^  Lam.p  N.  gi^hUui^  Leym* 
SehiMMShr  jMifimsis^  Dubois. 
Osim^  gti^Miea^  Braider. 
CiHOtium{Campiunii)figMkmm^ljmk 


%.  Coiiglomcrate»  with  fragments  of  the  underlying  rockt  nd 
pebbles  (nearly  as  laige  as  beans)  of  dialoedonyi  quartsite,  felsite- 
V^^^yVf*  quifts-porphyxy,  diabasci  quaitsoee  sandstone,  bitununom 
dotomitOt  &:c  (II  of  Karmirrank). 

6.    Pisnno-Triaasic  and  Bateoaoic  stnta,  aearfy  vertical. 

At  this  locality  and  eviddatly  from  the  sanie 
fossiliferous  iox^t  (^  yellow,  sandy  marls  underlying 
red  conglomen^te  )\  fsulukidse  t  found  the  following 
fossils : — 

NummHlttes  Lucasanus,  Defr.  N.  Immgatun^  Bn«., 
:V.  sM(Bmgatus^  d'AichM  AT.  Itamcnii^  Defr., 
H,  distans^  Desh. 
,  Comdypeus  conoideus^  GoUf.  fKressenberg) ;  ScAtV 
ostfr  Studeri,  Ag.»  S.  julfenais,  Dubois;  £11^- 
tagus  xmuUus^  Defr.  (Biarhtz). 

*  Dmi  JMnsift  6v.»  Mem.  Acad.  ScL  St  Pftefsb.»  s6r.  6»  ScL 
math,  k  phys.  viL  6y ;  and  his  <7m/.  Urui.  m  dm  Aimk.  Lsidirm^ 
VL  So»  8c. 

t  GhI.  QM.  lySs  Mutmrn^  loe-tb]. 


Middle  &  Upper  Eocene  405 

Gryphaea  Brongniarti^  Deth.;  Spondylus  rarispina^ 
Dcik.;  Vulsella  fatcata^  Mfon.  (Kresseoberg), 
V.  exogyra^  d'Aidi.;  Modkla  (Brackidontes)  p^- 
inata^  Lam.;  Area  (Barbatia)  barbatula^  Lam.; 
Cardium;  Ckama  cf.  gigae,  Deah.;  Cdpsa; 
Teredo  Toumali^  Leym.;  Ludna  [DentilucimaJ 
eoncentrica^  Lam.;  Acropagea  subrotunda,  Desh. 

Solarium;  Phasianella  Otienif  d'Aidi.;  Nerita  con- 
oidea^  dtSuAuj;  NaUca  angUlffera,  d'Orb.,  N. 
(Cepatia)  cepooM^  Lam.;  Ampullina  patula^  Lam., 
A.  sigaretina^  Soir.;  Diostoma  Grateloupi,  crOrb. 
(=:: Melania costellata^ Lam., 9F  Abich) ;  7 urritella 
Deehayesif  ljm.f  T.  cJF.  anguUxta^  Sov.:  Cerithium 
(Campanile)  giganteum^  Lam.;  .  Seraphs  {Tere- 
helium)  convomtum^  Lam.;  Cassis. 

This  assemblage  of  fossils  seems  to  indicate  the 
Kressenberg  facia. 

These  ywow  Nummulitic  beds  form— according  to 
Abich  ^ — the  bed  of  the  Araxes  when  entering  the 
Aji  plain  on  its  way  ,  to  Ordubad.  These  strata 
continue  to  be  overlain  by  red  trachytic  conglomerates 
of  Upper  Eocene  or  01i|focene  age,  which  are  covered 
disj(^rdantly  by  Sarmatian  g)rpseous  marls.  I'owards 
the '  Dary  Dagh  (a  few  miles  N.E.  of  Julfa)  the 
trachytic  conglomerates  increase  in  thickness  and 
contain  more  numerous  fragments  of  ^'trachytic 
milktone-porphyry.*'  The  Dary  Dagh  seems  to  me 
to  be  the  southernmost  member  of  a  S.E.-N.W.  series 
of.  trachytic  laccolith  (Arabkev  Daeh,  Nagashir 
Dagh,  &c.),  which  have  broken  through  the  Eocene 
becb  to  a  greater  or  less  extent  (cp.  Abich's  views 
of  the  Nakhichevan  distridl,  op.  dt.^  Atlas). 

According  to  Abich  (I.c.)  the  succession  of  the  strata 
on  the  Dary  Dagh  is  as  follows : — 

^  Ctm/.  Fcmi.  im  dm  hmk.  LimiirH^  iL  Si. 


4o6  Geology  of  Armenia 

u    Red  and  grey,  sandy  tnft. 

s.    Light  yellow  **  Suinnurgei^*'  r^gukriy  bedded. 

3.  Very  compact  conglomerates,  sermd  fathoms  thick,  with 
pebbles  of  a  homblende^uartz-trachyte  in  a  feUtic  cement  Thqr 
abruptly  overly 

4.  Red  Nagdjluk^  with  pebbles  of  Pialeosoic  limestones  and 
also  of  sanidine-oligodase-trachy te.    (Cp.  IX  of  Karmirrank  section). 

^.  Red  breccia,  very  hard  and  compact,  composed  of  fragments 
of  siliceous  chalk,  chalcedony  and  alunite  m  a  pditic  groundmass. 

6.  A  thick  series  of  red,  calcareous  sandstones,  merging  into 

7.  Yellowuh  Nummulitic  Limestone. 

According  to  Freeh*  Sarmatian  strata  occur  only 
at  the  base  of  the  Dary  Dagh,  in  horizontal  beds 
overstepping  the  tilted  Eocene.  No  trachyte  actually 
appears  to  view  in  the  Dary  Da^h  as  in  the  other 
laccolites  to  the  N.W.,  but  considerable  quantities 
of  realgar,  orpiment  and  alunite  have  been  deposited 
in  vertical  cracks,  which  doubtless  extend  from  a  still 
hidden  intrusive  mass  of  igneous  rock. 

Near  Yaiji,  on  the  Araxes  below  Julfa,  a  conral* 
limestone  of  Eocene  age  crops  out — according  to 
Tsulukidse.!  A  little  to  the  S.W.  of  Ordubad  Radde 
and  Sievers  X  found  Eocene  beds  with  Ottrea^  Nuticaj 
C6rithium  and  Turritella^  and  they  have  recorded 
Nummulites  perfnrdtus^  d'Orb.,  fi^om  Julfi.  In  the 
Araxes  TOrge  of  Ordubad  Abich  §  noticed.  *'  Num- 
mulitic Limestone  raised  up  on  end,  forming  steep 
ridges  of  rock/*  and  from  Ordubad  '*  the  Eocene  (with 
N.W.-S.E.  strike)  passes  over  into  Persia,  gaming 
an  extensive  distribution  on  the  plateau,  overlying 
M^sozoic  and  Palaeozoic  beds.^* 

On  the  other  hand,  towards  the  granitic  chain 

^  Das  PaUatmam  in  Huhormmem  6*  Parnn^  Beitr.  BaL  CEitr.* 
Ung.  xiL  iSo. 

t  Gml.  CaioL  Tifiis  Miumm^  104. 

t  Mi.  178. 

{  Gtol.  PonA.  inimkauk.  Limdem^  it.  17. 


Middle  &  Upper  Eocene  407 

of  the  West  Karabagh  the  Eocene  consists  chiefly 
ol  sandstones  with  impressions  of  plants  and  oc- 
casional  seams  of  lignite.  It  is  probable  ^baX  these 
beds  mark  the  topmost  horizon  of  the  Upper  Eocene, 
-like  the  similar  beds  in  the  Akhalzik,  Rion  and  Kur 
basins  (pp.  395t  397^.  They  clearly  show  a  littoral 
charadler,  and  the  Central  Karabagh  was  evidently 
dry  land  from  the  Turonian  (p.  360)  down  to  the 
Pliocene. 

Tsulukidse  ^  found  that  the  foot-hills  on  the  banks 
of  the  Araxes  at  Ordubad  consist  of  this  series  of 
I)Iant-sandstones  and  marls  with  variegated,  sandy 
limestones ;  and  at  Andamich  (a  little  N.  of  Ordubad) 
he  observed  dark  grey  sandstones  with  plant-im- 
pressions, overlying  marls.  Proceeding  further  N.W., 
isulukidse(Z.c.)colle(5led  plant- impressions  and  teeth 
of  Lamna  eUgans^  Ag.,  at  Gyansa  in  similar  sandstone, 
much  uptilfed,  alternating  with  dark  marls,  and  at 
Anabat  plant-impressions  also  occur  in  dark  gr6y, 
calcareous  marL 

Further  N.,  at  Bilev  (Belev)  Tsulukidse  f  met  with 
the  Nummulitic  series,  consisting  ot  giey,  marly, 
fossiliferous  sandstones,  imderlying  red  conglomerate 
(cp.  IX  of  Karmirvank,  p.  400).    The  fdssik  are 

Nummulites  Umngatus^  Brag. 

Pecten;     Venus  sulcataria,  Deth.:    Panopea  inter- 

media,  So«r.,  P.  cf.  ehngata,  Lejm. 
Turritella  imbricataria^  Lam.)  T.  edita^  Sow. 
Dentalium  strangulatum,  Desh. 

The  country  N.  of  Bilev,  up  the  Ustupi  valley, 
consists — ^according  to  Valentin  ± — of  "  pyramidal 
and  roof*like  mountains,  red,  yellow  and  grey  in 

*  Gipi.  CeiaL  TUKi  Museum,  too. 
t  IbU.  loj,  104. 
4  Btriihi  Stuikmik.  Naiuff.  Gts.  Fninkfurt  a/M.  1891,  ttt. 


4o8  Geology  of  Armenia 

colour,  and  of  Tertiary  date.  In  general  the  stiata 
dip  steeply  to2S.W.  In^istinifl  fossils  oor^irred  in  the 
greenish  giey  *  Steinmsrgel*  aod  in  the  yellow  and 
broii(*n  limestoiies.**  On  the  way  from  Ustupi  to 
Paraga,  near  the  pas8»  Valentin  found  ^  thin  beck,  full 
of  ciius^^  of  Tertiary  shells.*'  lliese  are  doubtless 
Eocene,  in  the  light  of  the  fossils  from  Bilev,  recorded 
above.  \ 

About  a}  miles  W.  dP  Bilev  Tsulukidse*  found  a 
green  sandstone  interbedded  in  grev,  marly  sandstone, 
priesumably  Eocene,  whilst  a  small  ridge  near  Shuirut 
(N.W.  of  oih&X  consists  ^f  a  gr^,  mudi  weathered 
sandstone,  in  which  he  found  plant^remains  with 

OfhitoUtjfs  Fortisii^  d*An^;  NummuUtes  perf&ratus, 

N.fAssilina)  exponent^  Sow. 
Mytilus  nmostts.  Lam. 

After  crossing  the  auriferous  Alinja  rivecv  still 
proceeding  to  N.W.,  Tsulukidse  (Le.)  found  a  yellow 
sandstone  with  Turritella  imbricataria^  Lm.,  about 
I  mile  &E.  of  Keinik^  (Koinuk) ;  this  sandstone  was 
interbedded  in  a  system  of  strata  endosing  lignite. 
Here  also  he  collected  Nummulites  and  Teredina 
personata,  Lam.;  and  at  a  spot  a  mile  N.E.  of  Vaikhir, 
on  the  right  bank  of  the  Nakhichevan  river,,  he  found 
an  argillaceous  limestone  with 


Nummulitr,  perforatus^  <rOrb.>    N.  leevigatm^  _ 
N.  Ramonai^  Defr.,  N.  (Assilina)  exponemi^  So». 

Above  Jagxy,  near  the  springs^  Tsulukidse  f  records 
Eocene  limestone  with 

Canoelyteus  eonoideus.  GoMf.  (Kressenberg). 
Spandytus  tarispina^ 


*  Gipl.  CUat  Tlfis  Musmm,  io«. 
f  Mi.ios. 


Middle  &  Upper  Eocene  409 

Bettveen   Nakhichevan    and    Sutab    Abich  *   has 
d^idled'  a  se<%ion  showing  the  following  succession : 

1.  Red,  aigilboeous  marls  with  fiagmeiits  of  gypsum* 

a.  lUd,  hvd,  calcaieous  nnurls. 

3.  Smdy.  brownish  yellow  Itmestbnes  full  of  JVummtfUn. 

4.  I4i^t»  sandy  marls  with  MtmmuIHis^  OrUtMtes  uid  Echino* 


On  proceeding  up  the  Bechinakh  valley  towards 
the  West  Karabiig[h  Range,  Abidi  (I.e.)  observed  that 
the  Eocene  calcareous  beds  pass  into  dark,  detrital 
sandstones  and  brownish,  sandy  ^arls,  which  prevail 
over  the  whole  Bechinakh  (Karmalinovka)  valley  and 
sldrt  the  mountain -chain.  These  clastic  beds  have 
been  derived  from  qpntemporary  andesites  and  tra- 
chytes ;  many  of  these  Eocene  Volcanoes  seem  to  be 
still  represented  by  the  lofty  trachytic  heights  oi 
Salvarty,  Kechal,  Akdaban  (I^uki),  and  Kyabyrli  (all 
over  10,000  ftiet),  forming  a  N.W.-S.E.  series  in  the 
West  Karabagh  Range.  Beds  of  dark,  argillaceous 
limestone  crowded  with  Nummulites  Ueuigatus^  Brug.» 
occur  intercalated  in  the  clastic  series.  Lower  down 
the  Bechinakh  valley  (probably  in  the  same  series), 
near  Sheikh  Mahmut,  Radde  and  Sievers  f  found 

Nummulites  perfaratus^  d'Oib.,    N.  IcevigatuSf  Brqg. 

At  Kulidara  near  Masra  (8  miles  N.E.  of  Nakhiche- 
van they  colledled 

Nummulites  perfaratus^  d'Qib.,  N.  Ramondi,  Defr.» 
N:  complanatuSt  Lam.,  N.  biarritMensis^  d'Arcfa., 
N.  granulosus,  d'Aicfa. 

Further  N.W.,  in  the  Daralagdz  regipn,;the  Eocene 
is  generally  transgressive  on  Toronian,  and  still 
further  W.  on  Palaeozoic  strata.  The  hB$e  of  the 
Nummulitic  series  in  th^  Daral;agOz  region  is  marked 

*  Geei.  Bteb.  wm.  Kwr  6*  Araxa^  >S-'7* 
t  GeU.  OeieL  TH/Ks  Mumm,  177. 


4XO  Geology  of  AriuMia 

— according    to    Abich  ♦ — by 

**  oompact  conglonientesv  comislnig  of  mmUI  pebUes  of  maifale,  ind 
fraffuioau  of  quarU  and  poq>hjn7  ^  *  Iv^^  S^T*  cjdctfeout  cement 
However,  the^  congloaiefates  occaaonauy  aitttae  a  reddifeh  colooTy 
and  pass  itoto  a  dnnabar-ied  bitoda,  bjr  which  the  division  between 
the  Qret^ceoua  and  the  Nummtilitic  limestone  becomes  ns3>fe  as 
a  dark  red  looe  oti  the  predpitoiis  ^pes  pi  the  deep  valleys.*' 

In  this  region,  at  a  wot  2  versts  N.E.  of  Martlros^ 
near  the  source  of  the  rashalu  river  fa  left  tribiitary 
of  the  Arpa-chai),  Tsulukidse  t  found  limestone  with 

//fiiiiiiiiiZites  perforatus^  dX)ib.»  N.  Utmgatus^  Bn«^ 
N.  planulatus^  dXkfa. 

Ab>ch|  has  observed  that  Nummulitic  Limestone  is 
particularly^weli  developed  further  N.  at  Malishkent 
(Malishke),  where  |t  is  overlain  by  Oligocene  ooral- 
limestone  The  terraced  Nummulitic  strata  show 
steep  escarpments  facing  the  Dalyldi*tilpa  volcano 
in  a  semicircle,  and  consist  chiefly  of  impuce»  detrital 
limestones  wiUi  a  ovstalline  groundmass^  alternating 
with  grey,  oompatt  limestones,  whidi  hi|^ef  up 
develop  into  a  tene^  of  copaiderable  thickness  wi3th 

(Mdtoliti^   di$cu$^    Rfitim.,    0.    patettans,    lUMm., 

0.  sella,  d*Aich.;  {Nummulites  perfora$m^  4fOA. 
Ottrea;  Poctem. 
Natica. 

Tsulukidse  §  siibsequently  collected  ilie  following 
fossils  from  the  same  locality : — 

Ofbitolites  Fortmi^  d'Aich.r  0.  discus,  lMm^\ 
Nummulites  UBtngatus^  Bn«.,  N.  perforMus  dV)ib.f 
N.   UafritmnsU^   itAxA.,     N.   KUfkondi,   Defr., 

•  Dot  Sabuobin.,  Uittu  Acad.  Sd.  St  Paeysb.,  s6r.  6,  So. 
nmth,  k  phys.  viL  C 

t  Gift.  QUoL  Tlfii  Mumm.  105. 

t  ty.  WiC  14. 

f  Gml.  Coiol.  TyKtMm$mm^  113, 1x4. 


Middle  &  Upper  Eocene  41Z 

N.  granulosus,   d'Aich.,    N.   variolarius,  d'Oib., 
N.  Leymerid^  d'Arch. 
Cidaris   serrata,    d'Aidi.;     Porocidaris   veronensis^ 
MdriftB. 

Ostrea  latissima^  Deth.;   Pecten ;   Ckama  rusticula, 

Dtdi. 
Turbo;   Xenophora  agglutinans.  Lam.;  Ampullina 

cf.  Edwardsit   Dah.;     Natica  hrevispira^  Leym.; 

Melania  Cuvieri^  Desh.;  Rosteltaria  sptrata^  hcpaA 

TerAellopsis  Braunt^  Leym.;    Ob'va  nitidula^  Deth. 
Scaphander  Fortisi^  firongn. 
Dmtalfum  (Fustiaria)  ehumeum^  Liiui. 

Fucoidal  sandstone  colleAed  by  Tsulukidse  *  from 
Malishkent  (Malyshka)  and  from  keshishkiant  (a  few 
miles  to  the  W.)  may  perhaps  indicate  the  presence  of 
Upper  Eocene. 

At  Moss  (Mass)  on  the  Arpa-chai  (S.  of  Malishkent) 
Radde  and  Sieversf  found 

OrbitoUtes  discus,  Ritiin.;  Nummulites  planulatus^ 


Between  Bielav  (Belev)  and  Khachik  in  the  Kasan- 
yaila  region  (farther  S.)  Tsulukidse  |  has  recorded 
an  Eocene  marl  with  Copula. 

At  Oktaberd  (a  few  miles  E.  of  Erivan)  Abich  § 
found  Nummulites  intermedins^  d'Arch.,  which  seems  to 
indicate  the  presence  here  of  the  Priabona  horizon. 

The  Eocene  of  the  Daralagoz  distridt  becomes 
concealed  to  the  W.  and  S.  beneath  the  lavas  of 
the  Armenian  plateau  and  the  loess  of  the  Araxes 
plaiui  but  to  the  S.  of  the  Araxes  and  Ararat  it  rises 
up  in  the  Gedargan  Dagh  (between  Basyrgan  and 

^  Gt^L  Catal.  Tlflis  Mfusmm^  1 13. 
t  JNd.  177. 


4ia  CMogy  of  ArmMia 

Baya^dl  forming  the  southei^  wall  of  the  Bayazid 
plain.  Here— according  to  Abtch^ — the  Eooene  is 
overlam  by  Miotene  and  exhibits  the  following 
suooession: — 

I.  OoSlic  Eaeitoiie  with  wiq^tod  fEiiiit;  u  appftcialik 
part  is  foroMd  of  Fonuninifcim,  «^.|  MuMhnt/mm  and  MmwmHin 
mUghUt  Kdit. 

».  light  mnjt  Tvj  compact  Ifmntonft  irith  Echinodenat 
(Cm$dfpm$)  $ai yriA  BiUimUmm  and  THUem/im^,  aa  ia  i.  But  hcve 
MuiuHiUittt  appear  widi  graatcr  diftJnctacti  althw^  ifcry  maOy 
acarori^  i  oua.  in  diametcTp  #^.t  Jifi  iiitrmMu,  d'AidL.  or  Jf. 
gmmmsimtii^  J6tf. 

Nummulitic  limestone^,  and  sandstones  form  a 
narrow  selvage  on  both'^Qanks  of  the  Aghri  Dagfa, 
from  Takjaltu  to  Kagizman  and  even  fiirther  west. 
They  overlie  Hippuritic  Limestone  (p.  36A  and — 
according  to  Abicht — have  been  much  altered  by 
intrusive  rocks.  In  the  Gyngdrmas  valley  (between 
Takjaltu  and  Kulpi)  the  Nummulitic  limestone  is 

*'feddith  jtUaw  aad  marmoraceous,  and  rery  diick,  oontaining 
Nkmmmiiiii  Umifthu^  Kug-i  N.  pitftrmiiu^  tfOrth,  Al  nAvr- 
medimif  d'Aiai. 
It  rites  op  like  a  wall,  with  the  lame  strike  (E.  jo*  S.)  as  the  directioa 
of  the  Araxes.  .  «  The  limestone  ia  travened  bjr  nmneroiB  veins 
of  jreUow  aragonite  filling  np  a  network  of  cracks.  .  •  This  Kiae- 
atone  is  overlain  on  the  W.  by  a  iine-giained.  calcareous  sandstone, 
which  rises  np  to  fonn  the  sommit  of  the  neigfabonring  Gyogonnu 
Dagh.  •  .  •  •  In  descending  from  Pariy  Dagh,  down  the  Akiiak 
▼allef,  to  Nakhirkhran  and  fay^shlar  metamorphosed  limestones 
and  sandstones  of  Eocene  date  appear  to  view  from  nnder  their 
covering  of  lava.  .  .  •  The  Kabagh  Dagh  (4  versts  W.  of  Takjaltu)  is 
chiefly  composed  of  sandstone  and  Nummolitic  UmcatoiMt  for  the 
most  part  changed  into  marUe/  penetrated  by  red,  aomdaloidal 
ophite  [dialiase].  .  •  •  In  the  Pemaut  valley  (axis  B*  so*  N.)  Eooene 
limestones  and  sandstones  appear  uptilted  verticalJ^y  and  have  been 
penetrated  by  vertical  and  hoiuontal  veins  of  veqr  fine-fraincd, 
tabular  tiachyte^porphyries  (liparites).   •   .   Soon  after  leaving  the 

*  D0i  SmnMU  drv.,  M6n.  Acad  Sd.  Sl  Pfteisb.,  s£r.  «,  ScL 
math,  ft  phys.  vii.  iS^  19^ 

t  Giri.  H^rtck.  m  4m  kmtk.  ZSmbm,  iL  149,  I5}*i5&  t6o-iit. 


A  upper  Eocene  413 

^uringt  of  Jumaly  (5710  feet)  the  Penuut  TaDegr  penetrates  the 
eraptive  aone  of  Takjaltn.  •  •  The  hoists  on  the  right  ride  of 
the  vallejr  conrist  of  impure  Nommulitic  limestone  containing 
Mmmmmlhis  HarrUgemsis^  d'Aicfa.»  i|nd  with  geodet  of  green  evth. 
The  other  tcansverse  valleyi  of  the  Shatin  [Ariiri]  Dach  as  fiu  as 
yigSf^^  thow  the  same  structure  as  the  PemMt  iraiOej. 

The  Eocene  of  the  Aghri  Dagh  evidently  extends 
still    further    west,    for    Koch^    states    that 

<*  the  great  P^tfin  t>arin,  especially  on  the  S.E.  ride  of  the  river  [Kala- 
6hai|t  has  batn  much  interrupted  by  sedimentary  rocks,  espedsJly 
by  n  marmoraceous  NummuUtic  limestone." 

On  the  same  line  of  strike  as  the  Aghri  Dagh,  but 
about  80  miles  due  W.,  in  the  Frat  vsj^ey  near 
Mamakhatun,  Abicht  has  recorded  a  nuiiaber  of 
distinctively  Eocene  Foraminifera,  together  with 
several  Sica- urchins  (much  crushed  and^  damaged), 
which  he  iidentified  with  Miocene  and  Oligocene 
forms,  and  he  ascribed  a  Miocene  age  to  the  series. 
Until  more  complete  material  becomes  available  it 
seems  to  me  advisable  to  rely  on  the  evidence  of 
the  Foraminifera  rather  than  on  the  imperfedllv 
preserved  Echinoids  which  need  revision,  lliey  v^ill 
probably  be  found  to  be  more  closiely  allied  to  the 
Indian  Khirthar  species  than  to  the,  Mediterranean 
forms. 

Abich*s  sedlion  of  the  cliff  (300  feet  high)  of  the 
right  bank  of  the  Tuzla  Su  near  Mamakhatun  shows 
the  following  succession : — 

u    Yellow-white  limestone  with  Ofbitoides  dis^ 
pansa.  Cart,  var.nov.,  0.  ephippium^  Cart  (Kressenber^^). 

2.  Bryozoa-Umestone  with  abundant  Foramuii- 
fera.    Ceriopora  intricata^  d'Aidi.,  is  very  abundant. 

3.  Echinoid-limestone  with  Foraminifera : — 
Polystomella ;    Amphistegina  nummulincefarmis^ 

•  Xiiu  impmUiHhm  GMrgi^  iL  jsA 


414  G$otogy  of  Armenia 


AMch  (Eocene  of  Thriatetic  Mountains,  p.  396, 
and  in  the  N.E.  part  of  the  Pontic  Range, 
vide  infra);  Orbitaides  (Discocycliha)  papynicea, 
BooMe,  0.  Fertisi^  Sow.;  Heterostegina. 
Clybeaster  Martinianus^^  Da  Moalms  (in  Khirthar 
Of  Cutch  with  Orbitaides  dispansa  and  0. 
ephifpinm),  C.  Mtchdotti,  Ag.  P;  Schio],  C. 
^rifarmi$.  Ahich  (itilated  to  the  last) ;  Echiwh 
lampas  gMndus^  Laube  [?;  Vicentinian  Mio- 
cene], E.  Matherani^  D«s  Moulina  [?;  Crosara], 
£•  acuminatus^  Abidi;  SchiMaster  Parkinsom^ 
Defr.,  with  the  Eocene  Orbitaides  papyracea 
attached  [a  very  doubtful  determination,  for 
this  sea-urchin  is  charadteristic  of  the  Mio- 
cene of  Malta ;  cp.  the  species  oi  SchtMOSter 
in  the  Eocene  of  the  mid-Araxes  gorge,  p. 
404],  S,  eurynatus^  Ltubei  much  crushed  [?; 
Miocene  of  Malta] ;  Spatangus  delpkinus^  Defr. 
r?;  Miocene  of  Malta]. 
4*  Orbu/ffia-limestone,  white,  oolitic  and  tra- 
yertinous  in  appearance,  with  black  flints  in  regular 
bedding.    It  is  full  oiOrbulina. 

All  these  beds,  which  dip  30^  N.W.,  rest  discordant- 
ly on  strongly  inclined,  green,  serpehtinous  psammites 
and  pelites  (p.  387). 

To  the  S.W.  of  Mamakhatun  Tchihatcheff*  found 
a    dtstindl    outcrop    of    Nummulitic    Limestone:— 

**  $  leagues  E.  of  K41aracfa  and  N.  of  Shekltt  this  KmesUme  tret 
(overlying  serpentine)  is  announced  at  first  by  rodcs  and  rounded 
lieights  completely  impregnated  by  iron  osdde ;  Ac»  the  rock  passes 
into  bluish  or  yellowidi  niarls  and  dark  grey,  siliceous  limestones ;  dit 
latter  occur  at  a  height  of  1680  metres  and  contain  the  foUoiriiig  fotrib : 

(MHoliUs  Fartid^  d'Arch.;  Nmrnwrnlihs  fttfMmius.  tfOib.,   M 
ZMOuamts^  Defr. 

(hirm  jirMda\  BeD. 
Noaunulitic  Limestone  still  forms  rounded  hOls  at  a  hio^  of  16S9 

ii  ilinmrit  GMgii^  ir.  433. 


Middle  A  Upper  Eocene  4x5 

Further  W.,  in  the  valley  of  the  Upper  Kuru*chai, 
W.  of  Kemakh^  Hommaire  de  Hell^  found  Eocene 
strata  in  the  following  succession :— > 

1.    ConiMct  limestOM. 

t.     LimesloBe  (**  Cakaiie  gronier  **)  wiUi  MmmeUkt^  A^teUe^^ 


^    Cottie  nadftoiie. 

4.    CoQ^omeimte  al  the  bottom  of  die  TiOej. 

In  the  Pontic  region  at  Baibtirt  Abich  f  found  the 
foUowiog  thick  succession  of  Eocene  strata  overlying 
Cretaceous  marble  (p.  340) : — 

l^    Calcareous  marls  and  sandstones. 

3.  YelloWt  arenaceous,  or  compaA  limestone, 
ixrith  the  following  fossils : — 

Nummulites  btarritzensis^  d'Axcfa.  (abundantX  N. 
perfaratus^  d'Oib.,  N.  ketngatus^  Bng.^  N.  cbesus^ 
hefOLt  N.  Lucasanus^  Dtb.  (abundant  and  very 
variable  in  size),  N.  Guettardi^  d'Arch.,  N.  Ley- 
merieiy  d'Aich.,  N.  cf.  Ramondi,  Dcfr.,  N.  granule^ 
sas,  d'Arcfa.,  N.  jdanuUaw,  d'Orb.,  N.  (Assilina) 
exponenSf  Sow.;   Otbtteides  (Discocyclina)  papyra* 

cea^  BouMe. 
Ostrea  gigantea^  Bmnder. 

Abich  (I.C.)  traced  these  Nummulitic  beds  .S.W.  of 
Baiburt  to  a  point  S.  of  the  Keikid-Chorokh  water- 
shed; tiiey  form  low  hills  (with  E.-W.  strike)  on 
the  way  nom  Varisma  to  Pullarik  (5470  feet),  on 
tl^  S.  border  of  the  great  Kachevan  or  Varzahan 
plain*  The  limestone  at  times  becomes  more  sandy 
and  less  compadl,  and  then  contains  innumerable 
Nummulites.  Further  W.,  as  far  as  the  low  pass 
of  Sadagjno-Syrty  (5859  feet)— on  the  Kelkid-Chorokh 

^  Vpfogim  nvpdfdc.p  Gialcigii,  it.  73 ;  and  Viqneiiiel,  SuU,  Su. 
GIfi.  JFhmHf  t<r.  t,  tu.  soS. 
t  (fW.  Fifnek,  ifidmkattk.  JUniam^  vL  115-1171  •4i-s49. 


4i6  Geology  of  Armoma 

watershed  between  Pliir  and  Sokman — ^Abich  (Ix.) 
found  NUmmulites  in  extfaordinary  profusion  and 
variety,  especially  of  the  following  forms : — 

Nummtdites  biarritMensis^  <rAi€h.»    N.  Lucasanus, 
I>efr.|  iV.  Guettardit  d'Axch. 

They  lie  in  a  formation  of.,  whitish  yellpw  limestone 
and  calcareo-argillaceous  sandstone. 

It  was  evidently  this  Numinulitic  series  to  which 
de  Chancourtois  referred  in  a  verbal  communication 
to  d'Archiac.*  He  states  that  he  found  ''beds  10-12 
metres  thick  between  Gumiishkhaneh  and  Baiburt, 
forming  plateaux  1000  metres  in  hdght  [Le.,  the 
plain  of  Varzahan].    They  contain 

OfbitoUtes    Fartm^    d'Orb.»    0«    subtnedia^    d'Aidu, 
Nummulites  spira^  d'Arch.,  N.  Ramondi^  Defr. 

The  result  of  my  own  observations  on  the  geosyn- 
clinal  plain  of  Varzahan  may  be  thus  summarized : — 

I.  Greenish^  serpendnous  sandstones  between  Kersi  and  Vana- 
|ian  (pi  139)9  unfosstliferousy  passing  into  pale  yellow,  calcareous 
sandstones  between  Varsahan  and  Baiburt  ^p.  3$,  139),  and  into 
greenish  and  grey  shales  between  Varsahan  and  Osluk  Khan  (p.  36). 

8.  Nodular,  buff  limestone  of  Kem  (p.  S39X  with  a  narrow 
fossiliferous  band  containing  abundant  individuals  of  NummubUs 
petfmraius^  d'Orb,  (var.  B.  adtmbrauis^  sub-var.  e.  oi  d'Archiac),  and 
ill-preserved  specimens  of  Sfo$idyIus  to.  indet. 

3.  Cateareous  flagstones  of  ,  (a).  Pale  brown,  sandy  lime- 
KersL  stones,    thinly   bedded,    between 

Baiburt  and  Varzahan. 

(b).  Massive,  yellow,  odcateous 
sanostones,  overlying  grey  dudes 
near  Baiburt  (p.  338). 

This  Eocene  area  is  continued  westwards  down 
the  Kelkid  valley.  Tchihatcheff  j  met  with  it  on  the 
S.  slope  of  the  Kangly  Dagh ;  a  league  S.  of  KSseh 
(1 712    metres)    he   entered   on    **a   rath^   uniform 

*  IBsMn  iisfrogrh  ie  ta  gto^giif  vi.  194. 
t  AsU  Mimun^  GMtgii^  iv.  431. 


Middle  &  Upper  Eoceiie  417 

country  sunounded  by  heights  of  yellowish  limestone^ 
dipping  S.E.,  as  seen  in  the  heights  od  which  Salyar* 
dak  and  Persar  are  situated;  the  surface  of  the 
country  is  strewn  with  limestone  fragments  with 
Ostrea  Arckia^^  BeU." 

Nummulitic  Limestone  has  played  a  considerable 
^ut  in  the  strudture  of  the  Pontic  Range.  Near 
£mberek»  in  the  Kharshut  valley  below  Ardasa, 
TchihatchefiE  *  found  black  limestones  containing 

Nutnmulites    Rdmondi^    Defr.;    Oybitoides   (Disco^ 

cyclina)  $ubfnedia^  d'Arch »  0.  dispansa,  Cart. 
Trochoseris  (casts  and  impressions). 
Pecten  solea^  Desh.,   P.  cf.  pleheius^  Lam.,   P.  de- 
cussatus,  Munst. 

The  beds  dip  steeply  to  N.E.,  and  pass  into  bluish 
or  white  marls  with  the  same  dip,  but  infinitely 
poorer  in  organic  traces.  The  fossiliferous  zone  only 
constitutes  a  narrow  band  scarcely  i  kilometre  in 
breadth  from  N.W.  to  S.E.  It  is  merely  one  of  a 
series  of  seams  of  limestone,  alternating  with  zones  of 
•*pyroxenic  porphyries,"  from  near  Giimiishkhaneh 
down  the  Kharshut  valley  to  the  Hassanly  plateau,  S. 
of  Tireboli.  Tchihatcheff  indeed  points  out  that 
the  numerous  alternating  bands  of  ''pyroxenic  por- 
phyry** and  limestone  have  so  narrow  an  outcrop 
that  he  could  not  represent  them  all  on  his  map. 

Still  further  W.  he  noticed  a  similar  alternation 
of  limestone  and  igneous  rock  in  the  transverse 
valley  of  the  Ak  Su,  which  enters  the  Black  Sea 
at  Kerasun,  the  sedimentary  beds  consisting  here 
of  ^*  whitish  or  yellowish  limestones,  passing  into 
blue  and  yellowish  marls."  The  coast  between 
Kerasun  and  Tireboli  reveals  the  same  alternation  of 
beds. 

^  Asie  Mimufi^  Ghlagii^  L  26^^  ii.  428. 


fi 


4i8  Geology  of  Armenia 

Nfy  own  observations  in  the  Pontic  Range  further 
&•  (pp.  ig-aa,  349-257)  disclose  a  similar  series  oi 
black  or  grey  limestones  and  shales  alternating  with 
andesites.  In  the  Machka  Dereh.  between  Aiama  and 
Hamsikoi  large  boulders  of  pale  grey  limestone  en- 
cumbered the  road,  having  lallen  from  the  heights 
above  (p.  3i^.  It  is  very  hard  and  splintery,  with 
numerous  thm  veins  of  calcite.  Only  an  obscure 
indication  of  a  Pe£Un  could  be  seen,  but  the  micro- 
scope  revealed  numerous  Foraminifera,  chiefly  of 
Miliolina  with  a  less  number  of  Textularia.  The  rock 
in  situ  by  the  roadside  was  a  dark  green,  compact 
andesitic  lava. 

On  the  S.  side  of  the  Range,  in  the  Khacshut  valley, 
about  halfway  between  Ardasa  and  Kharava,  I  found 
a  very  similar  grey  limestone,  which  had  been  in- 
truded by  a  homblende-pyrozene-porphyxite.  The 
microscope  xevealed  Foraminifeca  very  similar  to  the 
above-mentioned  forms  but  in  larger  and  more 
numerous  individuals,  as  follows : — 

Miliolina  cf.  trigonula,  Laa.,  of  the  Paris  Eocene ; 
Globigerina;  Textularia. 

Further  E.,  near  Chorak  Khan  Q)etween  the  Kitowa 
and  Kazikly  ranges,  p.  xig),  I  noticed  a  dark  purplish 
grey  marble,  dipping  30^  S.  by  E.,  alternating  with 
contemporary  Hows  of  dark  green,  amygdaloidal 
augite-andesite.  The  microscope  revealed  traces  of 
Lithothamnion  and  a  sedtion  of  an  Echinoid  spine, 
but  no  Foraminifera.  The  rock  has,  however,  under- 
gone some  re-crystallization.  It  is  probably  Eocene, 
and  may  be  parallelled  by  the  **  brecciated  limestones 
with  numerous  microscopic  remains  of  Lithothamnion 
and  shell-fragments^'*   which  Inostranzeff^  records 

^  Au  trovers  ie  h  eMm  prgfu^ak  iulCaue^u^  947. 


Middle  &  Upper  Eocene  419 

on  the  S.  slope  of  the  Caucasus,  regarding  it  as  a 
littoral  deposit  contemporary  with  the  Nummulitic 
Limestone.  A  o^mpanson  may  also  be  made  to 
Giimbers  ''  granite- marble  **  with  Lithothamnion  from 
the  lx>mbardy  Alps  (Montbrfano). 

To  the  N,  of  this  locality  and  on  the  N.  side 
of  the  Pontic  Ran^e»  in  the  Meiriman  Dereh  between 
Sumelas  and  Jevizlik,  1  noticed  numerous  seams 
of  shales  and  black  limestones  alternating  with 
contemporary  homblende*p3TOxene-'andesites  (pp.  255, 
2Slb).  The  limestone  at  Merdosh  revealed,  under 
the  microsicope,  numerous  Foraminifera,  very  small 
and  apparently  dwarfed  by  adverse  conditions.  They 
consist  mainly  of  Miliolina^  and  Textulariti.  The 
series  is  doubtless  continuo\is  with  that  of  the 
Machka  Dereh  (p.  418),  only  8  miles  to  the  W.  If,  as 
seems  likely,  the  whole  series  is  continuous  with 
Tchihatcheff's  Nummulitic  Limestone  of  Emberek 
(p.  4:17)  the  Pontic  Range  consisted  of  a  series  of 
Eocene  volcanoes^  which  poured  out  their  lavas 
into  a  sea  extending  both  to  N.  and  S.,  and  this 
mountain*  belt  bias  apparently  remained  dry  land  ever 
since  the  close  of  the  Eocene  period.       '  / 

This  conclusion  receives  additional  support  from 
the  fa(%  that  in  the  Karchkal  region,  at  the  N.E. 
extremity  of  the  Pontic  Range,  Eocene  strata  pre- 
ponderate, likewise  associated  with  contemporary 
lavas  and  tuffs.  Here — ^according  to  Batsevitch^ — 
the  Nummulitic  beds  rise  up  to  the  height  of  11,248 
feet — ^an  even  greater  altitude  than  that  to  which  the 
Nummulitic  Limestone  has  been  raised  in  the  Hima- 
layas. Near  the  source  of  the  Machakhelis-tskhali  on 
the  N.W;  slope  of  these  mountains  Batsevitch  {l.c) 

*  CM.  Researtkis  in  ttu  Baium  6«  Arhin  DistriOi,  Mat  Geol. 
Cauc.  TSilit,  1887,  Ser.  s,  w  197.  [Id  RomabV  See  also  GM. 
Cetel  nfib  Museum^  119-113. 


420  Geology  of 

found  dark  grey  tuff  alternating  with  marls  and 
iandstbnes,  which  rest  on  a  thick  series  of  light 
reddish  and  greenish,  crystalline  limestones  crowded 
with  the  following  Foraminifera  :— 

Amphistegina  nummulinceformis^  Abich ;  Nummuiites 
Icevigatus^  Brag.;  Ofbitoides  ( Discocycltna)  cbV 
pansa.  Cart. 

The  same  horizon  occurs  on  the  opposite  side 
of  the  Karchka]  Dagh,  near  Sinkot,  on  the  left 
bank  of  the  Imerkhevi  gorge.  Here  Batsevitch^ 
found  a  reddish  sandy  limestone  crowded  with 

Globigerina ;  Amphistegina  nummulinoeformis, 
Abich :  Nummuiites  laevigatus^  Brag. 

This  region  adjoins  the  Eocene  area  of  the  Imeretic 
and  Thrialetic  Mountains,  which  form  the  Northern 
Border*Ranges  of  Armenia  (p.  3()6). 

Nummulitic  Limestone  is  extensively  developed 
in  the  Taurus  Mountains,  the  Southern  Border- 
Ranges  of  Armenia,  but  there  are  no  modem,  precise 
geological  observations  covering  this  important  area. 
In  the  plain  of  Apaud  (Arpaout),  between  Kharput 
and  Keban  Maden,  Hommaire  de  Hellf  discovered 
Nummulitic  limestones  and  fine-grained  sandstones, 
containing  the  following  fossils,  which  were  deter- 
m  ined  by  Viquesnel  % : — 

Nummuiites  Rdmandif  Defr.,  AT.  laevigatus^  finig.> 
v&r.,  N.  intermedius^  d'Aich. 

It    i$    more  than  probable  that  the   last-named 

*  CM.  ItistanJUs  im  ihi  Bahtm  ^  Artvin  DiOridi^  Mat  GeoL 
Gauc  Tiflia,  1887,  Ser.  t,  i.  iit.  [Tn  Rosmn].  See  aiio  Ckti. 
CatttL  lifiis  Mmsimmt  12a. 

t  Vbyapen  Titnfm'stk.f  Cfiahgie^  W.  8a 

i  Butt.  Sec.  GM.  B^Ma,  afr.  a,  rii.  507. 


Middle  &  Upper  Eocene  421 

species  occurs  in  a  higher  bed  than  the  other  two. 

Aca>rding  to  Smyth  ^  the  higher  parts  generally 
of  the  Taurus  between  Kharput  and  Arghana  consist 
of  calcareous  strata  abounding  in  Nummulites,  over- 
lying Cretaceous  beds.  The  rock  of  the  Arghana 
monastery  (2000  feet  above  the  plainj  and  of  the 
neighbouring  heights  is  a  '*compa($t,  light-coloured 
limestone  abounding  in  Nummulites  and  sometimes 
with  fragments  of  Pecten  and  O^trea.Y'  At  Arghana 
de  Chancourtois :{;  too  recognized  the  same  Nummuli- 
tic  Limestone  which  he  had  already  observed  twice 
before,  viz.,  to  N.  and  N.W.  of  Jezireh  (vide  infra). 
Ainsworth§  also  states  that  the  summits  both  pf 
the  Kala'at  Dagh  (the  Arghana  bluff)  and  of  the 
Ali  Dagh  (N.  by  W.  of  this)  consist  of  "a  limestone 
with  Ostracea,  Cones  and  Nummulites,''  overlying 
carbonaceous  marls  (probably  Lower  Eocene^  p.  384). 

Ainsworth  (I.e.)  also  states  that 

'*  in  the  mountains  4  miles  E.  of  Arghana  Maden  the  silico-magnesian 
rocks  (euphotides,  actinolitic  and  diaUage  rocks)  are  accompanied 
fri^>eri<Miy  by  an  indurated,  argillaceous  limestone  with  abundant 
fossils,  especially  Cythena,  Vemu,  Cones  and  Cmihium,  and  below  by 
sandstones  with  larae  Osiraaa  and  PicUH.  .  .  •  In  the  Mihrab  Dagh 
[N.W.  of  Arghanaj  the  summit  is  of  limestone,  strike  N.  80*  £., 
which  stretches  far  down  the  slope,  where  it  rests  on  carbonaceous 
stea-schista^  diloridc  slate  and  highly  ferruginous  diallage-rocks.  The 
limestones  contain  few  shells,  but  Cones,  abundant  Catdiaua  and 
impressions  of  PtcUn  are  easily  procured.'' 

To  the  S.E.  of  Arghana,  up  to  the  basalt  plain 
extending  from  Sherbet  Khan  to  Diarbekr  Ainsworth 

'*  traversed  3  ranges  of  low  hilli  of  compact  granular  limestones  with 
few  shells,  chiefly  Cones  and  Ostraeites  and  impressions  of  Piciem.** 

At  C^rger  (3724  feet)  on  the  Euphrates  (S.W.  of 

^  Quart.  Joum.  del.  Soc.  L  331. 

t  nu.  11%. 

X  Comfiis-nnims  Atad.  Set.  xnvu  S30. 
{  Ruionha  in  Assyria,  ek.,  tl%  ei  sqq. 


423  Geology  of  Armenia 

Arghana  and  the  Mihrab  Dagh,  and  on  the  same  line 
of   strike)    Ainsworth*    ob%rved    that 

*'  the  lower  beds  ooMit  of  ted  laiidrtoiie  and  taiid$tone-ooiigloiiierater 
fiupportitig  KmaiCone  with  PecteiUi  Cones,  Madreporites,  Sec  The 
stnu  dip  in  oppodte  dneedons  at  the  two  adet  of  die  valley." 

A  similar'  series  of  Eocene  strata  can  be  traced 
in  tlie  hills  of  Mardin,  which  frame  in  the  basin  oE 
Diarbekr  on  the  S.  Ainsworthf  states  that  the 
Jebel  Mardin  commences  at  Derek  on  the  W.*  rising 
out  of  the  basalt  sheets  oi  the  Karaja  Dagh,  and  that 

*'it  oonaistaof  a  range  of  Hmettpne  hilis*  which  terminate  rather 
abnipdjr  in  the  plain,'*  evidently  forming  a  h^ld  efcarpment  to  the  S. 

The  limestone  range  is  continued  eastwards  into 
the  Baarem  hills,  which  bound  the  Jezireh  plain 
on  the  northi  and  consist  of  **  granular  and  indurated 
limestone  in  curved  and  contorted  beds,  dipping 
mostly  S.  and  S-W."  This  range  (also  called  Jebel 
Tur)  is  composed  of  "limestones  with  Nummuiites, 
limestones  with  Pe^Hnides  and  Ostracea^  and  the 
ChaIk-formation/\ 

At  Mardin  (380P  feet)  ''a  sandstone  is  capped  for 
about  109  feet  with  a  ^st raw* yellow,  friable  limestone 
abounding  in  Numniulites^  which  also  occupies  the 
summit  of  Kala*at  Marah  and  C^ir-i-Yakub  hiiW 
^nd    Ainsworth,^    whilst « at    Mardin^    was    shown 

"  laige  Osirucitis  and  other  iheUs,  as  alto  the  teeth  of  a  shark,  said  to 
bave  been  obtained  fitoa  the  summit  of  hiUs  to  the  eastwaid.  Ont 
bed  here  was  fntt  of  tOUihs  [MWdiina]  like  an  o<«ttc  or  i^Utic 
rook^  and  another  near  ICaidin  was  fonned  almost  endrely  of  broken 
or  comrainnted  bi?diQa-Aells.  .  .  The  casde  andiown  of  Mardin  are 
built  upon  the  same  fsBoir  limestone  with  fissile  and  argillaceous  beds 
bt!neatn ;  the  whole  nfOOff,  on  comfMict  and  often  granular  limestone 
with  Ostrgata.  «  .  At  IHra  [E.  of  Hardin]  the  limestone  is  ex- 
cttAoifif  fiMsittfbiows,  containing  Osirm  and  Cotnts  (both  of  Isige 

*  frntm.  Ay.  (fSitr.  Set.  x^  33s. 
t  Jtemtfd^  in  Attjffim^  sft;^  150. 
I  IM.  S5S ;  and  slao  his  Sufkraiis  Bj^eHHon,  n.  333,  345.  3|t* 


Middle  &  Upper  Eocene  423 

meX  /Vflteii,  Aim  {laigeX  -P^f^^mmt  Vtnus^  Cvi^ta  and  CdrAum. 
•   •  •   Noith  of  ICazdtn  the  next  hUl-iaBge  wpleiys  carbonaceous 
measures  at  die  foot  of  the  hlBs,  below  t£e  Dara  and  NummuKtic 
linwilnnr.    Crosaog  this  nui^e  the  Upper  Chalk  formation  occurs.'' 

From  these  rather  va^e  observations  of  Ainsworth, 
in  ooiijun(5tion  with  his  secftion,  the  following  suc- 
cession would  appear  to  hold  good  with  regard  to  the 
Eocene  of  this  part  of  the  Taurus : — 

I.  Upper 'Eocene.  Dara  and  Mihrab  Kmestonoi  #ith  Osirea^ 
PtUuit  Vtrnu^  Q^kina^  Cardium^  Cmthimm^  Fums^  Cmim^  Pieure* 
imm  and  shark's  teeth. 

a.    Middk  Eocene.    Numninlitic  Limestone  with  Mmmuiita 

3.    LoKfer  Eocene    Marls  and  sandstones  with  seams  of  coal 

(pp.  3«4,  3«S)- 

4*    Upper  Oetaoeous  (p.  367). 

Further  E.,  de  Chancourtois  *  in  travelling  from 
Bitlis  to  Jezireh,  found  the  Lower  Eocene  sandstones 
(p«  385)  to  be  overlain  in  places  (not  specified)  by 

*'  thidc  beds  of  greyish  limestones  with  Mmmaiftts,  JiOm  and  sea- 
urchins  Ifte  hats  [?  a  Canoclypeus^  cp.  p.  404],  but  Uttle  disturbed 
compared  to  the  metamorphic  rocks.  limdhr,  to  the  S.|  one  sees 
a  regubur  chain  of  limestone  [probably  Eocene],  strongly  raised  to  the 
N.p  foraung  a  wall  pierced  by  a  single  gate,  the  defile  of  the  Jennet 
Su  [Tigris],  and  then  the  deiert  [of  Mesopotamia]  begins.'* 

The  locality  is  evidently  in.  the  open  region  of 
the  Lower  Bohtan.  Su  valley,  between  Sert  and  tj^ie 
Tigris.  De  Chancourtois,  in  a  verbal  rqmmunication 
to  d'Archiac^t  has  eiven  the  additional  information 
that  on  this  part  of  his  route  he  found  **  very  fine- 
grained, argillaceous  sandstones,  overlain  by  beds 
of    limestone    with 

Nummulites  variolarius,  d'Orh.,  N.  scaber^  Lam/* 

-Aansworth,!   in   travelling  on   this  route   in  the 

:    •  Ott^iu-mubts  Aud.  Set.  zviiL  Ss;. 
t  HuMn  iisfrogris  de  la  ^gii^  iii.  194- 
%  Trmueis  ^  JUsmnka  in  Asia  Mhm^  ii.  %i^. 


4^4  Geology  of  Armrniia 

reverse  direAion,  from  Jezireh  to  Sertp  states  that 
between  Konakti  and  Finduk  he  pa^ed  through 
*'a  narrow  glen  with  fossils  of  supra-Cretaceous 
limestone  with  Ostracites  and  Serputce^  overlying  red 
sandstones  with  carboniferous  beds  on  the  hills 
between  Finduk  and  the  Tigris/'  At  Guellek 
(Chelek)  he  passed  through  ^'  lofty  clife  of  limestone/* 
At  Jaminiyah  he  again  met  with  ''carbonaceous 
measures,  but  no  true  coal/'  Still  further  N«,  between 
Redvan  and  Sert,  on  the  right  bank  of  the  Bohtan 
river  he  passed  through  ''another  narrow  pass  of 
limestones  on  red  sandstones/' 

It  was  probably  hereabouts  that  de  Chancourtois 
found  his  Nummulites  (p.  433),  especially  in  the 
light  of  the  statement  by  Loftus^  that  after  leaving 
Sert  (for  Jezireh)  he  "skixted  the  N.W.  base  of  the 
inaccessible  Bohtan  Mountains/'  composed  of  a  lime- 
stone, which  he  considered  to  be  '-undoubtedly 
Nummulitic,  a  rich  cream  colour,  compa<ft  and 
crystalline  as  at  Kirrind  and  elsewhere  '  in  the 
Zagros  Range — tiie  south-easterly  continuation  of  the 
Taurus.  In  the  Zagros  the  Nummulitic  Limestone 
reaches  a  v^ry  sreat  thickness  and  development. 

to  the  IS(.  (^  Sert,  but  still  on  the  S.  side  of  the 
metamorphic  axis  of  the  Taurus  (pp.  377,  278), 
Loftus  (Uc.)  found  Alveotina  subpyrenaica^  Lqrm.,  in  a 
limestone  at  Varkhan  (Werkhantz),  S.  of  Dukhan  and 
on  the  N.  bank  of  the  Keser  Su,  on  the  Sert-Bitlis 
road.  It  overlies  "  blue,  altered,  contorted  limestone^" 
and  is  unconformable  overlain  by  the  gypseous  marls 
(probably  Upper  Miocene,  q.v.)  of  the  Keser  basin« 
The  fossil  would  indicate  the  presence  of  the  lower 
division  of  the  Nummulitic  Limestone/ 

On  the  N.  side  of  the  Taurus  Eocene  strata  were 

^  Quari.  /mm,  CM.  Ar.  xL^^ts^ 


Middle  &  Upper  Eoc$Me  435 

found  by  de  Cbancourtois  at  Surb  Karapet  (Chan- 
geli  or  **  Tdianlu-kilissa,''  6480  feet)  in  the  valley 
of  the  Kuro-chai,  N.W.  of  Mush.  Here,  according 
to  a  verbal  communication  made  by  de  Chancourtois 
to   d*ArchiaCy*   he    observed   limestones  containing 

OfiatoUtes  Fartisi^  d'Aichc^  0.  subtnedia^  d'Axch. 

Kochif.  who  also  visited  this  locality,  evidently 
refers  to  these  Eocene  beds,  and  his  rather  vague 
4escrijption  may,  I  think,  be  interpreted  to  include 
overlying  Miocene  strata,  which  I  found  to  be  so  well 
developed  in  the  Murad  gorge,  only  x6  miles  to 
the  W,SW.^  along  the  same  line  of  strike  (pp.  193- 
196).    Koch's  account  is  as  follows : — 

**  On  lemng  Sorb  Kuapet  the  zigsag  path  towards  Mush  traverser 
bare  rock,  showing  various  Tertiaiy  formations  of  varjing  thickness, 
underlying  the  trachytes  and  ^olerite-porphyries  {i.i.t  the  basalts  of 
the  Surb  Karapet  plateau].  The  hmestone,  almost  dsiiling  in 
whiteness,  contami  PetUacrimtts  and  EnirinUes^  besides  Foramimferst 
and  NummuUies^  while  in  other  phces  OOraeUes  and  other  shells  were 
intermingledy  and  in  a  tiiud  locality  actual  cdral-ieefii  occurred  of 
a  Und  of  nb^am.  A  fine-grained  molasse-sandstone  was  inter- 
bed&ed.    Trachyte  and  basalt  had  broken  Ihroi:^  these  strata  in 


disfregriiielag^giifVSui^    At  the  ttme  of  prin|&if 
p.  196  I  was  not  aware  of  this  reference. 

t  jBmsr  imfamtiMhm  GMr^^  iL  393. 


CHAPTER  XXII 

THE  OLIGOCBNS  OP  AJtMENIA 

During  the  Upper  Eocene  a  considerable  area  of 
Armenia  had  become  dry  land,  and  in  the  Oligocene 
a  belt  of  land  separated  the  dorthem  sea,  which 
covered  the  Akhalztk  basin,  from  the  warmer  Medi- 
terranean sea,  in  which  corals  flourished  near  Erivas. 
In  the  northern  area  (Akhaizik,  Rion  and  Tiflis 
regions)  sedimentation  appears  to  have  been  con- 
tinuous from  the  Upper  Eocene  to  the  Upper  Oligo- 
cene, and  the  Oligocene  sea  was  a  gulf  of  the  North 
European  sea.  Its  fauna  shows  a  close  connedion 
not  only  with  the  Oligocene  of  Lake  Aral  but  also 
with^  that  of  the  Mainz,  Belgian  and  Hampshire 
basins.  The  upper  beds  indicate  a  gradual  shallow- 
ing of  the  gulf  with  brackish-water  epndttious  and 
final  evaporation.  Since  that  time  the  Akhaizik 
region  has  remained  dry  land.  I  have  compiled  the 
following  succession  of  the  Oligocene  in  this  basin 
from  the  writings  of  Abich,*  Simonovitc  h  and  Soro- 
kin,t  and  Foumier4  and  the  colledlion  of  Bayem.§ 

^  VlarMckimU  geoi,  GrundtSge^  143,  144 ;  and  his  Ge^i.  F0nch,  U 
dm  ktuA.  Ldndem^  ii.  284-307,  314. 

t  Gwloiy  of  the  Tertiary  Akhahik  Basin ^  Mat.  Geol.  C««ic.  TUis, 
1886,  1-39.    [In  Russian]. 

t  Description  ghi,  du  Caucase  central ^  169. 

\  Geol*  CdtaL  Tiflis  Museum^  32,  55-70* 


OligOCMB  437 

The  Lower  OUgoceoe  (Liguriaa  and  Tongrian)  can 
be  divided  into  three  secftions : — 

1.  Dark  brown,  oQpcretionary,  argillo-calcareous 
sandstones,  with  the  fbllowing  fossils,  chiefly  as  casts 
with  a  bright  red  coating  of  iron  oxide : — 

Pedien  arcuatus^  Bnicchi  (Cardium  subienuisulcatum^ 
tfjntf  of  Abich,  corredted  by  Oppenheim  ^)«  at 
Ab;   Cardium  araUnse^  Abich;   Arctica  tumida^ 

Nytt. 

Rostellarta  (Hipfocrene)  macroptera^  Laoim  (of  the 
Aral  beds) ;  oamsia  (Cassidaria)  atnbigua^  Sol. 

2.  Soft,  argillaceous  sands,  occasionally  ferru- 
ginous (with  fragments  of  augite-porphyry  reddened 
by  iron  oxide)  sometimes  pelitic,  containing  the 
following  fossils : — 

Ostrea  ventilabrum  w.  prona^  Wood  (Brockenhurst) ; 
Arctica  rotundata^  Brown ;  Isocardia  crassa^  Nyst 
(at  Akhalzik,  Ab  and  Arikhe),  /.  bipartita,  Abtch 
(near  to  /.  multilatnella  of  the  Aral  beds),  L 
harpa^  Gf.;  Meretrix  aff.  nitidula^  Lam.  (of  the 
Aral  beds ;  at  AkhalziK,  between  Akhalzik  and 
Suflis,  at  Ab  and  Arikhe). 

Natica  hantaniensis^  Sow.;  Cerithium  /inui,  Desh.  (at 
Arikhe). 

3.  Soft,  loamy  sandstones  in  confused  bedding, 
sinking  down  steeply  to  the  Pozkhoff  valley  below 
Akhalzik ;  they  contain  the  following  fossils : — 

Harpactocardnus  cAhalzicus^  Bittner  (between  Akh-^ 

alzik  and  Suflis). 
Aturia  ziciac^  Sow.,  A.  Burtini. 

The  Upper  Oligocene  (Aquitanian)  of  the  Akhalzik 
basin  shows  the  following  succe^ion : — 

*  Die  FriaiomauUehiem^  Palcontograpluca,  ztvii.  (1900)  305. 


438  Gtohgy  of  Amunia 

1.  Loamy  saadstoaes  with  argillaceous  septaria, 
alternating  with  variegated  marls,  dark  green,  gypsi- 
ferous  days  (with  weathered-out  Glaub^'s  saltsX  in 
which  seams  of  lignite  occur,  containing  pyrites  and 
hence  useless.    Weak  salt  springs  occur  sporadically. 

2.  Grits^  tufe,  volcanic  breccias  and  conglo- 
merates of  a  Nagtlftuh  chara<fter,  with  intercakLn- 
beds  of  sandstone  with  many  small  Cyrena. 

3.  Argillaceous  sands  and  grits  with  intercalary 
calcareous,  fossitiferous  beds,  altogether  150  feet  in 
thickness.  The  following  fossils  have  been  recorded 
by  Abich,  Bayem  and  Simonovitch  from  the  cal- 
careous beds  o!  Karatuban  (N.  of  Akhalzik) : — 

Pectunculus   obovofus,    Laa.    (abundant);     Venus 
nutura^  Brongn.  (probably  K.  lugensis  according 
to  Oppenheim,  op.  cit^;    Cyrena  semistriata, 
Dedi.  (abundant). 
Natica  Deskayesit  Njrsi,   N.  helicina^  Brocchi  (abun- 
dant) ;  Ceritkium  piicatum^  Bn^.,  C.  margarita- 
ceum,  BiMchi  (according  to  Oppenheim),    C.  cf. 
troclUeare^  Lam.,  C.  eubeUgans^  Abkh  (abundant), 
C.    karatubani^    AWcfa;     Buccinum    costulatum, 
Brocchi ;  Eburna  Caronis,  BroBfii.  (abundant). 
In  an  earlier  paper  Abich  ^  gave  the  following 
additional   species,   which   were  possibly  erroneous 
determinations,    for   they   did   not   appear    in    his 
later  list : — 

Ampullina  ertissatina.  Lam;  Ceritkium  Cordieri^  Desb. 
Pectunculus  terAratularis^  Lam. 

In  the  uppermost  beds,  which  are  well  developed  in 
the  Telil  Dagh  in  the  Arzian  Mountains  (bounding 
the  Akhalzik  basin  on  the  W.),   Valentin  f  found 

^  MtM.  ma  iem  JBnk,  Vah.  k.-k.  geoL  SMhaaiist,  ttyy,  jj. 
t  Biricki  Smda$t.  JfUmf.  Get,  FnakTurt  a/M,  189X,  aji. 


Oligoceiu  429 

silicified  tree- trunks  on  the  W.  side  of  the  Ganfy  pass, 
near  the  source  of  the  Ajaris  river.  In  the  same  series 
Bayern^  also  found  silicified  wood  on  the  right  bank 
<>i  the  Koblian  river  opposite  Mokhe,  and  at  Gora 
(S.  of  Abastuman),  and  he  has  recorded  Kgnite  from 
Benara  close  at  hand. 

Lower  Oligocene  beds  occur  in  the  Pontic  region, 
between  the  Karchkal  Mountains  and  Jhe  Ardahan 
plateau.  Here,  above  Satlel  [i.e.,  just  oelow  Okro- 
baket],  in  the  gorge  of  the  Samjel-robat  river,  a  left 
tributary  of  the  Imer-khevi,  Batsevitch  f  found  **  grey 
marls  and  a  pale  yellow,  very  <:ompa<!l,  crystalline 
limestone  with  Harpactocarcinus  ^cp.  p.  42I7]»  Certthium^ 
Turbo^  ConuSf  &c.,  (Morals  and  spines  of  Cidarts.'* 

To  the  S.  of  Akhalzik  Abich  X  I^^  recorded  fucoidal 
beds  of  Upper  Oligocene  age  in  the  synclinal  valley  of 
Oshora  near  Aspinja. 

To  the  E.,  in  the  Tiflis  region  Fournjer  §  states  that 

'*  the  Oligocece  is  lepresented  by  a  very  thick  (ornuL^m  of  arenaceous 
ctajTii  aigillaceoui  slates  and  volcanic  tnflh,  breccias  and  cjn- 
glomerates.  It  forms  all  the  hills  of  the-e*wirons  of  Tiflis,  and 
contains  many  lignitic  beds,  /^.,  in  the  Vaghja  Dagh,  S.  of  Tiflis, 
frequently  accompanied  by  days  with  gypsnm.  The  argillaceous 
sandstones  contain  impressions  of  CAottdriUs  Afywiri.  Brongn.  .  .  In 
this  basin  of  the  Kur  thit  Flysch  fades  of  the  OUgocene  rests  con- 
coidaAtiy  on  th^  Upper  Eocene,  and  partidpates  in  its  folds,  which 
are  bent  over  to  the  N.  It  is  ev<:rywhere  overlain  discbrdantly 
by  Sarmatian  deposits.'* 

ITie  conditions  seem  to  have  been  slightly  less 
inimical  to  animal  life  on  the  W .  side  of  the  2)irula 

*  Gtai.  CMial.  TiJIis  Musium^  52,  54. 

t  Geol.  Resiarchis  in  the  Batum  ^  Artmn  Distmts^  Mat  Geol. 
Cauc.  Tiflis,  1887,  Ser.  s,  i.   las.    [In  Russian]. 

X  Ged.  Faruh.  in  dm  kmtk.  ZAmdem^  vL  Jj. 

{  Dtuription  gM.  dm  Cnncau  anirui,  i}8.  See  also  Abi^h,  Jf/erfu 
de  mu$  veyagis  tn  Trantcaueam  m  1864^  Bull.  Acad.  &d.  St  P^tersbw 
V.  (1865)  17  ;  Abich,  MiUh.  aus  dew  AnuJt.^  Verb.  k.-k.  geoL  Reichs^ 
aost,  1877,34. 


43^  Geology  of  Armenia 

dome  (Meschic  Mountains),  in  the  basin  of  the  K\  irila 
and  Lower  Rion,  viz.^  on  the  N.  side  of  tiie  Northern 
lk>rder*Ranges  of  Armenia;   according  to  Fournier'*' 

**  bene  die  Upper  Eocene  ZaMMM-beds  overlie  the  manganese  depodtiy 
s«nd  are  in  turn  covered  by  calaureous  sandstones  with  ftaf^ents 
gf  Cardimm  and  Qynma,  These  sandstones  are  ideniical  in  faaes  and 
iftiatigraphj  to  lime  with  Cardium  ataUnU  and  Lyrcna  usmiUrkM 
in  the  Akhaliik  basin  and  seem  to  be  OUgooene." 

The  next  occurrence  of  Oligocene  to  the  S.  of 
Akhalzik  is  the  totally  di£Ferent  facies  of  Ok ta  herd, 
a  few  miles  £.  of  Erivan.  The  strata  have  been 
folded  in  a  broad  anticline;  I  have  summarized 
Abich*s  observations  f  in  the  following  succession : — 

X.  Horizontal,  trachytic  conglomerates  (Pleisto- 
cene), unconformably  overlying 

2.  A  very  thick  series  of  light-coloured  marls 
(Upper  Oligocene?)  with  much  gypsum,  which  is 
quarried  between  Oktaberd  and  Envan. 

3.  Light  yellow  coral-limestone  of  inconsiderable 
thickness,  not  very  compacft.  Many  of  the  spectes  are 
identical  with  corals  from  Castel  Gomberto  and 
Oberburg : — 

Porites  minuta^  Reust ;  Rhabdophyllia  tenuis,  Rehss ; 
Ulophyllia  acutijuga^  Reuss,  u.  macro gyr  4,Reiiss ; 
Desmocladia  septifera^  Reuss;  DimotphopkylUa 
oxylopha,  Reuss;  Favia  confertissima^  Reuss; 
Agathiphyllia  conglobaia^  Reuss ;  Heliasircea  col- 
umnaris,  Reuss,  //.  Boueana,  Reuss ;  Prionastrcta^ 
gibhosa^  Abich;  Isastraea  elegans,  Reuss;  Ploco- 
phyllia  calyculata^  Reuss,  P.  jlabellata.  Reuss; 
Stylina  Suessiif  Reuss ;  Stylocoenta  lobato-rotunda- 

*  DesiripHan  giol.  du  Caucasi  cmiml,  184. 

t  Das  SUinudM  6v.,  M^m.  Acad.  Sci.  St  P^teisb.,  s^.  6,  Set 
math.  &  phjrs.  vii.  10 ;  see  also  his  CM.  Btob.  wm.  Kur  6r*  Anuus, 
16-si,  and  Butt.  Sac,  GM.  Frtma,  Ui.  s,  xai.  217,  an^  espedsUf 
Geo/.  ForuA.  in  dm  kauk.  ZSmderm^  ii*  S50-S64. 


OUgoe$ne  431 

ta,  Micb.»  S.  micyophthtttmai  Retm;  Phyllocoenia 
irradians^  M-E.  k  ]L»  P.  macrocantUt  Abich. 

4.  A  considerable  series  of  dark  grey»  argilla* 
ceous  limestones,  calcareous  sandstones  and  foramini- 
ieral  limestone*conglomerates ;  this  series  contains 
the  following  fossils : — 

PectufUulM  tugM$i%^  Lam.  (San^onini);  Crassa* 
tella  trigonula^  rndm  (Sangonini) ;  Cardium  per* 
versum^  Abich  (frequent;  near  to  C.  lallax  of 
Sangonini),  C.  puudo-fragile^  Abich  (very  simi- 


lar  to  C.  irapU^  Bioccbi);  Arctica  campressa^ 
Fvcht  ipmr.,  C.  Tytarrisi.  Sow.;  Venus  vindobonensis^ 
]iayef'(Vienna  basin),  V.  umbonaria^  Lam. 
Ampullina  maxima^  Gnit.  (abundant  in  calcareous 
sandstones  and  yellowish,  calcareous  itiarls), 
A.  crassatina^  Lam.  (less  frequent,  in  overlying 
dark,  calcareous,  concretionary  sandstones); 
Cerithium  Charpentieri^  Batt;  Conns  Mercati, 
Bioodii* 

5*    Eocene  with  Nummulites  intermedins,  dVlreh. 

These  beds  probably  extend  into  Persia,  for  Tletze 
found  Cerithium  Charpentieri^  iftc,  At  Siokuh,  S.E.  of 
Teheran. 

The  coral-limestone  of  Oktaberd  recurs  in.  the 
Daralagdz  region  at  Malishkent  (S.E.  of  Oktaberd), 
where  it  evidently  overlies  Eocene  and  is  overlain 
by  Miocene.  Here  Tsulukidse*  coUeAed  the  follow- 
ing corals : — 

Cyathophyllia  annulate^  Rcim  ;  Dasypkyllia  difor- 
mis,  Retm;  Calamophyllia  fasdcnlata^  Reust; 
Plqcophyllia  calycnlata.  Rant;  Stylina;  Stylo- 
coehia  liAata,  Mich.;  Trochoeyotkus  comutus,  Hatme. 

At  Gerger  (a  few  miles  E.N.E.  of  Malishkent)  this 


43^  Geology  of  Armenia 

coral-limestone  occurs  undec  similar  conditioas  and 
was  found  by  Tsutukidse  {l.c.)  to  contain  the  follow* 
ing  fossib : — 

Multtloculina. 
Milleporidce. 

PolytromaoM  BMardi.  Haime. 
Podobacia  prisca,  Reuss;   Heliastroea;    Phyllocoenia 
irradians,  M-E.  ft  H. 

A  higher  (Aquitanian)  horizon,  showing  a  littoral 
facies  was  found  by  Abich  *  at  Argachi,  at  the  S.W. 
foot  of  the  Kizilkaya  hills,  E.  of  Igdir.  These  heights 
have  just  escaped  being  overwhelmed  by  the  enormous 
lava-streams  which  proceeded  from  Kippgol,  the  N.W. 
outspur  of  Ararat.  The  steep  clifiEs  (pernaps  a  fault- 
scarp)  rise  679  feet  above  the  Araxes  plain,  and  show 
the  following  succession : — 

1.  Yellowish  sandstone,  alternating  with  fine 
and  coarse  conglomerates  fenclosing  geodes  of  limonite 
and  large  pebbles  of  felsite-porphyry),  with  bands 
of  brownish  red,  ferruginous  tufif — in  all  about  40  feet. 

2.  Yellowish,  rather  hard,  shaly  tuflb  with  plant- 
impressions,  alternating  with  fragmental  beds  (re* 
sombling  Bunter),  showing  no  distindt  bedding  and 
varying  in  texture.    The  plant-remains  consist  oi 

Fodocarpus  eocenicuSj  Unger. 

Agaoites  ararattea^  Abidi  (leaves  and  rhizomes); 

Cy pontes  canaliculatus^  Heer,  C.  Doucalionis,  Heer ; 

Phragmites  oeningonois^  Heer. 
Myrica ;  Salix  angusta^  Braun. 

3.  Reddish  brown  tuffs  with  the  following  fossils 
in  the  form  of  sharply  defined,  non -calcareous  casts : 

*  Doi  Sififuoh  6v.,  M^m.  Acad.  Sd.  St  PAenb.,  96r.  6,  Sd. 
math,  k  pXiys.  irii.  9,  69-89;  and  his  do/.  Forsch,  in  itm-  lunk. 
LUnitrn^  ii.  147,  251,  365,  t66. 


Oligocene  433 

Ostrea  (large,  smooth,  indeterminable);  Mytilus 
Haidingeri^  Homes  (very  abundant,  often  ver>' 
large) ;  Area  umh(mata^  Desh.,  A.  araratica^  Abich 
fbelonging  to  the  Noce  group) ;  Meretrix  (Pitar) 
tncrassata^  Sow.  (in  great  numbers),  M .  (Callista) 
erycinoides^  Lam.  (very  abundant) ;  Cyrena  unda- 
ta^   Abich   (very  abundant);    Saxicava  rustica^ 

Broochi. 

Cerithium  tricindiutn^  Brocchi,    C.  margaritaceum^ 

Broochi,    C.   pi6ium^   Defr.;    Fusus  burdigalensis^ 
Bast 

4.    Very  compa(5t,  conglomeratic  sandstones  or 
breccias,  with  many  fragments  of  felsite-poiphyry. 

According  to  Abich  ^  the  same  formation  recurs  in 
the  Kizilyeri  hills  (Kiziltash)  on  the  S.  bank  of  the 
Araxes,  at  the  Arpa-chai  confluence  N.W.  of  Kulpi. 
They  rise  400  feet  above  the  river  and  consist  of 

"  conglomerates  and  sandstones  with  a  femiginous,  calcareous  cement, 
and  are  overlain  on  the  S.  by  the  red  saliierous  marls  [Sarmatian]  of 
KulpL  The  sandstones  contain  distinct  traces  of  the  moliuscan  fauna 
of  K]2ilkaya  as  well  as  reedlike  grasses  and  dicotyledonous  leaves ; 
they  alternate  with  marls  chiefly  derived  from  felsite-porphyry.'' 

The  only  remaining  instance  of  Oligocene  in  Ar- 
menia is  recorded  by  Abich  f  in  the  Araxes  gorge  a 
few  versts  above  Julfa,  about  halfway  between  Julfa 
and  Darosham.  Here  the  horizontal  Nummulitic 
limestones  (overlying  Permo-Triassic)  are  conformably 
overlain  by  Oligocene,  consisting  of  a  very  thick  bed 
of  an  ash-grey,  calcareous  sandstone  passing  upwards 
into  brown  sandstones  overlain  by  Sarmatian  sand- 
stones and  conglomerates.  Intercalated  in  the  Oligo- 
cene sandstone  there  occur  rather  compad,  calcareous 


*  I>as  SieinstjiM  ^c,  M6n.  Acad.  Sd.  St  Pftersb.,  s6r.  6,  Sd. 
math.  &  phys.  vii.  9,  10 ;  and  his  Giot.  Fifnch.  in  dm  kauk.  Linditn^ 
tL  148.  Id  his  luge  map  to  the  latter  work  Abich  has  obviously 
made  the  mistake  of  transposing  the  names  Kizilyeri  and  ELisilkaym. 

t  Das  SkifuaU  A»c.,  67,  68. 


a? 


434  Geology  of  Armenia 

beds,  i~i|  feet  thick,  enclosing  a  number  of  shells 
(especially  Gastropods),  which  are  all  changed  into 
<*oarse-grained  calcite,  and  difficult  to  determine 
owing  to  their  fragmentary  and  altered  condition. 
Absence  of  all  Foraminifera  characterizes  this  group. 
Abich  distinguished  the  following  fossils : — 

Ostrea  ^mall,  smooth,  indeterminable);    Venus; 
Astarte  (small). 

Natka  glaucina^  (?)  Linn.;  Cerithium  pltcatum^ 
Bn«.,  C.  margaritaceum^  Broodii. 
Freeh  and  Arthaber,^  during  their  eKamination 
of  the  Araxes  gorge  between  Negram  and  Julfa,  seem 
to  have  overlooked  this  instance  of  Oligocene  recorded 
by  Abich,  for  they  expressly  state  that  no  Oligocene 
beds  occur  in  this  district  of  the  mid*.\raxes. 

^  jDos  Palaatmum  in  Hochtumntiim  6*  Ptnien^  Bcitr.  PaL  CEslr.- 
Ung.  xii.  i8i. 


CHAPTER     XXIII 


T££   MIOCENE   OF  ARMENIA 


The  Miocene  of  Armenia  consists  of  two  stroQgiy 
contrasted  members,  separated  by  a  marked  uncon- 
formity,  indicating  an  epoch  of  mountain^making. 
At  the  commencement  of  the  Miocene  a  more  wide- 
spread transgression  occurred  in  Armenia  than  ever 
before.  Even  regions  such  as  the  Palaeozoic  plateau 
of  Urmi,  which  had  remained  dry  land  during  the 
nearly  universal  Cretaceous  transgression,  now  sank 
beneath  the  waves.  The  N.  half  of  Armenia,  which 
had  been  covered  by  the  Oligocene  sea,  was  now 
upraised  to  form  the  N.  shore  of  the  Miocene  Medi- 
terranean in  these  latitudes  and  (together  with  the  N. 
Border- Ranges)  adled  as  an  effectual  land-barrier 
between  the  warm  southern  waters,  abounding  in 
corals,  and  the  colder  northern  sea  which  washed 
the  base  of  the  Caucasus. 

Immediately  subsequent  to  the  Lower  Tortonian  an 
era  of  mountain-making  took  place  in  Armenia, 
accompanied  by  extensive  intrusions  of  ultrabasic 
rocks,  indicating  the  deep-seated  nature  of  the  crustal 
disturbances.  The  whole  of  Armenia  became  raised 
above  the  sea,  and  the  lagoon  facies  of  the  Upper 
Miocene  (Sarmatian)  was  widespread,  resulting  in  the 
deposition  of  rock-salt  and  gypsum  in  a  series  of 


436  Geology  of  Armenia 

red  sandstones,  marls  and  conglomerates. 

I.      LOWSl  MtOOSMB. 

The  Lower  Miocene  limestones  are  particularly  well 
developed  in  the  basin  of  Lake  Urmi,  and  their  fauna 
has  been  recently  examined  by  Messrs.  Gregory  and 
Newton*  on  material  brought  by  Giinther  from  the 
coast  and  island  of  this  lake ;  since  they  have  at  the 
same  time  revised  and  corre<5led  many  of  Abich*s 
previous  determinations  f  it  is  convenient  to  begin 
with  an  account  of  this  distridl. 

The  majority  of  Abich's  specimens  were  obtained 
from  the  largest  island  Koyundaghi,  but  a  few  were 
coUedled  on  Ishakdaghi  atid  the  Shahi  peninsula. 
Gunther's  specimens  came  exclusively  from  Koyun- 
daghi. 

I  have  summarized  the  succession  on  Koyundaghi 
as  follows : — 

1.  Lower  Tortonian.  Compact,  rather  porous 
limestone,  occasionally  like  travertine^  containing 

Alectryonia  (Ostrea)  Virleti^  Desh.  [=0,  Virleti  and 
0.  excavata  of  Abich],  with  Thamnaroea  poly- 
morpha^  Abich,  attached,  0.  lamellosa^  Brocchi. 

Latirus  crispus. 

2.  Upper  Helvetian.  This  division  shows  no 
sharp  demarcation  from  No.  i.  Nos.  i  and  2  together 
are  at  least  ^00  feet  thick.    It  falls  into  2  sections : — 

(a)  Light  yellow  or  white  limestone,  more 
or  less  compadl,  with  spathic  portions,  capable 
of  high  polish,  sometimes  partially  translucent  and 
recalUng  the  yellowish  travertine  of  Azerbaijan.    It 

*  JomrfL  Lbm.  Sac.  xxvii.  4x8. 

t  Das  Sidmalt  6v.,  Mem.  Acad.  So.  St  Paenb.,  96r.  6,  Sd. 
math.  &  jdifB.  vii.  15-27,  31-67 ;  and  his  Geoi,  Fanch.  in  itn  ktmk. 
JUutdmtf  u.  274,  tS3« 


i 


Tjmmr  MiouM  437 

contains  the  following  fossils : — 

Poly^tam$lia  quat$rfunctata^  KHA. 

Ofhiulla  Defranceit  M-E.  &  H.  (Heltastrcea  Guettar- 
di^  Dcfr.,  of  Abich),  0.  Guntheri^  Qttgtxy  (H.  De- 
ftanui^  M..E.  &H.,  of  Abich);  Phyllangia  al- 
MOfam,  Catullo,  P.  grandis,  Reuss ;  Prionastrcea 
irregularis^  Defr.;  oolenastraea  turanensis^  Mich, 
(probably  includes  S.  astraites^  Abich) ;  Astrangia 
cf.  prince^f  Reuss;  Astraea  Ellisiatia,  Defr.,  A. 
trrandistetla^  Abich  (probably  a  Phyllocomia)  \ 
P.  Archiaci,  11.-E.  &  H.;  Cyathamarpha  conglobata, 
Reuss,  C.  cf.  gregaria,  Catullo;  Thamnaraea 
(Parites)  polymarpna^  Abich,  T.  leiophylla,  Reuss. 

Ceri0para  anomala,  Abich,   C.  globulus ;    Diastopara 
gemmifera^    Abidi;     Metnbranipara    fenestrata, 
Eiehw.;    Onychocella  [Cellepora]  gracilis,  Miinst., 
0.  depianata ;  Polytremads  spongiasa,  d'Oib. 

Spandylus  bifrtms^  Munst.;  Pecten  urmiensiSf  Abich, 
P.  sfi&ttrmiimsis,  Dep6ret,  P.  TaumcUi^  Serres« 
P.  lychnulus^  Fcntumes,  P.  coiff^exocostotfis,  Abich 
(the  larger  form),  P.  ftabelliformis^  Brocchi, 
P.  {Amussiopecten)  burdigalensis^  Lam.,  P.  {Fla- 
belllbecten)  cf.  Besseri^  Andiej. 
(b)  (jtreenish  grey,  coarse-giained  molasse-sand- 
stone  with  the  following  sea-urchins : — 

Clypeaster  GUntheri^  Gregory  (C.  altus^  Lam.,  var. 
turritus^  Philippit  oi  Abich) ;  Echinolampas  com- 
piUmatus^  Abidu 

3.  Lower  Helvetian.  Concretionary,  brownish 
yellow  limestone,  bresdcing  in  great  slabs  similar  to 
the  concretionary  Faluns ;  it  contains  the  following 
fossils : — 

Corab  (not  specified  by  Abich). 
PeHen  canvexocostatus^  Abich  (the  smaller  form), 
P.  subopercularis^  AUdi,   Chlamys  (Mquipecten) 


438  .  Geobgy  0/  Af$B€ma 

aff.  MalvifUB^  Dubois. 
Tutritella  Arcktnudis^  Bnmip^f  T.  marganta^  AbidH 
T.  turriz^  d'Orb.,  T.  o^.  rotifera^  Desh.,  T.  j^VMEa* 
^O,  Homes  iru; 

The  following  species  were  coUedled  by  Gfinther 
(Z.C.)  in  a  grey,  marly  rock  and  probably  come  from 
this  horizon : — 

Ostrea  pseudodigitalina^  Fnchs ;  Cardita. 

Strtmbus  cf.  Banelli^  Biongn.;  Canus. 

4«    Fragmental  deposits,  varying  in   thicknesSi 
derived  from 

^.    Carbonifbrous  Limestone  (p.  299). 
Abich*  has  also  stated  that  on  the  neighbouring 
island  of  Ispir  *' light  yellow   Qy^ast^r-Timestones 
overlie  Carboniferous  Bdlerophon  and  Fuitdina  lime- 
stone." 

The  determination  by  Messrs.  Gregory  and  Newton  f 
of  Giinther's  fossils  from  Guverchin-kala  (Gugarchin- 
kala),  on  the  west  coast  of  Lake  Urmi,  gives  an 
aulditional  value  to  the  sedlion  which  Loftos^  has 
recorded  of  this  picturesque  locality.  The  castle-rock 
is  about  400  feet  high  ana — according  to  Giinther  {Lc) 
— is  composed  of  Miocene  strata  resting  horizontally 
(7^  E.S.E.)  on  red  granites  traversed  by  felspar  dikes. 
Loftus*  sedlion  is  as  follows : — 

1.  '*  Compact*  hszd,  crystalline,  white  limestonep  beooming 
concretioDar)-  downwards,  afterwards  marly. 

2.  Light  Uue  marl,  with  hard  flesh  coloured  flints  a&d  nodoles, 
and  irregular  fragments  of  limestone.  Abundant  corals  occur  m  ^Blm 
and  in  layers,  below  which  are  numerous  Cij^atkr^  SMmdMmfas^ 
Peeten^  Sitpuia  and  casts  of  univalves."  This  u  evidently  the  Upper 
Helvetian  bed  whence  Gunther  obtained  the  foHowuiig  fosnb : — 

*  Verh^  k.^.  gfol.  Xekhsanst^  1877,  67. 
t  Joum.  Linn.  Soc.  xxvii.  418. 


Lower  Miocene  439 

Clypiaster  GuntJken\  GregOfy,  C  MarHni^  Desmoulins  (one 
specimen  from  Loftus'  collection  in  the  British  Museum); 
C  aff.  impmaiis^  Midi. 

PecUn  canvexoofsiatus^  Abich,  P.  urmienstSf  Abich,  P.  sub- 
mrmiensiSf  Dep6ret  (in  drab  limestone). 

3.  Compact  mass  of  highly  crystalline  nodules  in  hard  marl. 
Nos.  I9  3  &  3  together  amount  to  250  feet 

4.  fine,  reddish  ^vel  or  coarse  sand-conglomerate,  much 
hardened  and  full  of  fossil-ftagments  x8  feet 

5.  Friable,  yellow  sandstone  of  very  fine  grain,  with  fragmentary 
fossils ;  passes  into  XS  feet 

6.  Hard  ancl  compact,  grey,  marly  limestone,  filled  with  coraU 
and  casts  of  shells."  6  feet 

Nos.  4,  5  &  6  are  probably  on  the  same  horizon  as  the  Lower 
Hdyetian  Turriteila'limestont  of  Koyundaghi,  p.  437. 

7.  "  Hard,  reddish  marls,  abounding  in  shells. '*         ax  feet 
Tins  is  probably  the  bed  whence  Gtinther  (/.c.)  obtained  reddish 

brown  casts  of  the  following  fossils : — 

Viwu  Aglaura^   Brongn.;    Aferefnx  fersiensis^   Newton,    M. 

(Piiar^  cf.  incrassata^  Sow.  (Oligocene,  cp.  p.  433). 
Pyruia  cmgulaia^  Homes;  Cassis. 
These  fossils,  accordmg  to  Mr.  R.  Bullen  Newton  (Lc.)  **  indicate  a 
Burdi|pslian  or  even  Aquitanian  horizon." 

8.  ''  Brownish  yellow,  friable  sandstones  with  several  thin  layers 
of  gravel  and  conglomerate  consisting  of  variously  sized  rounded, 
pebbles.''    Probably  a  basement-conglomerate.  xoo  feet 

Total     4x0  feet 

On  the  S.  shore  of  Lake  Urmi  Abich*  established 
the  presence  of  the  Upper  Helvetian  coral-limestone 
by  means  of  the  fossik  coUedled  by  Khanykoff  from 
the  Yaghyn-Aly  hill  on  the  Jagatai  and  Tantivi  delta. 
The  rock  is  a  compa(5l,  crystalline,  marmoraceous 
limestone  rich  in  Foraminifera  and  Bryozoa,  with 
larg^e  masses  of  coral;  it  contained  the  following 
fossils : — 

Ofbicella  Defrancei^  M-K  &  H.  {Heliastrcea  Gnettar- 

*  Das  StdnsalM  6v.,  M&i.  Acad.  ScL  St  Ptenb.,  sdr.  6,  Sci. 
math.   &  phys.   viL   96,  s} ;   and  Ut  GwL  Fcnth.  in  im  kmtk. 


440  Geology  of  Armenia 

di^  Defr.i  of  Abich),   0.  Guntheri^  Gngocy  (H.  Di- 

franqei,  M.-E.  &  H.»  of  Abich) ;   Astrcea  EllisianQf 

Dffr. 
Ceriopora  palmatn^  d'Orh. 
Clypeaster  Scillve^  Detmoolint^ 
Pecten  (simplex)  Holgeri^  Geinito. 
Haliotis  Pkilberti^  de  Serret. 

These  beds  are  probably  indicated  by  Wagners 
statement*  that  the  cliffs  at  the  S.W.  end  of  Lake 
Urmi  consist  of  a  limestone,  which  is  *^  compadi,  hard, 
light  in  colour,  with  many  fossils,  especially  Pecteu 
and  Terebratula^'  [cp.  the  Terebratula  minor  bed  of 
Kardarich,  vide  infra]. 

These  limestones  pass  into  conglomeratic  beds 
towards  Urmi.  At  Seir  hill,  6-8  miles  W.  of  Urmi, 
Giinther  j  found  the  following  Upper  Helvetian  fossib 
in  a  light  brown,  calcareous  sandstone  in  a  con- 
glomerate : — 

OHncella   GUntheri,  Gregoiy,     0.    Haimei^  d'Axch.; 

Solenastrwa  turonensis^  Mich. 
Pecten  (Oopecten)  rotundatus^  Lam.;  Cardium. 

This  sandstone  is  probably  the  same  as  No.  2  In  the 
following  sedion  given  by  Loftus^  of  the  *^Seyr 
Dagh,"  which  he  approadied  up  the  Shaher  valley, 
proceeding  W.  from  urmi : — 

I.  Loose  gntvd  or  gravel-coQglomerate  with  adcareous  ptite, 
chiefly  of  pebbles  from  the  froctier-chain ;  oonfoniuiblj  ovedjing 

a.  Yelloirish  or  brownish  saadstooes,  dipping  (*  N.E,,  alter 
Bating  with  conglomerates,  containing  pebUes  of  bhie  and  white, 
altered  Umestone^  quarts  and  micaceous  sandstone^  bat  none  of 
igneous  rodcs. 

3.    Maris  and  impure  limestones,  alternating  with  sandstones. 

It  was  doubtless  from  the  same  bed  that  Abich  § 

*  RdumiA  Artimf  iL  118. 

t  /mmL  JUnm.  Set.  xzviL  414,  443. 

t  Qfuirt  Jmm.  CM.  Sec.  zi.  S98. 

{  DMt  Skimufb  &x.,  M6n.  Acad.  ScL  St  Pftersb.,  sfr.  6p  Sd. 
math,  ft  phys.  viL  tj. 


i 


Ijnmr  Mioeem  44Z 

received  from  KhanyJkoS  a  specimen  from  near  Seir 

"of  aaeitven^  fine-fruned  aikote,  widi  more  calcaieops  tbaa 
dkeoat  oemenl,  containing  aa  angular  fragment,  4  tnehes  laige» 
of  die  maraMxaoeoas  Bijoioa-limealone  of  huut  Urmi,  with  a  variety 
of/WteiMfAtf.'' 

Near  Urmi,  in  a  ^'oompad,  oeam-coloured,  Fore* 
miniferal  limestonei*^  Gfinther^  found  the  Helvetian' 
Pecten  Beudanti^  Baat 

Fragmental  beds  similar  to  those  near  Urmi  occuiT 
also  at  Khoi  further  N.  According  to  Abicht  they 
oonsist  of  confusedly^  bedded,  reddish  brown  and 
greenish  grey  marls,  mixed  with  gypsum  and  pebbles, 
and  containmg  large  blocks  of  fine-grained,  dolomitic 
coral-limestone,  enclosing  casts  of  the  following : — 

Ofbicella  GUntheri^  Gregcvy,  (Hflia^troBa  DefrancH^ 

M.-B.  ft  H.,  nur.  X  of  Abich),  0.  Haimei,  d'Aidi., 

{H.  Defrantiiti  var.  a  of  Abich) ;  Astrcea  Ellisiana^ 

Deb.  (with  burrows  of  LAnodamus  lithophagus, 
linnA 

'Haliotis  manilifera,  BonelE;   Cyfrtea  inter  Uparina 
&  Brouhi ;  Cmius  cf.  mediterraneus^  Brocchi. 

This  list  of  derived  sjpecimens  is  only  of  value  in 
showing  the  close  proximitv  to  Khoi  of  the  Upper 
Helvetian  coral-limestone ;  its  fragments  are  probably 
derived  from  a  limestone  at  the  base  of  the  hill  of 
Khoi,  or  from  the  ranges  on  the  W.  and  S.  of  the 

flain.     Loftus,^  however,  was  of  opinion  (mistakenly 
think)  that  some  of  the  corals  in  the  coarse  gravel 

«« undoubtedly  lived  during  the  depotitton  of  tha  giavel;  others 
are  decidedly  derhfed  from  an  old  bed  of  crystalline^  yellow  limestone^ 
and  are  much  rolled  and  worn.  Large,  waterwom  blocks  of  t&e 
fimestone  lie  at  the  base  of  the  hill.  .  '.  At  the  base  of  the  hill  the 
hard»  crystalline  limestone  is  filled  with  fossils,  especially  beautifuUy 


^  /oMrrn.  Lbm.  S0C  xxA  441. 

t  Gici.  Fanch.  in  dm  ktmk,  Ububm^  &.  syt-syj,  310. 

X  Quart.  Jaum.  G»U  So€.  xL  308.  309. 


44^  Geology  of  Amufda 

pmerved  oonk  wich  cattt  of  dieOs,  amoQg  wkidi  tie  ^fm,  UUk§^ 
4$mm  and  Trackm.    The  oonb  «gne  giectelly  villi  apedei  Jboa 
Gttveirhinkah.'* 

This  brownish  yellow  limestone  was  also  found 
by  Loftus*  to  flank  the  S.  border  of  the  plain  on 
the  way  to  Oilman. 

A  great  part  of  the  mountainous  country  between 
Khoi  and  Ararat  seems  to  consist  of  Miocene  lime* 
stone.  In  particular  Abich  has  investintted  the 
region  SJE.  of  Bayazid,  in  the  Gedargan  l3agh»  and 
the  valleys  of  Maku,  Taoali  and  Zangimar. 

Near  the  Karakilissa  monastery  (^577  feet),  at  the 
E.  end  of  the  Kurbittu  valley  (a  right  tributary  of 
the  Zangimar),  Abich  f  has  established  tiie  following 
succession,  which  also  holds  good  feu:  the  Zangimar 
and  Tanali  valleys : — 

1.  Red  sandstones  and  marls  (Upper  Miocene). 

2.  Fine-grained,  compadt,  light  yellowish  lime* 
stone,  capable  of  taking  poli^;  occasionally  dolo- 
mitic  (i|  %  of  magnesia) ;  corals  form  reef-like  masses 
and  consist  of 

AttrcM  BUistana^  Defr.;    PrionastroM  irreguUms, 

$.  Conglomerate  of  shell-  and  ooral-fragments, 
with  an  arg^illac^us,  sandy  intercalary  layer.  Some^ 
times  there  is  a  gradual  passage  between  2  and  4. 

^  A  series  of  calcareous  marls,  associated  with 
greenish  grey^  calcaraous  sandstones,  dry,  san<fy 
marls  and  ash-grey  clays.  In  the  dark  aey^  slaty, 
very  compadfc,  calcareous  marls  of  Karakilissa  Abioh 
found  the  following  fossils : — 

QuinqudoculitM  cf.  ahmfioMO^  tfOib.;  RAuUna  d$^ 

^  QfUHf.  /mm.  Gmt.  Ar.  iL  308. 

\D0m  SulmtU  6%.,  M6n.  Acad.  Sd.  St  FCMrik,  stir.  (,  Sd. 
math,  fcphyi.  viL  ly,  it,  ja,  |j,  rf,  m^  48,  s>,  «it  6a,  66;  aad 
hk  Cfmt.  F(mck.  tm  im  hmk  Zimbm^  &  a6  j«a70. 


J 


444  Geology  of  Armenia 

(b)  Echinoid  limestone  with 
Quinqueloculina  cf.  akneriana^  d'Oib.;  RcbMna  de^ 

Pressa^  Blich. 

ClypeasUr  GUntheri^  Gttmj  (C.  altus^  hun^  yv. 

turritus^  Philippii  of  ADichj;    Schizaster  Genet. 
Sismonda. 

4.  Basal  conglomerates,  overlying  Carboniferous 
and  Devonian  limestoj^es  not  far  S.  of  Little  Ararat 

The  range  on  thef  N.  side  of  the  Karakilissa  or 
Kurbittu  valley  is  called  the  SogmanaRra  Dagh, 
with  escarpments  facing  the  Araxes  valley.  It  is 
continued  to  the  N.W.  through  the  Bashkent  Dagh 
into  the  Gedargan  Dagh  (between  Basyrgan  and 
Bayazid),  which  bounds  the  valley-plain  of  Bayazid 
on  the  S.  All  these  heights  are  composed  of  the  same 
Miocene  limestones,  which  here  overlie  Nummulitic 
Limestone,  according  to  Abich's  observations.^  The 
succession  is  as  follows : — 

I.  UrpiB  MiociHB : — red  sandstones  and  marls,  ooconing  on 
the  Bayasid  side  of  the  pass  (79S8  feet)  over  the  Gedargan  Dagh. 

s.  Lowut  MiocnnK  (tipper  Helvetian) :— Coral  and  Brfosoa 
limestones.  Thej  <fip  near^  vertiodljr  above  Bayasid,  Imd  seem 
to  have  been  thrust  over  the  Nummulitic  limestone.    Thejr  contain 

Bilo€9dinai  Tniantlimt. 

jOMastrMt  EUisuiita^  Defr.  (in  reeMike  masses)^  LUkamm 
iwsMjrtfy  !&«•&  at  n* 

3.    EooBKx : — Nummulitic  limestone  (p.  4rs). 

To  the  W.  of  BaVazid  these  Miocene  limestones- 
according  to  Abich  f — occur  in  the  basin  of  the  Upper 
Murad,  near  Diadin,  '*  overlying  Cretaceous  rocks,  and 
both  series  are  much  traversed  and  dislocated  by 
labrador-rocks  and  amygdaloidal  rocks  rich  in  calc- 
spar/*    Wagner  j:  gives  the  additional  information 

*  Das  StdmsaU  6v.,  M6n.  AouL  ScL  St  PAersb.,  s6r.  6,  ScL 
math.  &  phys.  viL  7,  i8»  19,  js,  44,  49. 

t  BwO.  fkys.'ma^.  Aud.  SeL  St.  Pfitra.  nv.  141. 

X  JUiu  muk  t^artiin^  L  191. 


IjHver  Miocene  445 

that  these  limestones  between  Diadin  and  Bayazid 
contain  Pecten. 

llie  same  series  seems  to  occur  in  the  north  of 
Azerbaijan,  for  the  Shagadi  Mountains  (just  N.  of 
Tabriz)  are  said  by  Fraser*  to  consist  of  '*  loamy 
sandstone,  compadl  limestone  and  a  shell-limestone 
with  Pecten^'*  rising  up  in  heights  above  the  saliferous 
marls. 

On  the  N.  side  of  the  Araxes  valley  the  Miocene 
limestones  do  not  reach  any  great  development,  yet 
the  Coral  and  Bryozoa  liviestone  is  still  100  feet  thick 
at  Malishkent  in  the  Daralagoz. region.  Here  Abichf 
has  established  the  following  succession : — 

I.    Compact,  fine-grained,  brown  limestone,  a  true  coral-reef  with 

Qum^tUiocu/iMa  rf.  Akmriana^  d'Orb. 

Cladacara  mamfuUUa^  Mich.;    TkuPtmiia;    Tkamnaraa  leia- 
f^Udt  Reius. 

Cifwpora  palwiaia^  d'Orb, 
3.    OuoocBNi : — Shaly  mark  and  coral-limestone  (p.  431). 

To  the  S.W.,  on  the  N,  slope  of  the  Gyneshik 
valley,  Abich:}^  found  a  limestone  very  similar  to 
that  of  Malishkent,  but  thicker  and  impure.  It 
contains  numerous  Foraminifera  and  Bryozoa  together 
with  Clypeaster  and  other  Echinoids,  and  uncon- 
formably  overlies  Devonian  limestone. 

Still  further  W.,  on  the  Jafarlu  Dagh,  between 
Nurashin  and  Chanakhchi,  Abich§  again  refers  to 
this  Helvetian  limestone,  as  follows : — 

X.  White,  marmoraoeous  limestone,  with  the  appearance  of 
an  actual  cond-reef.  It  consists  of  Foraminifera,  Bryozoa  and 
mainly  of  Corals  (Cladocora  articuhia^  Abich,  Paritida^  ^c.\  and  it 

*  Travels  (f»  Advenhtres  in  ihe  Persian  provinets^  i8s6. 

t  Das  SUinsalz  6v.,.  Mem.  Acad.  Sd.  St  Ptersb.,  wts.  6,  Sd. 
math.  &  phys.  vii.  14,  36,  43,  47,  48. 

X  IbU.  14. 

§  IHd.  14.  38,  3J. 


446  Geohgy  of  Armenia 

alio  ooDtains  a  ftw  isolatad  NmmtmUiiiS  imtirmtdimi  d'Arch,  wUdn-io 
mj  opinion— aie  probably  derived  tpedmeiif. 
%•    EooDfK  : — Nammiilidc  Limestoiie. 

The  Miocene  area  has  been  considerably  extended 
by  mv  observations  in  Turkish  Armenia.  It  is  parti- 
cularly well  exposed  in  the  cli&  of  Lake  Van,  and 
would  repay  a  detailed  investigation. 

The  Miocene  limestones  form  a  nearly  E.-W.  block 
of  high  downs,  breaking  off  in  tliffs  on  the  W.  coast 
of  Lake  Van^  and  show  a  conspicuous  white  escarp- 
ment facing  S.  I  traversed  this  block  firstly  on  the 
way  from  Melaskert  to  Akhlat  (pp.  95-99),  and 
secondly  along  the  coast  from  Akhlat  to  the  base 
of  Sipan  (pp.  156-166,  172).  My  observations  on  this 
Miocene  zone  may  be  summarized  thus  (pp.  165,  x66) : 

I.    LowBK  ToRTOMXAM : — ^Whitc  limestone,  rather  trftvertmoas  in 
•ppeanmoe,  with  Akaiy0mm  (Ostrm)  tVrUti,  Desh.  (p.  z6i). 

9.      UPFBlt  HSLVlTEikH:— 

(a)  Compact  Coral-limestone  with  the  foDowing  fossils : — 
OrUalla  Deftmrnti^   M.-E.   &  H.;    Hdiasirtm\    SoUtmOfm 

tufrnmsit^   Mich.;   SiyUfk^ra  smh^tHcmlai^^  Renss;    Tkam^ 
marmm  poipmarpka^  Abich* 
^Om  urmUmU^  Abich  (pp.  169,  163). 

(b)  Cream-colotned  limestone  full  of  LMatkatrndaH  ramauUd* 
nrnm^  Reuss,  and  with  numerons  Foraminifers  and  Bryocoa  sach  as 

MilMina  \  ThshUtma ;  Hiiirasi^Ma  i  Amfhisi^um  cf.  Mnuri' 

nm,  d'Oib. 
Cm9f0m  anvmaUf  Atrich ;  Rdtpora  (pp.  97,  163). 

3.    Lower  Hblvktiam  : — 

J  a)  ConMimestone   with 

Cladocora asHctdaiaf  Abich;  TlvtkogmUhff  cf.  nartmm,  Doncsa 
(pp.  160,  165). 

(b)  Basal»  red  sandstones  and  conglomeiates  of  Akhlat  00 
the  S.  of  the  limestone-sone  (pp.  98,  145,  156)1  and  the  greenish  grqr 
sandstones  and  conglomerates  ct  Akhyiran  (p.  95),  and  of  ShctA 
(p.  17s)  on  the  N.  of  the  high  downs. 

The  Miocene  limestone  of  Adeljivas  disappears  from 
view  beneath  the  great  volcano  of  Sipan,  l)ut   it 


Lower  Miocene  447 

appears  on  the  £.  side  of  the  mountain,  extending 
to  Arjish.  The  observations  which  Loft^*  made 
on  this  part  of  the  coast  clearly  refer  to  the  same 
Miocene  limestones.  Between  the  Kanzag  plain  and 
Aksuraf  (Axaraf)  he  crossed  a  limestone  ridge,  which 
showed  the  following  succession : — 

I*    White  limestone. 
s.    White  ckys  #ith  flteUm. 

3.    Loose»  grey  sandstone  (q).  the  greenish  grejr  sandstone 
of  Shebu,  No.  3  (b)  of  p.  446). 

^  On  the  N.  side  of  Aksuraf  lies  another  ridge  of 
limestone:  *'upon  this  are  strewed  numerous  flint 
pebbles  derived  from  the  limestone  and  containing 
Operculina**  Another  low  ridge,  which  projects  into 
Lake  Van  W.  of  Madravank  (Madghavank),  is  a 
''compadl,  white  limestone  with  a  few  casts  of 
corals."  Still  further  N.,  to  the  W.  of  Erishat 
(Irishat),  Loftusj  noticed  a  'Mow  range  of  white 
limestone/*  which  f:his  section  clearly  shows  to  be 
identical  with  the  above. 

On  the  E.  coast  of  Lake  Van  the  limestone  cliffs 
were  clearly  once  in  continuity  withthe  Miocene 
series  of  Adeljivas,  but  it  requires  investigation 
to  demonstrate  this  conclusively.  Mr.  H.F.B.  Lynch, 
however,  collected  in  1895  a  few  specimens  on  this 
side  of  the  lake,  and  Professor  T.G.  Bonney,  F.R.S., 
who  examined  them,  has  kindly  allowed  me  to 
publish  his  descriptions.  The  rock  from  the  little 
island  of  Ktutz  is — ^according  to  Professor  Bonney — 

*'a  rather  compact,  very  pale  cream-coloured  limestone,  with  some 
small  cavities  and  indications  of  organisms.  Under  the  microscope, 
I  have  identified  Gichtgtrina^  firagments  of  cakareons  a]g»  (?  Litho^ 
ihamnioH)^  and  some  of  a  bivalve.  One  piece  has  a  tubular  aspect 
recalling  that  of  Giroanella.  Organic  fragments  are  evidently 
numerous  but  are  difficult  of  identification." 


*  Qpuurt  Jwnu  Gtal.  Sot.  xL  3so»  jsi. 
t  Op.  iU.  3x7. 


44^  Geology  of  Armenia 

This  limestone  probably  belongs  to  the  same 
horizon  as  the  Li^AotAaifinioif-limestone  which  I  found 
in  the  great  escarpment  N.  of  Akhlat  (p.  97,  and  2  (b), 

P-  44^)- 
A  somewhat  similar  specimen  comes  from  Toprak 

Kala,  a  limestone  height  overlooking  the  city  oi  Van : 

''  Tlie  rock  is  a  compact^  Tcrjr  pak  buff  limftone  vith  tnoet  of 
minute  oiganitnu.  Under  the  mioroecope,  tbe  rock  pravct  to  be 
full  of  frigments  of  oiganitmi,  among  which  I  find  tivo  fonns  of 
calcareous  a^gB,  tubular  masses  suggestive  of  GirwameUa^  and  frag- 
ments of  Foraminifera.  I  cannot  identify  with  certainty  anj  of  t£« 
last-named,  though  one  much  resembles  Timforus.  There  is  also 
a  fragment  of  a  I^meOibranch.'' 

The  limestone  of  the  castle-rock  of  Van  not  impro- 
bably belongs  also  to  the  Miocene  series.  Professor 
Bonney's  description  of  the  rock  is  as  follows: — 

"A  pale-ooloured  limestone,  mora  conqwcl  and  "steatitic Mocking 
than  the  Ktuts  limestone,  very  like  some  of  those  in  the  Jura. 
Organic  fragments  are  numerous,  but  most  of  them  are  small^  the 
most  readily  recpgnised  being  Foramiiufera.  Among  these  are 
Opiftulina^  k^taita,  and  a  fragment  .of  (?)  (^iUdyptm^ 

The  limestone  of  Akhtamar  (a  small  island  off  the 
S.  coast  of  Lake  Van)  shows  a  somewhat  different 
charaAer  from  the  preceding,  and  may  ^  perhaps 
belong  to  an  older  series;  its  close  proximity  to 
the  metamorphic  axis  of  the  Taurus  is  a  point  of 
some  interest.    Professor  Bonney  describes  the  rock  as 

'*  A  compact,  grey-buff  jimestone.  Under  the  microeoopei  it  conabts 
laigely  of  little,  rather  ovoid,  daifc,  j{ranutar  bodieSt  about  0.01  inch 
long,  set  in  a  minutely  granular  matnx  of  calcite,  tiaversed  by  slqjhtly 
coarser  veins  of  the  same.  There  are  also  a  few  Foramiiiiferav 
chiefly  nxhOaria." 

In  the  light  of  these  descriptions  it  is  probable  that 
the  white  or  pale  yellow  limestcmes,  which  Loftus^ 
traced  between  Van  and  Lake  Archag  and  further 
E.  are  also  Miocene  in  age,  especially  since  his  setSlions 
indicate  their  close  similarity  to  the  timestone  he 

*  Qmui.  /mm.  CM.  Soe.  n.  31 1-31  j* 


Lower  Miocene  449 

found  on  the  N.W.  coast  of  Lake  Van,  near  Aksuraf 
(p«  447).  He  states  that  the  white  limestone  of  the 
Ak  kspri  Dagh  (adjoining  Toprak  Kala),  which  has 
an  escarpment  facing  Van  plain,  joins  on  to  the  W. 
extremity  of  the  Kuzulje  (Kizzuljah)  Dagh  fW.  of 
Lake  Archag).  This  range  is  also  composed  ot  white 
limestone,  and  an  outer  ridge  of  white  limestone  lies 
on  the  N.  side  of  the  Varak  volcano  (E.  of  Van). 
Several  low  limestone  spurs  also  extend  from  Varak 
into  the  Anzub  plain  (S.  of  Lake  Archag).  At 
Derkashen  (on  the  S.  bank  of  the  Mermid  river, 
almost  due  N.  of  Van)  the  yellow,  compadl  limestone 

*^  rises  up  in  thin  beds  from  beneath  the  red  and  blue  marls  of  the 
Kusulje  Mountains.  This  limestone  also  rises  in  peaks  on  the  N. 
side  of  the  Mermid  river-plain,  and  forms  the  summit  of  the  easy  pass 
overlooking  the  N.E.  arm  of  Lake  Van,  but  it  soon  becomes  con- 
cealed by  basalt'' 

To  the  E.  of  Van,  on  the  road  to  Kotur  and  Khoi, 
Loftus  (Ix.)  found  limestone  of  the  same  age  occupying 
the  lower  ground,  and  resting  on  granitic  rocks  on  the 
N.  side  of  the  Mehmedik  river,  and  on  old  "blue 
and  grey  limestones  and  slates"  on  the  S.  side  of 
the  river.  The  ridge  between  Serai  and  Astuji,  and 
the  undulating  heights  enclosing  the  freshwater  lake 
of  Sharab  Khani,  likewise  consist  of  this  later, 
presumably  Miocene  limestone.  It  will  probably  be 
found  eventually  to  belong  to  the  same  zone  of  Mio- 
cene limestone  which  occurs  further  E.  at  Khoi  and 
Urmi  (pp.  436-^^2). 

The  block  ot  Miocene  limestone  W.  of  Lake  Van 
is  interrupted  in  its  westerly  extension  by  Lake  Nazik 
and  the  Bilejan  volcano,  but  it  soon  rises  up  again 
W.  of  Liz,  and  I  found  it  to  be  cut  through  by  the 
transverse  gorge  of  the  Murad  N,  of  Mush  (pp.  193- 
iq6).  My  observations  indicate  a  great  thickness 
of  buff  conglomeratic  limestone  evenly  bedded  and 

so 


^5^  G^logy  of  Artnema 

vealed  under  the  microscope  (p.  75)  remaixis  of  Litho- 
thamnian^  together  with  Bryozoa  and  Foraminifeia 
(Miliolina,  O^culina),  similar  to  those  in  the  Al  Khes 
limestone.  The  specimen  also  contains  a  well-rounded 
pebble,  i^  inch  in  diameter,  of  a  dark  green,  much 
altered  okvine-diabase,  of  a  kind  which  I  nowhere 
found  in  situ  during  my  journey. 

Further  £.,  near  Iwanjean  (p.  81)  the  Miocene 
limestone  is  a  pale  buff  with  a  pinkish  tinge.  Under 
the  microscope  I  found  it  to  contain  the  following  :— 

Lithothamnion  ramossissimumj  lUiw. 

Amphistegina  cf.  Hauerina,  d'Orb. 

Ceriopara  anomala^  d'Oib.,  and  other  Bryozoa. 

A  few  miles  still  further  £.,  at  Al  Khes  in  the  Kersik 
valley  (p.  82),  the  soft,  chalky  limestone  contains 
the  same  fossils,  together  with  a  large  Otbitoides 
(Lepidocyclina)y  which  forms  the  main  constituent 
of  the  rock. 

Near  Kali  (a  few  miles  £.  of  Al  Khes)  the  limestone 
is  pinkish  and  crystalline,  with  abundant  Foramini- 
fera,  chiefly  Amphistegina  (p.  83). 

It  was  evidently  the  same  series  of  Miocene  lime- 
stone which  Koch*  traversed  further  S.  between 
Gargalik  and  Karajviren,  where  he  found  *'  a  Tertiary 
limestone,  with  abundant  Foraminifera  and  fossils 
like  Serpulce,**  overlain  by  the  basalt  of  the  Torlu 
steppe. 

This  zone  of  Miocene  limestone  extends  also  to 
the  N.  of  the  Elmali-dereh  distridl,  rising  up  to  form 
the  white,  E -W.  range  of  the  Sharian  or  Mergemir 
Dagh  (p.  9).  It  was  doubtless  from  this  range  that 
the  Araxes  derived  its  pebbles  of  compaA^  yellow 
limestone,  which  Abidi  |   foimd  in  the  gravek  of 

*  Heiu  im/MtiscAen  GMrgft  u-  437* 

t  Geoi.  Fomh.  Pi  dm  ktmk.  IMmdim.  ti.  59. 


Lower  Miocene  453 

this  river  in  Pasin;  in  these  pebbles  he  found  the 
Lower  Miocene  corals  Heliastrcea  Ellisiana^  Defr., 
and  Prionastroea. 

I  met  with  the  Miocene  belt  of  the  Sharian  Dagh 
on  the  W.  side  of  the  Araxes  basin,  at  Madrak  (p.  52), 
on  the  same  line  of  strike  ;  the  succession  is  as  follows : 

I.     Massive  white  limestone  with  ThamnarcMi polymarpha^  Abich. 

9.  Brownish  limestone,  mainly  composed  of  Lithothamnion^ 
Foraminifera,  and  Bryozoa.  Only  the  upper  part  is  visible.  It 
contains  the  following  forms  : — 

lAthothamnion  ramcssissimum^  Reuss. 

MilioUna ;  OrbitoliUs ;  Opercuiina ;  Amphistegina  cf.  Hduerina^ 
d'Orb.;  Orhitoides  (Lepidocyclina)  Mantelli, 

Cidaris  (occasional  plates  and  spines). 

Ceriopara  anomala^  Abich,  C  cf.  palmata^  Abich ;  Retepora, 

The  Miocene  of  Madrak  becomes  concealed  to  the 
W.  beneath  the  lava- plateau  of  Pasha  Punar  and 
the  Khalkhal  Dagh  but  it  re-appears  again  on  the 
same  line  of  strike  in  the  valley  of  the  Tuzla  or 
Mushlu  Su,  down  to  its  confluence  with  the  Frat. 
Thus  Tchihatchejff,*  whilst  proceeding  downstream 
from  Pardi  (Bardak)  to  Komentzor  (Kamsor,  Kemler), 
noticed  the  occurrence  of  numerous  Miocene  corals. 

*' Between  Kemler  and  Sarykaya  [halfway  between  Chikhnis  and 
Kotur]  the  limestones  are  not  only  impregnated  with  salt  but  contain 
Miocene  corab.  •  .  The  whole  of  the  plain  between  Chikhnis  and 
Sarikaya  is  strewn  with  blocks  of  limestone  containing  coral-fragments, 
among  which  HeUastraa  EUisiana^  Defr.,  occurs,  evidently  proceeding 
from  the  mountains  bordering  the  plain  on  the  N.£.  and  S.W.  .  . 
The  N.  bank  of  the  river  at  Kemler,  near  its  confluence,  is  bordered 
by  limestones  from  which  a  salt  spring  issues.  Close  to  this  spring 
the  yellow  limestone  contained  innumerable  03rsters  of  a  Miocene 
fades,  together  with  corals,  e.g.^  Heliastraa  Ellisiana.  .  .  .  Higher 
op  the  stream,  about  a  leagues  N.W.  of  Bardak,  near  a  salt  spring, 
limestone  hills  rise  up  above  the  surrounding  dolerite,  and  contain 
fragments  of  the  same  coral.    The  strata  dip  N." 

This  zone  was  traced  further  W.  by  TchihatchefF 
{I.e.)  "in  the  Mustav-deressi  gorge  of  the  Frat,  5 

*  Am  Mimurgf  Gtoiagkf  L  266,  iti.   zoi-zo8. 


454  Geology  of  Armonia 

leagues  E.  ci  Kelarik  (Kalarach)/'  Here  he  found 
Heltastrcea  Ellisiana  in  a  white  limestone  alternating 
with  red  or  black  marls^  dipping  N.E.  when  not 
vertical. 

In  the  upper  part  of  the  Frat  valley  Abich  has 
described  the  highly  fossiliferous,  Miocene  limestone 
of  Kardarich,  a  tew  miles  above  Ashkala.    Here  the 
Erzerum  plain  is  closed  in  on  the  W.  by  a  wall 
of  limestone,  through  which  the  river  has  cut  a  short 
ravine.    Little  Kardarich  lies  at  the  S.  base  of  a 
conspicuous  hill  of  this  limestone  from  which  Abich 
coUe^ed   numerous  fossils  of  Lower  Miocene  age. 
DeyroUes*  has  figured  this  locality  where  the  Fiat 
*'  is  narrowed  between  two  enormous  rocks  of  shcdl- 
limestone/'    in    which   he   noticed    «<  giant   oysteis, 
Pedtens  and  numerous  other  shells,  madrepores,  &c." 
The  cliff  on  the  south  side  of  the  Frat  gorge  is 
called  Gaban  or  llanjik  according  to  Abich's  account. 
Abich's  sedtionsf  seem  at  first  sight  confused  and 
contradidlory,  but  his  mingling  of  older  beds  with 
later  can,  I  think,  be  brought  into  line  with  all  the 
previous  stratigraphy  of   the  Armenian  Miocene  if 
we  assume  that  some  of  the  beds  which  he  observed 
were  repeated  by  step-faults.    In  facft,  a  few  miles 
further  E.,  at  Karabairkh  (p.  43),  I  noticed  a  series 
of  similar  step-faults  in  the  Pleistocene  lake-deposits 
of  the  plain  of  Erzerum.    Abich*$  second  and  more 
detailed   section   (which   reads  a,  b,   c,  cl,i^e,    /,   in 
downward   succession)    may   be   arranged,    on    this 
assumption,  in  the  following  maimer : — 

^  Itmr  At  Jt^rndtf  1875. 

t  Dot  StrimstfU  ^v.,  M^m.  Acad.  ScL  St  PAenb.,  tfr.  6,  ScL 
math.  &  ^ju  Tii.  Tab.  X,  S;  and  his  Gt^L  HmcA.  Ai  dm  kmJL 
Umiim^  iL  loj. 


Lower  Miocene 


455 


FIRST  RBPBTmON 

d.  Fine-grained,  cal- 


a.  Dull  green,  sandy 
limestone,  with  small, 
rounded  fragments  of 
serpentinous  rocks;   idide;   with  small  frag 
shows  occasional  layers  men^    of   serpentine, 


careous  conglomerates|tnoved  by  denudation, 
coloured  with  iron  ox- 


indistinct  bivalves  and 
numerous  Bryozoa. 

d,  Abich  repeats  this 


SECOND  REFSnnOH 

Absent;    probably  re- 


/.  Light  grey,  compact 
limestones.  Layers  oc- 
cur with  large  Echinoids, 
oyster-beds  and  shell- 
ragments.  Gastropods 
re  absent. 


€.  Friable,  light  greyl  These  limestones  pass 
limestones,  which  arelgradually  downwards  in- 

|to  more  compact,  azoic 
[beds  of  limestone. 


of  compact,  grey,  con 
cretionary  lunestone  full 
of  Terebratula  minor. 

b.  Thick  beds  of  con 

glomerate  (with  manykiivision  in  his  section, 
fragments  of  Astraean 
corals)  alternating  with 
oyster-beds  of  consider- 
able thickness,  with 
large  oysters. 

c.  Sandy  friable  lime- 
stone, with  extremely 
numerous    £chinoids,|actual  PecUn-htds. 
abundant   PecUn  ana| 
other  bivalves. 

In  accordance  with  this  re-arrangement,  which 
gives  a  natural  sequence,  I  have  classified  Abich's 
fossils  *  in  a  corresponding  manner : — 

1.  Arenaceous  limestone  with  small  pebbles  of 
serpentine.  Intercalary  beds  occur  of  grey,  concre- 
tionary limestone,  full  of  T^ebratula  minor^  Suess. 

2.  (a)  Oyster-beds  with 

Ostrea  (Alediryonia)  Virleti^  Desh.,  0.  crassisstmay 

Lam.,  0.  crassicostata^  Sow.  (  =  0.  excavataj  Desh.) 

bored  by  Lithodomus  lithophagus^  Linn. 

(b)  Coral-limestone,   with    great    numbers    of 

Astraean  corals,  especially  Heliastrcea  Ellisiana^  Defr., 

and  Astraea  crenulata.    Great  numbers  of  Bryozoan 

fragments  occur  between  the  corals. 

3.  (a)  Pectew-limestone  with 

Pecten  dubius^  Brocchi,   P.  planocostatus^  Abich,  P. 

*  Das  Sfeinsaig  &*c.,  Mem.  Acad.  Sci.  St.  P^tersb.,  s^.  6,  Sd. 
math.  &  phys.  vii.  31 ;  and  his  GeoL  Forsch.  in  den  kank.  L&ndim^  ii. 
aiz-223. 


45^  Geology  of  Armenia 

burdigalensts^  Lam.  (in  different  sizes),  P. 
dtfficilis^  Fuchs. 
{b)  Echinoderm-limestone  with 
ClyPeaster  Guntheri^  Gregory  (C.  altus^  Lam.,  w. 
turritus^  Phillippi,  of  Abich),  C.  diversicostatits^ 
Abich;  Echinolampas  complanatus^  Abich,  E. 
pyvamidalisj  Abich;  Spatangus  euphratensis^ 
Abich  (in  the  lowest  beds). 

Abich  *  met  with  the  same  P«:ten-limestone  further 
W.,  between  Mamakhatun  and  Erzingau,  on  the  pass 
(6462  feet)  between  Elmali  (Almaly)  and  Dibija,  near 
the  Nummulitic  Limestone  recorded  by  Tchihatcheff 
(p.  414).  Further  W.,  at  Sarydash  (on  the  S.  side 
of  the  Sipikor  pass  from  Kelkid  Chiftlik  to  Erzingan), 
Abich  t  found  massive  conglomerates  of  Upper  Miocene 
age  overlying  a  series  of 

''reddish,  marmoraceous  limestones  in  well  milked  beds,  contaimng 
abundant  individuals  of  Fecten' pradubius^  Abich,  and  P.  eupkraUnm, 
Abich.  The  loose  texture  of  the  pale  reddish,  marly  limestones  of 
the  /'(tf/lrff-beds  at  Sarydash  and  the  intercalation  of  loose  beds  of  fine 
fragments  of  shells  and  Echinoids  render  it  probable  that  the  whole  is 
a  beach-deposit." 

Below  this  series  there  occur  about  80  feet  of 
yellowish,  marly  limestones,  and  the  entire  system 
has  been  intruded  by  ultrabasic  rocks. 

The  N.  shore  of  the  Miocene  sea  of  Armenia  seems 
to  have  been  marked  by  the  line  of  Lower  Cretaceous 
heights,  which  form  at  present  the  Frat-Kelkid  and 
Frat-Chorokh  watershed.  It  is  possible  that  a  gulf 
extended  into  the  Tortum-Olti  district,  for  Abich  J 
noticed  here  limestones  (unfortunately  unfossiliferous), 
similar  in  appearance  to  the  Miocene  limestone  of 
Kardarich,  and  overlain  by  gypsiferous  and  saliferous 
marls  of  presumably  Upper  Miocene  age. 

*  GeoL  Farsch,  in  den  kauk.  Landtm^  ii.  IS5. 
t  Ibid.  ii.  isxy  215,  ai6. 
X  Ibid,  iL  176-179. 


Lower  Miocene  457 

To  the  W.  of  the  great  bend  of  the  Frat  Miocene 
strata  are  probably  indicated  by  the  "  soft,  cretaceous 
limestone  with  flints,  large  Pecftens  and  large  Echi- 
nids,"  which  Ainsworth^  found  extending  trom 
Kirchu  N.W.  to  Arabkir  and  beyond  Berastik;  it 
is  about  600  feet  thick  at  Arabkir  and  overlies  mica- 
schist  (p,  273).  The  limestone  series  contains  inter- 
bedded  sheets  of  basalt,  and  is  nearly  horizontal. 

To  the  S.E.,  Smyth!  found  that  "superimposed 
on  the  .older,  grey  limestones  [of  Keban  Maden]  there 
occur  beds  of  a  white,  softer,  calcareous  rock,  with 
a  strong  resemblance  to  the  Calcaire  grossier  or  Grob- 
kalk  of  the  Vienna  basin,  and  containing  oyster 
shells."  This  is  quite  different  in  character  to  the 
hard  and  compadl  Nummulitic  Limestone  £.  of 
Keban,  and  also  on  the  heights  above  Arghana. 

Hommaire  de  Hell  J  also  refers  to  this  white,  chalky 
limestone  at  Sovuk  (Zaouk)  on  the  Frat,  where  it 
discordantly  overlies  an  older  limestone,  which — ac- 
cording to  the  index  of  his  specimens — is  probably 
Eocene.  Taylor  §  too  speaks  of  the  formation  between 
Arabkir  and  Egin  as  consisting  of  ''chalk,"  and 
also  of  having  traversed  "  chalk  downs"  from  the 
Frat  to  Chemishgezek. 

A  gulf  of  the  Mesopotamian  Miocene  sea  probably 
washed  the  S.  foot  of  the  Taurus,  for  Hommaire  de 
Hell  II  foimd  a  Clypeaster  ("  belonging  to  the  C.  alius 
group,"  according  to  Viquesnel  5f )  in  the  Tigris  basin, 
evidently  from  some  point  between  Zokh  and  the 

^  Xesearckss  in  Assyria^  ^c.^  aSi-aSj. 

t  Quari.  Jnum.  GtoU  Soc.  L  338. 

X  Voyagi  em  Tkrfuii  ^.^  GhlegU^  it.  77. 

{  Jcum.  Hay.  Gugr.  Sac.  3X4t  3x5* 

0  Op.  ai.  87. 

IT  BttU.  Soc.  GM.  Firana^  wti.  s,  vu.  308. 


A 


458  Geology  of  Armenia 

Batman  river,  probably  close  to  the  latter  near  Is^» 
Rezou,  where  "  a  white  marble  "  occurs. 

It  is  possible  that  a  Lower  Miocene  age  may  be 
ascribed  to  the  "low  and  regular  hills  of  soft  and 
fissile  chalk  **  of  Samsat  and  the  neighbouring  country 
along  the  Euphrates  valley.    Ainsworth  *  states  that 

"from  Samsat  to  Urfa  the  undulatory  character  is  broken  up  by 
volcanic  ridges  and  cones  and  then  come  deep  valleys  divided  l^ 
*  long,  parallel  hill-ranges  of  chalk  capped  by  basalts.'* 

H.      UPPER  ICIOOINS  (SARlCiLTUN). 

Although  for  the  most  part  unfossiliferous  on  the 
plateau-region,  the  Upper  Miocene  which  occurs  along 
the  outer  margin  of  the  Northern,  Eastern,  and 
Southern  Border-Ranges  of  Armenia  encloses  a  rich 
fauna  of  lagoon  and  brackish  water  molluscs. 

In  the  region  of  the  Central  Caucasus  the  Sar- 
matian  deposits  are  the  earliest  representatives  of 
the  Miocene  and  occur  in  notable  transgression 
on    Palaeogene    deposits.    According    to    Fournierf 

''The  Sarmatian  extends  along  the  S.  slope  of  the  Caucasus  unin- 
terruptedly from  the  Black  Sea  to  the  OMpian,  which  were  united 
both  to  the  N.  and  S.  of  the  Caucasus.  This  chain  then  formed 
a  vast  island,  and  the  folding  has  not  yet  finished^  for  the  Sarmatian 
all  along  the  principal  chain  is  overfolded  on  the  S.  slope. 

The  upper  part  of  the  Sarmatian  of  the  S.  Caucasus  contains  the 
Mceotic  fauna  of  Andrussow.  .  .  The  Pontian  stage  does  not,  in 
the  Central  Caucasus,  contain  Congeria  beds;  it  is  represented  by 
sandy  clays  and  unfossiliferous  sands,  with  thin  layers  of  lignite 
and  passes  insensibly  into  the  conglomerates  and  alluvia  of  the 
Pliocene.  Hence  there  was  a  gradual  retreat  of  the  sea  firom  E. 
to  W.  during  the  Sarmatian  period. 

In  the  basin  of  the  Lower  Rion  and  Kvirila  Sarmatian,  calcareous 
sandstones  and  arenaceous  marls  are  transgressive  over  aU  lower 
deposits.     In  the  N.W.  part  of  the  ZiruU  dome  and  in  the  W.  part  of 

*  Researches  in  Assyria^  <S*r.,  290. 

t  Description  gkol,  du  Caucase  central^  187 ;  and  see  also  Abich, 
J/i/M.  ausd.  Kauk.^  Verb.  k.-k.  geol.  Reichsanst  1877,  34. 


460  Geology  of  Armenia 

Dary  Dagh  (p.  406).     In  these  yellow,  sandy  marls  of 
Juifa  Tsulukidse  *  found  the  following : — 

PeAen  sartnaticuSf  Gf.»    P.  lasuicostatus^  Sow.,  P. 
Hopkinsi^  d'Arcb« 

At  Yaiji,  lower  down  the  Araxes,  similar  saliferous 
marls  form  a  plain  between  the  Dewil  and  Itushem 
Mts,  or  Nisham-kuh  (Kharzan  Dagh).  Here — accord- 
ing to  Grewingk  t — rock-salt  is  adively  worked  (as  at 
Nakhichevan),  and  at  the  N.  end  of  the  quarry  '*a 
high  cliff  of  rock-salt  rises  up,  covered  by  clay,  which 
is  overlain  by  granular  gypsum."  This  salt-hill 
(Kupchivan)  is  clearly  figured  by  Abich  %  in  his  view 
from  the  Dary  Dagh,  looning  S. 

To  the  S.  of  Ararat,  on  the  S.  side  of  the  Araxes 
basin,  Abich  §  noticed  red  sandstones  and  marls  on 
the  Gedargan  Dagh  near  Bavazid,  overlying  nearly 
vertical  Helvetian  limestones  (p.  444). 

The  saliferous  marls  are  extensively  developed  in 
Azerbaijan,  especially  in  the  valley  of  the  Aji  river  on 
which  Tabriz  is  situated.  Here  Abich  ||  e^ves  the 
following  succession  of  these  Upper  Miocene  beds : — 

I.  Sandstones  and  marb. 

«•  Red«  femiffinons  conglomerates  and  sandstones. 

3.  Ysri^ated  marls. 

4*  Ydlow  and  grey,  gypsiferous  maris. 

A  very  similar  succession  was  also  observed  by 
Abich  ^  in  the  Shorr-dara,  between  Sufian  and 
Maraud,  and  the  marls  are  probably  also  saliferous, 

*  Gi^.  Caial.  Tijtis  Mus€um^  los. 

t  Gtognoit.  6*  9roiraph.  VerhaliH.  d.  nSrdl.  PenUms^  34^  44. 

X  Gtol.  Fonch.  in  den  kauk.  Landim,  ii.  80. 

I  Das  SietfUMb  ^c,  Mem.  Acad.  Sd.  St  P^tersb.,  a£r.  6,  Sd 
matli.  &  phjs.  viL  7. 

B  Md.  Taf.  X,  Section  zo. 

t  IbuLX,  II. 


upper  Miocene  461 

Sot  near  Marand  Muller-Simonis  ^  noticed  that  the 
Zilbis  (Zunus)  river  flows  '*  between  hills  of  salt" 

To  the  S.  Loftusf  has  given  a  general  description 
of  the  salt*hill  of  Khoi,  rising  15a  feet  above  the 
plain  of  saliferouSi  Upper  Miocene  marls.  Abich:^ 
nas,  however,  described  the  succession  more  precisely, 
as  follows : — 

I.  ^  Con^omerates  and  graveb  of  Khoi  plain ;  the  predominant 
terpendnous  dements  give  these  conglomerates  a  pecufiar  variegated 
appeazanoe.  Adjoining  the  Salt  Hill  (Diulak)  these  fragmental  beds 
consist  of  confusedly  bedded,  reddish  brown  and  greenish  grey  marls, 
mixed  with  gypsum  and  pebbles,  and  containing  large  blocks  of 
Kdyetian  coral-limestone  [No.  6],  fragments  of  dolomitic  limestone  in 
great  numbers  [probably  PalaeoxoicJ,  and  numerous  fragments  of 
fireshwater  limestone  [No.  5]. 

s.  Marlj,  fragmental  sandstones  and  fine  gravels  with  consider- 
able masses  of  gypsum,  both  massive  and  in  loose  crystals. 

3.  Bed  of  rock-salt,  externally  with  a  fragmental  appesrance 
like  coarse  ^lader-ice,  but  internally  compact.  On  the  steeper,  E. 
side  of  Khoi  hill  only  impure,  reddiih  or  grey  rock  salt  appears. 

4*  Red,  saline  clay,  with  gttBi  numbos  of  small  pebbles  of 
(Efferent  colours. 

$.  Freshwater  limestone  with  Pianorh's  Jieussi\  Homes,  and 
Pu^i  niusa^  Sandb.    It  was  probably  quite  a  local  formation. 

6.    Helvetian  oond-limestone  (p.  441 X 
There  seems  to  be  an  unconformity  between  1  &  a,  and  between 

4&S- 

The  variegated  marls  of  Khoi  plain  extend  west- 
wards up  the  Kotur  river,  as  far  as  2  miles  up  its 
ravine,  ipiccording  to  Loftus.§  On  the  S.  side  of  the 
Gadykh-tapa  (bordering  the  plain  of  Dilman  on 
the  N.)  Abich  (Z.c.)  observed  that  the  variegated  marls 
and  conglomerates  contain  intercalary  beds  of  g>'p- 
sum  and  show  a  southerly  dip.    On  the  S.  and  W. 

*  Du  Caiicau am gotfi peniqui^  iti. 

t  Qupri.  J9um.  GtoU  S§c  d.  308,  309. 

X  GfoL  Foruh.  in  dm  ktnA.  Limkm^  vL  S71,  sf  4,  175. 

I  Cf.€U.  310. 


463  Geology  of  Armenia 

borders  of  this  plain,  in  the  valley  of  the  Zuba  (Cheh- 
rik)  river,  Lof tiis  *  found  the  same  gypsiferous  series  of 
sands  and  gravels  (dipping  N.)*  covered  by  extensive 
sheets  of  leucite-basalt.  Further  S.  Loftusf  found 
that  the  Seyr  Dagh  (7260  feet)  consists  of  the  **  gypsi- 
ferous  series/*  which  is  here  composed  of  yellow  or 
brownish  sandstones,  alternating  in  their  upper  part 
with  conglomerate- gravels,  and  in  their  lower  part 
with  marls  and  impure  limestones  (p.  440).  The 
presence  of  fragments  of  Helvetian  limestone  in  these 
sandstones  has  already  been  referred  to  on  p.  441. 

To  the  E.  of  Lake  Urmi  and  on  the  S.  side  of 
Sahend  Monteith  t  has  recorded  that  salt-mines  occur 
12  miles  S.  of  J^ohak,  probably  in  saliferous  beds 
of  Upper  Miocene  age,  wliich  become  increasingly 
developed  on  proceeding  further  into  Persia.  One  d 
the  most  important  salt-mines  of  Persia  is  situated  to 
the  S.  of  Savalan,  at  Maman,  i  day  east  of  Mianeh,  in 
the  valley  of  the  Kizil  Usen  river.  Here — according 
to  Gdbel§ — the  rock-salt  is  accompanied  by  thici 
beds  of  gypsum  and  by  yellowish*  calcareous  marls 
and  variegated  clays,  forming  a  sjmcline  overlying 
Lower  Miocene  limestone  ("  Supra-Nummulitic  '  of 
Abich).  The  rock-salt  (over  25  feet  thick)  is  associat- 
ed with  red  nodules  of  carnallite,  enclosing  diatoms 
(Gallionella)  and  spicules  of  Spongilla  fiuviatilis.  It 
seems  probable  that  these  organisms  concentrated  the 
slight  amount  of  iron  in  the  water  and  hence  caused 
the  red  colour. 

Returning    to    the    Araxes    valley,   we   find   that 
saliferous  marls  are  well  developed  at  Kulpi,  situated 

•  Quart  Jaurn.  GtoL  Sac.  zi.  301. 

t  Op,  cit  998,  299. 

X  Joum.  Roy.  Geogr,  Sae.  iii.  5. 

{  Bull.  Acad.  Set.  St.  PiUrsb.  ix.  1-26. 


Upper  MioetM  463 

^  the  exit  of  the  Araxes  from  its  deep  canyon  into 
the  wide  depression  between  the  huge  volcanoes 
Ararat  and  Alagoz.    According  to  Abich  *  the  salt-hill 

*' rises  400  feet  above  the  level  of  the  Anzes,  with  an  area  of  1.5 
▼exsts  [5/6  mile]  ^  it  is  steq>  on  3  sides  and  has  a  gently  synclinal 
surface.  The  pure  ra^^-salt  is  in  beds  s^so  metres  in  thickness; 
slight  clayey*  intercalary  layers  occur  only  in  the  upper  part ;  the  total 
thickness  <»  the  salt  is  s8o  feet.  It  rests  directly  on  reddish,  ar- 
gillaceousi  marly  beds,  with  gypsum  and  Glaub^s  salts,  dipping 
18*  N.£.«  with  a  strike  of  N.  39**  W.,  overlying  Eocene  beds  un* 
conformably.  It  is  dear  that  the  whole  salt-hill  has  been  isolated  by 
fluviatile  erosion  from  the  low  slopes  at  the  foot  of  Takyaltu  Dagh." 

Abich  t  traced  these  red  saliferous  and  gypsi- 
ferous  marls  right  up  the  Araxes  canyon  past  Kagiz- 
man  (where  rock-salt  is  quarried)  to  Alija-krak  and 
Chiftlik,  where  the  conglomerates  contain  pebbles 
of  Helvetian  coraUlimestone.  The  series  appears  to 
be  unfossiliferous  and  is  overlain  by  Pleistocene  lake- 
deposits.    The  marls  are  often  highly  variegated. 

The  saliferous  series  is  widely  distributed  in  the 
basin  of  the  Tortum  and  Olti  rivers,  and  also  in 
the  valley  of  the  Lower  Chorokh  below  Ispir,  but 
no  fossils  have  been  recorded.  Rock-salt  is  quarried 
near  Olti,  and  the  general  succession  in  the  district 
seems  to  be  the  same  as  in  the  Araxes  canyon,  viz.: — X 

X.    Grey  days  enclosing  fibrous  gypsum  and  alabaster. 
a.    Grey  day  with  rock-salt  in  layers  and  nests. 

3.  Red  sandstones  and  variegated  marls. 

4.  Yellow  and  light  grey  limestones  of  presumably  Lower 
Miocene  age. 

Hamilton  §  has  remarked  on  the  prevalence  of  tl  e 


*  Gv/.  Fonck.  in  dm  ktmk.  Ldndim^  ti.  \i\ ;  and  see  also  his 
Das  StitnstUM  df^.,  M^m.  Acad.  Sd.  St  P^tersb.,  s6r.  6,  Sd.  math. 
&  phys.  viL  s« 

t  Geol.  Fcruk.  6v.  vL  1369  139,  I4S*I4S9  X47« 

X  Ibid.  iL  i76*i8o* 

§  Xiuarchts  in  Ada  Minar^  i.  ao7*sio. 


■V 


464  Geology  of  Armenia 

variegated  marls  in  the  Nariman  and  Bardus  valleys. 
Abich  ^  found  these  variegated  marls  to  contain  much 
gypsum  and  salt  between  OIti  and  the  Pennek  con- 
iluence,  and  Koch  f  observed  them  also  in  the  Pennek 
valley. 

To  the  W.  of  Olti,  near  the  head  of  the  Tortum 
valley,  Abich  J  noticed  salt  springs  at  Bar  (6goo  feet), 
suggesting  the  proximity  of  the  saliferous  series,  and 
Koch  §  records  a  salt-working  at  Tuzla,  lower  down 
the  valley.  Red  sandstones  are  well  developed  in  the 
surroundings  of  the  pidluresque  Tortum  lake. 

The  same  variegated,  saliferous  marls  were  obser\^ed 
by  Koch  ||  in  the  Lowei>  Chorokh  valley  below  Ispir, 
from  2^gos  through  ^armenik  and  Kudrashen  to 
Pertekrek.  The  white  alabaster  found  by  Batse- 
vitch  ^  in  the  Belisor-deressi  near  Khodulia  (lat.  41° 
N.)  may  perhaps  indicate  a  still  more  northerly 
extension  of  the  gypsiferous  series. 

Turning  southwards,  on  the  other  hand,  my  own 
observations  (pp.  41,  42)  show  that  the  broad  Frat 
valley  between  the  Lower  Miocene  limestone- heights 
of  Pirnakapan  and  Ashkala  is  occupied  by  clays  with 
thick  seams  of  white  gypsum.  The  Upper  Miocene 
age  of  these  deposits  in  the  Frat  valley  is  indicated 
by  Abich's  statement  •*  that  the  Helvetian  limestone 
of  Kardarich  (p.  454)  on  the  N.  side,  and  that 
of  Gaban  (Ilanjik)  on  the  S.  side  of  the  Frat  are 

^  Geoi.  Forsch,  in  dm  kattJL  Zandtrm^  ii.  176,  177,  180.. 

t  Xiisi  im  fontischin  GMrgt^  ii.  t^yz^^ 

X  Op.  at.  174. 

{  Op.  at.  S64. 

n  Ibid.  50,  52,  s9-6i. 

S  Gtol.  Catal.  Tiflit  Musnm^  laa. 

••  Dai  SUtnsaU  6v.,  Mem.  Acad.  ScL  St  Pftmb,,  a6r.  «,  Sd 
injuh.&  phys.  viL  20,  &Taf.  X,  8. 


upper  Miocene  465 

*'  in  places  overlain  by  light  grey,  tuik^ua  mohssey  belonging  to 
the  upper  memben  of  the  variegated  marlt,  with  indistinct  fiagments 
ctJ'aim  and  other  marine  sheUL'' 

My  observations  also  corroborate  Abich's  state- 
ment* that  the  gypsiferous  clays  are  underlain  W. 
of  Ashkala  by  red  marls  and  conglomerates  with  salt- 
springs  at  Chishveran  (2  miles  S.W.  of  Ashkala). 

Lower  down  the  Frat  valley,  below  Mamakhatun, 
Tchihatcheff  f  ascribes  a  Miocene  age  to  the  ^*  snow- 
white  hills  of  gypsum  in  linear  series  about  3  leagues 
N.W.  of  Kemier  and  also  round  Chikhnis/*  They 
probably  stand  in  close  relation  to  "  the  very  massive, 
nearly  horizontal,  dark  brown  and  red  conglomerates, 
which" — ^according  to  AbichJ — ^** stretch  across  the 
whole  valley  of  Mamakhatun  in  connedtion  with 
the  similar  formation  in  the  Frat  plain/'  Abich's 
observations  §  show  that  this  series  is  extensively 
developed  round  the  borders  of  the  Ercingan  plain, 
in  the  form  of  marls,  brown  sandstones  and  serpen - 
tinous  conglomerates,  often  gypsiferous,  and  with 
fragments  of  Helvetian  Pecten-limestone — the  whole 
with  a  S.W.-N.E.  strike. 

The  saliferous  series  is  extensively  developed  to 
the  W.  and  S.W.  of  Erzingan.  Thus  in  the  Gerjanis 
distridl  Wiinsch  ||  has  recorded  salt-pans  at  ''  Tuzla, 
behind  Chaush,"  a  little  below  Kurdkoi  (Elpkurdkoi). 
Major  F.R.  Maunsell  kindly  informs  that  salt-pans  are 
also  present  at  Aderik  (3  miles  below  Ak-kdi  and  8  or 
9  miles  below  Gerjanis). 

The  Frat  appears  to  cut  through  red,  saliferous 
marls  from  Kemakh  to  Egin  and  its  waters,  previous- 

^  Geol.  Farsch.  in  den  tank.  Zandtm,  iL  105,  106. 

t  Asie  Mituun^  Ghl^gk^  vL  108. 

X  op.  €it.  119. 

§  Ibid,  issy  115. 

H  MUih.  i.-i.  gmir.  Gis.  xxfiL  aoS. 


> 


466  Geology  of  Armenia 

ly  yellow,  now  become  red.*  At  Komur  (Keumur), 
a  few  miles  N.  of  Kemakh,  Wiinschf  mentions  the 
occurrence  of  much  rock-salt.  Major  Maunsell  also 
tells  me  that  '*near  Kemakh  are  salt-pits  yielding 
a  fine,  white  salt/'  and  that  '*  the  road  from  Kemakh 
to  Egin  leads  over  coni<::al  hills  of  gypsum  and  rock- 
salt.'  The  observations  of  Taylor^  and  Hommaire 
de  Hell§  show  that  gypsiferous  shales,  associated 
with  sandstones  and  conglomerates,  are  extensively 
developed  in  the  vall^  of  the  Kuni-diai^  and  as  far  as 
the  volcanic  mass  of  J6imara. 

Upper  Miocene  deposits  are  present  in  the  region 
S.  of  Erzerum.  Salt-pans  occur  at  ICaragdl  Tuzia 
— ^according  to  Major  Maunsell  and  Koch  || — ^about 
4  or  5  miles  £.  of  Chat,  and  just  N.  of  the  Pasha 
Punar  pass.  This  place  is  on  the  same  line  of  strike 
as  the  red  marls  and  sandstones  which  I  observed 
lying  to  the  £.  of  the  lava-plateau  of  Madrak  and 
Pasha  Punar  (p.  54J.  These  beds  overlie  the  Lower 
Miocene  limestones  just  as  at  Ashkala,  and  Uttty  are 
probably  continuous  with  the  Upper  Miocene  deposits 
lower  down  the  Araxes  basin  (p.  463).  It  is  among 
these  dark  crimson  strata  that  another  TuzIa  is 
situated  near  the  bridge  over  the  Araxes,  "7  hours 
N,E.  of  Khinis/'  evidently  between  Kulli  ana  Tash- 
kessan,  with  a  '^deposit  of  rock-salt,"  according  to 
Brant,  5r  or  merely  salt-pans,  according  to  DeyxoUes.^* 

On  the  S.  side  of  the  plain  of  Khinis,  the  Bashkent 

^  Yorke,  Geogr.  /cum.  viiL  534. 

t  Mitth.  i.-i.  geogr,  Ga.  zzrii.  aoi* 

X  Jomm.  Roy,  Gtcgr,  Soc.  xzxviii.  305,  306. 

S   Vcyag€€n  I^rquitetc.f  GtobngU^  m  74. 

II  Xeisi  impofiHscken  GMrge^  iL  333. 

T  Jaum.  Roy.  Gtogr.  Sac.  z.  34a. 

^"^  TcuriuiiomUt  1875,  40<* 


upper  Miocene  467 


ridge  (joining  Khamur  to  the  Bingol  Dagh) 
according  to  Ainsworth* — composed  of  "conglo- 
merates of  the  saliferous  red  sandstones,  various - 
coloured  sandstones^  sands,  marls  and  gypsum/'  over- 
ling Miocene  limestone.  It  is  only  at  the  £.  end 
of  the  Khamur-Zemak  range  that  salt  is  actually 
worked.  Here — according  to  Kochf — "at  Krimizi 
Tuzla,  on  the  £.  bank  of  the  Bingol  Su,  about  20-30 
salt-springs  rise  up  out  of  a  low  spur  of  the  Lala 
Dagh,  consisting  of  red  and  black  trachyte/'  which 
probably  overlies  saliferous  marls. 

Saliferous  marls  also  occur  on  the  S.  side  of  the 
Taurus.    Both  Taylor  ±  and  Belck  §  mention  the  salt- 

ns  at  Duzla,  3  miles  S.  of  the  Bakirein  tunnel  of  the 
ibeneh  (Dibeneh)  river.  Taylor  ||  also  states  that 
Milejrfan  further  E.  is  "  famous  for  its  salt-works/'  and 
at  Hazo  THuzu),  still  at  the  foot  of  the  Taurus,  "  near 
the  old  bridge  tliere  are  several  excavations  made 
for  salt,  which  is  procured  in  slabs  about  i  inch  thick, 
singularly  pure.1[ 

Between  Zokn  and  the  vallev  of  the  Bitlis  river 
near  Ziaret  Hommaire  de  Hell**  found  red  and 
grey,  shaly  sandstones  overlying  coarse  conglomerates, 
which  continue  as  far  as  Dukhan.  This  series — 
according  to  Viquesnelft — belongs  to  the  Upper 
Miocene  gypsiferous  series.    Near  Sert  Loftus  Xt  found 

*  IVaveb  &*  Reseanhet  im  Asia  Minor^  ^c.  iL  385. 

t  Reise  impmUisdien  GMrge^  iL  35SU363. 

X  /(mm.  Roy.  Gengr.  Soc.  jxrf.  4a 

)  MUih.  gtogr.  Ga.  Bamhifg.  xvi.  46. 

n  Op.  at.  45. 

T  IHd.  46. 

**  V(fyag!t  €H  TStrquiiitc.^  GioUgU^  iv.  89-93. 

tt  BM.  Soc.  ilot.  Fnma^  w6t.  2,  vii.  508. 

Xt  Qftari.  Jourm.  Geot.  Soc.  xi.  247. 


i 


S^K^^Mi^M^ter  oocumng  in 

^-''-^^^giSfceeds  to  state  that 

itocted,  lolt,  anioKOi- 

OB  dw  ikizti  of  the 

pans  occur  in 

ifra,  NX.  of  Sert, 

^__n  this  neighbour- 

'^Sura.ted  upland  be- 

rs^^es^stiie  a  rapid  descent 

MfSk^i  banks  of  a  salt 


'■•rlj^giliferous,  red  sand- 
StlkgfiHi  appear  to  overlie 
lllirH«>^ly  o^  Upper 
QO'^*^*^^^^^^  with 
'gtgkgS..  which  underlie 
il|ii|S-  doubtless   Lower 


>^P<i^^"^^^:^  u-  364. 


CHAPTER    XXIV 

THB  PLIOCBKB  AND  PLEISTOCENE  OP  ARMENIA 

The  mountain-snaking  which  had  mainly  taken  place 
in  the  middle  of  the  Miooene  period  was  continued  in 
Armenia  after  the  Sarmatian  in  the  form  of  a  general 
uplift  of  the  ground,  accompanied  by  fradluring  and 
settling-down  of  large  blocks,  and  the  plateaux  of 
Armenia  and  Azerbaijan  have  ever  since  remained 
above  the  sea.  This  region,  together  with  the  con- 
tiguous plateaux  of  Anatolia  and  Persia,  formed  a 
connedted  land-surface  already  in  the  Middle  Pliocene 
diversified  by  lakes.  The  vegetation  at  that  time 
was  luicuriant  enough  to  support  mastodons,  giraffes 
and  antelopes,  whidi  could  wander  imchecked  from 
Greece  to  the  shores  of  Lake  Urmi. 

Most  of  the  Pliocene  lakes  persisted  into  the  Pleisto- 
cedie,  and  [even  at  the  present  day  several  remnants 
still  occur — either  of  fresh  water,  such  as  Lakes 
Gokcha  (Sevanga),  Chaldir,  Toporavan  and  Nazik, 
or  with  vanning  degrees  of  salinity,  such  as  Lakes 
Khozapin,  Van  and  urmi. 

L     PUOOBNX. 

Abich  *  was  the  first  to  call  attention  to  the  Plio- 

^  Das  Sinnsmlk  6v*,  M^m.  Acad.  Sd.  St  PAenb.,  Ui.  6,  Sd. 
mmth.  &  phyi,  viL  a6,  a8,  83. 


470  Geology  of  Armenia 

cene  fauna  of  Mara^ha,  on  the  £.  coast  of  Lake  Unni, 
by  describing^  Kmains  of  Hipparion  (anterior  cheek- 
teeth), Certms  rtooth  of  upper  jaw),  and  Mastodon 
(bone-fragments).  They  had  been  colledled  by  Khan* 
ykofiE  **  from  a  gorge  near  Maragha  in  a  trachyte- tuff 
mixed  with  eaithy  gypsum/' 

Pohlig  *  investigated  the  localit}*  in  1884^  and  com- 
pared the  faima  of  Maragha  with  that  of  Pikermi ;  it 
is  also  closely  related  to  the  faunas  of  Samos  and 
Troja,  but  has  yielded  some  rather  later  forms.  The 
Maragha  valley  cuts  transversely  through  the  low 
Jurassic  (p.  32a)  and  Cretaceous  ranges  (p.  338),  whidi 
frame  in  the  basin  of  Lake  Urmi  on  the  £.,  and 
upon  which  Sahend  has  built  up  its  huge  pile  of  lavas 
and  tuffs.    The  Middle  Pliocene  deposits  consist  of 

''pale  reddiih  marb,  loose  at  the  suiiacey  but  rtxy  compact  at  a 
depth,  forming  the  hills  near  the  town,  and  showing  stngnlar  forma  of 
erosion.  They  are  the  detritus  of  tiie  volcanic  ashes  and  tufis  of 
Sahend,  and  accumulations  of  pumice-stones  are  frequently  found 
in  them,  usually  associated  with  the  bone-beds.  The  marls  also 
contain  numerous  strata  of  pebbles,  the  d&xiB  .of  the  Sahend  focksy 
and  these  are  sometimes  up  to  more  tfaaa  i  metre  in  diameter. 
Nearer  to  Sahend  the  pebble-beds  prevail^  and  the  bouldcss  axe 
still  hrger." 

The  fossil  bones  occur  in  the  reddish  marls,  in  mote 
than  6  places,  up  to  a  ^  kilometre  t  ftom  Maragha, 
and  at  different  horizons,  not  differing  ^m  eadi  other 
in  faunistic  characters.  The  bones  are  whitish  with  a 
reddish  coating;  those  from  a  depth  have  much 
vivianite,  quite  similar  to  the  Pikermi  bones : — 

Sus^  Palasohyus  maraghanus^  Pohlig  (one  of  the  few^ 
points  of  resemblance  with  the  Sivallk  fauna)  ; 
Cervus ;  Giraffa  attica^  Helladotherium ;  Anti* 
lope  cf.  major  (Antidorcas  ?)^  GcuUlla  hretncomts^ 

*  Zdtsck,  iemtsck.  gtoL  Ga.  xzxviL  iosa*ioa7,  translated  in  (^fort, 
Joum.  GtoL  S^.  UL  179. 

t  Not  *'  30  miles,''  as  erroneouslr  translated  In  tihe  Q.J.GJS^ 


Pliocene  471 

Palceoreas  cf.  Lindermayeri  (Pikermi),  Trago- 
ceros  fvery  common;  allied  to  the  Pikermi 
species),  Bitbalus. 

Hipparion  gracile  (the  commonest  form,  occurring 
with  several  skulls  and  all  the  bones),  Onager  (?), 
and  smaller  species  of  Equidce^  Rhinoceros  per- 
Slop,  Pbhlig  (very  common),  R.  Blanfordi. 

Mastodon  (pentelici  ?),  less  common  than  the  last. 

Canis^  Hyaena  cf.  eximia^  Felis  brevirostris. 

There  is  no  trace  of  bone-caves  or  of  a  Pleistocene 
fauna  {Rhinoceros  tichorhinuSj  Hyaena  spdaea,  &c.), 
as  maintained  by  Brandt  ^  and  others. 

According  to  Rodlerf  these  "loss- like  marls  of 
Urmia  cover  the  watersheid  S.E.  of  Sahend  and  extend 
far  into  the  Karangu  valley.  .  .  The  marl  of  Maragha 
is  certainly  not  a  freshwater  deposit,  for  it  contains 
abundant  crystals  of  g>'psum.'* 

n.    plustocbns. 

In  the  later  Pleistocene  Lake  Urmi  has  shrunk 
several  metres,  for  PohligJ  found  ^^  Neritina  and 
Dreissensta  on  the  islands,  corresponding  to  the  later 
travertines  on  the  Shahi  peninsula,  with  Neritina  and 
Dreissensta.**  Pohlig§  also  noticed  travertine  with 
Neritina^  Hydrobia  and  Dreissensta  along  the  coasts  of 
the  lake ;  he  also  states  ||  that  the  shores  are  strewn 
with  shells  of  Planorbis^  Limnaea  and  Bithinia  as  well 
as  the  above-mentioned  genera. 

With  regard   to   Lake  Van,   so   far  as  my  own 

*  Demkschr.  Natuff.  Vtr.  Jtiga^  1870. 

t  SitMungsbtr.  i.-i.  Akad.  Wiss.^  JlfaiAem.^naiufw.  CUuse^  xcvii. 
S03. 

t  Verk.  Nat.  Vkr.f  Bonn,  1886,  SiiMung^er.  19. 

§   Verh.  k.-Jk.  gwt.  Rikhsanst.  ^884,  981. 

n  yerh.  Nat.  V^*%  Bonn,  1884,  173. 


47^  Geology  of  Armenia 

observations  on  the  W.  coast  are  concerned  (pp.  157* 
163)  there  is  no  trace  of  any  shells  in  the  high-level 
beach-deposits.  Loftus,^  however,  noticed  in  the 
precin(5ls  of  the  citadel  of  Van  '*many  blocks  of 
comminuted,  white  shell-sand,  in  which  is  a  minute 
species  of  Myttlus.  The  last  is  probably  derived  from 
the  shore  or  from  one  of  the  islands,  and  of  very 
recent  formation.  At  present  there  are  no  molluscs  in 
the  lake.*'  He  is  doubtless  referring  to  a  Dreissensia- 
limestone,  similar  to  that  which  I  found  in  situ  near 
Demian,  at  the  W.  base  of  Sipan  (p.  173). 

Reference  may  here  be  made  tQ  the  3  old  beach- 
lines,  which  I  observed  clearly  marked  on  the  S.  coast 
of  Lake  Van  (p.  102),  at  about  15,  40  and  100  feet 
above  the  present  level,  and  to  the  recent  travertine 
with  walnut-leaves,  which  I  found  in  the  Gozel-dereh, 
at  the  S.W.  angle  of  the  take  (p.  |o6). 

Pleistocene,  lacustrine  dqx)sits  are  well  developed 
in  the  basin  of  the  Arazes.  Near  the  upper  end  of  the 
broad  Araxes  depression,  near  Erivan,  Abich  t  found 
loose,  travertinous  limestone  with  Dreissensi^  P^^y^ 
morpha^  Pallas,  "  immediately  overlain  by  the  doleritic 
lavas  [basalts],  which  extend  from  the  volcanic 
highlands  of  Alagdz  down  to  the  plain.^ 

Lacustrine  bed^  also  occur  on  the  S.  side  of  the 
Araxes  plain,  in  the  G^ngormas  valley  (N.W.  of 
Kulpi.  Here  Abich's  secftion  %  shows  isolated  patches 
of  horizontal,  travertinous  beds,  overstepping  Upper 
Miocene  sandstones.  The  beds  are  pink  with  green, 
glauconitic  grains,  and  contain  casts  of 
Viviparus ;   Valvata  piscinalis. 

*  Quaff:  fffum,  Gtol.  Soc.  zL  311. 

t  Das  Sti$naat%  &v.,  Mem.  Acad.  ScL  Sc  Pftenbw,  a£r.  6»  Set 
math.  &  phys.  vii.  4 ;    his  VergUukitfdi  gM.   Grwmdtifgi  Av.    157 ; 
and  Quari,  Jamm.  Gtd.  Soc*  vL  (2)  47.. 

X  GioL  Fpnck.  m  din  kmk.  Zdndirm,  vL  X4S. 


473 

In  another  sedHon  figured  by  Abich^  of  the  edge 
of    the    neighbouring    Kuipi    valley    he    depidb 

'^jjoearly  horisontal,  fine  volcanic  conglomerates,  ashes,  rapilli,  sands 
and  tub  of  Diluviiil  age,  which  show  that  at  the  time  of  the  volcanic 
outbreaks  on  the  heights  between  Takyaltu  and  Anurat  the  Araxes 
pUtn  was  covered  iby  water.  These  tufi  are  covered  up  by  doleritic 
and  andesstic  lavas.** 

Further  N.,  Abichf  has  figured  a  section  of  the 
Ama-chai  at  Alexandropol  (5033  feet),  in  which  Nos. 
3  &  A  have  manifestly  been  transposed.  The  sedlion 
should  (I  think)  run  as  follows  : — 

I.    Blade  earth,  generallj  corering  the  plateau  of 
s.    Black  and  red,  ctnderjr  lava  (tuff-lava  of  Abidi). 

3.  Travertinous  beds  with  Dreisstnsia  pofymorpha^  Pallas. 

4.  Trachytic  and  andesitic  tuffs,  bedb  of  fine  gravel  (pebbles 
of  s^pentine  and  gabbro  from  Aglagan),  and  argillaceo-calcareous 
marls  with  bones  of  Mastodon  and  of  indeterminate  Camivoions 
and  Cervine  remains,  which  Abich  considers  to  indicate  a  fauna 
of  the  Sub-Apennine  period.    All  the  beds  are  horizontaL 

Pohlig4  however,  states  that  a  large  calcaneum 
(0.27x0.19  m.)  of  [Elephas  ^migenius  was  found  at 
Alexandropol  and  is  now  in  the  Tiflis  Museum. 

Karakash§  is  more  explicit  in  his  account  that 

^  the  first  discovery  of  the  mammoth  (EUphas  primtgemius)  in  the 
Litde  Caucasus  comprised  remains  of  the  tusks,  molars  and  other 
parts  of  die  skeleton,  together  with  the  tooth  of  a  hot^it  (Eqwts 
oabaUus)^  found  in  a  great  cutting  on  the  Tiflis-Kars  railway,  at 
172  versts  from  Tiflis  [i.^.,  on  the  plateau  of  Alexandropol].  This 
cutting  was  in  a  slope  S4  metres  above  the  level  of  th«  river-bank, 
formed  of  old  alluvium  of  variously  coloured  sands,  with  pockets 
of  volcanic  taiL** 

Lower  down  the  Arpa-chai,  at  Kigach,  below  the 
confluence  with  the  Kars-chai,  Abich  ||   has  figured 

*  Gtoi,  Fonch.  in  den  kank.  Ldndtm^  ii.  i6o. 

t  Ihid.  41,  4S. 

X  Virh.  Nat  Ver.^  Bonn,  1886,  StiMUHgsder.  48. 

{  /mm.  X.    Congress  of  Russian  NatuMists  6v.   Na   7,   341  • 

Kiew,  z8  97*    [In  Russian]. 

II  Op.  cii.  XL  43* 


474  Geology  of  Armenia 

the  following  se<5lion  of  the  river-cliff : — 

I.    Red  and  bbck,  streaky,  dndery  la^a  ["  tnff-Iava  "  of  Abich], 
in  wfaidi  the  bladcy  pitchstone-lQLe  slieaks  prepondcnte.     •••    20  feet 
9.    TVachjtic  tufb. 

3.  A  considerable  soooession  of  calcareo-ai^Haceons  marla  wiA 
Dmssimsia  fafywMfka, 

4.  Basaltic  toff  and  conglomerate,  not  so  thidi  as  No.  5. 

5*    Basalts  with  large  steam-holes  filled  with  tiavertiiie.40-50  feet 

Abich  ^  also  found  the  same  travertinous  beds  with 
Dreissensia  polymorphc^^  underl3ring  doleritic  lava,  at 
Ani  on  the  Arpa-chai.  Hamilton.t  in  travelling  up 
the  Arpa-chai  from  Ani  to  Alexandropol,  observed  the 
same  series  whilst  riding  northwards^  past  Aras  Oghlu 

«<  over  an  elevated  tableland  of  volcanic  scoriae  and  lapillL  A  little 
beyond  Maurek  a  thin  bed  of  pale  yellow  sand,  filled  with  nnmeroos 
shells  of  Myiilus  [/.«.,  I}reissensta\^  overlies  a  bed  of  concretionary, 
calcareous  marl;  these  beds  dip  a  little  to  N.W*  under  the  bla^ 
peperite  with  which  the  neighbouring  hills  are  capped,  and  oontain  no 
traces  of  volcanic  matter.** 

In  proceeding  from  Alexandropol  "over  the  same 
undulating  country  to  Kars/'  Hamilton  (I.e.)  "soon 
came  upon  the  same  shelly  formation  just  W.  of 
Uzunkilissa."  At  Pyrwaly  (4734  feet)  on  the  Kars* 
chai  the  river — according  to  Abich  J — ^flows  through 

''a  narrow  gorge  with  walls  so-80  feet  high,  diowing  several  sheets 
of  lava,  one  above  the  other.  Here  and  there  are  indications  of 
the  same  travertinous  limestone  with  Drmsensia  pefymoffka^  whidi 
near  Hamsakaiak  [$  miles  S.W.  of  Pyrwaly]  b  a  fiithoms  thic^ 
passing  into  whiter  hemicrystaUine  limestone  with  N.E.  dip.*" 

Between  Kars  and  Khorasan  Abich  §  noticed  that 
*'  the  calcareous  Myttlus  [Dreissensia]  marls  and  tufis 
become  visible  beneath  the  sheets  of  dolerite,**  whidi 
form  the  surface  of  the  plateau.  ^.    ^. 


*    Verglekkindi  gtoi.  GrundMuge  ^c.  157. 
t  Reseanhes  in  Asia  Minor^  L  S03. 
X  GeoL  Fanch,  in  dm  kauk.  Landem^  it  173. 
§  Ihid.  56. 


Pleistocene  475 

The  Ibllowing  succession  (which  I  have  compiled 
from  Abich's  writings  *)  exists  between  Jermik  (5342 
feet)  and  Khorasan  (5059  feet) : — 

1*  "Whiter  dusty,  trachytic  tufb,  with  fine  conglomerates  of 
pmnice  and  tuC  Amongst  the  tuf&  a  ydlow-brown  shell-sand  occurs 
with  Dreisstmsia  folymarfha^  NeriHna^  small  Viviparus^  Rissoa  and 
PlanoHns.  Hamilton  t  also  states  that  the  upper  bed  of  the  plateau 
between  Khorasan  and  Jermik  is  horizontal  and  contains  comminuted 
shells,  belonging  chiefly  to  one  species  of  Mytilus  \pT€iss€nsid\  and 
a  few  univalves. 

a.    Trachytic  tuflb,  like  sandstones,  with  />.  pofymorpha, 

3.  Light  yellow-ochre  strata,  alternating  with  white  beds,  which 
have  become  consolidated  by  travertine  at  Khalil  Divana  and  Sogutly, 
and  are  cleaved  into  slabs  for  tombstones  and  building  purposes. 
These  fissile  sandstones  occasionally  contain  carbonized  remains  of 
reed-like  grasses,  and  to  the  K  of  ELhorasan  predominate  in  the  series. 

4.  Pudding  stone  conglomerate  cemented  by  travertine,  with 
great  quantities  of  rolled  pebbles  of  rodcs  still  in  situ  in  the  Aghri 
Dagh  range  on  the  S.,  €.g,^  Miocene  and  Cretaceous  limestones, 
syenite,  gabbro,  porphjnries,  quartzites,  chlorite-schist,  trachyte  and 
olivine-basalt.  This  conglomerate  occurs  from  Hassankaleh  west- 
wards to  Devehboyun  and  northwards— according  to  Hamilton  (/.r.) 
— to  ELargabazar  and  the  Sogaaly  plateau. 

Parallel  beaches  of  this  old  lake  of  Pasin  have  ^een 
noticed  to  N.  and  £.  of  Hassankaleh  by  Hamilton,:!; 
in  some  cases  *'  extending  for  nearly  2  miles,  following 
the  sinuosities  of  the  hills." 

To  the  S.,  at  Choban-kopri  on  the  Araxes,  Abich  § 
has  figured  a  sedtion  of  the  river-bank,  showing : — 

X.    Greenish  Drdssensia  marl. 

s*  Yellowish  white,  compact  marl,  unfossiliferous,  both  dipping 
$o*S. 

There  is  no  break  in  geological  continuity  between 
the  districts  of;  Pasin  and  of  Garin  or  Erzerum.     The 

^  GioL  Fonch,  in  den  kauk.  Landtm^  ii.  59,  6S-3,  69,  76,  134-136 ; 
and  his  Var^tdchindt  geoL  GrundMUge  &*c,^  158. 

t  Rutarehis  in  Asia  Minor ^  L  x86. 

X  Ibid.  183.185. 

)  Gtol.  FerscJu  6v;,  ii.  66. 


47^  Geology  of  Armenia 

Deveh-boyun  ridge,  which  separates  the  two  plains,  is 
composed  of  the  same  white,  lacustrine  maris;  but 
its  presence  as  a  dividing-ridge  is  due  to  its  being 
capped  here  and  there  by  andesitic  lavas,  which  have 
issued  along  a  N.N.E.-S.S.W.  fissure. 

According  to  Abich^  these  calcareous  marls  with 
Dretssensia  polymorpha  occur  at  the  same  level  along 
the  whole  margin  of  the  plain  of  Erzerum,  and 
also    that    on    the    south    side    of    the    plain 

''the  hot  springs  of  Uija  (5740  feet)  issue  out  of  calcareous  mails 
filled  with  Drmsensia^  underlying  a  tradiyte-poiphjry  conglomerate 
with  a  travertinous  cement,  which  forms  the  surface  of  the  plain  up  to 
Aigovao  (£•  of  Ilija),  and  borders  the  lavas  of  the  Aghmeser  volcano.'* 

Koch,t  however,  observed  that  beneath  a  sheet  of 
"  trachyte,*'  forming  a  terrace  behind  Ilija,  there  lies 

''an  Infusoria-bedy  outwardly  very  similar  to  aJunite;  it  forms  a  bed 
4-6  feet  thick,  and  is  visible  for  about  20  minutes  along  the  maipn  of 
the  terrace  but  is  probably  of  greater  extent.  •  •  It  is  a  fine,  whidsh 
grey,  consisting  of  amorphous  diatom-flour  for  about  half  its  volume." 

£hrenberg:|:  thinks  it  possible  that  originally  it 
consisted  entirely  of  diatom-f rustules  and  that  perhaps 
the  heat  of  the  overlying  lava  caused  them  to  break 
up  into  very  fine  particles.  However,  he  was  able 
to  determine  the  following  diatoms,  three  of  which 
were  new  species : — 

Achnanthes  (?);  Amphora  angusta^  A.  lybica; 
Campylodiscus  clypeus ;  Cocconeis  piacentula ; 
Discoplea  compta ;  Eunotia  granulata  (very  ab- 
undant),  E.  monodon  (?),  E.  zAra ;  Fragilaria 
pe£tinaUs^  (very  abundant),  F.  rhabdosoma  (?); 
Gallionella  cYenulqtOy  G.  granulata  \  Gompho- 
nema  gracile ;  Navicula  silicula,  N.  Semon  (?) ; 
Pinnularia  esox  (?),  P.  Kochi  (n.  q>.) ;  Surirdla 

*  GmL  Farsch*  in  dm  kauk.  Lmdam^  xL  73,  99,  roa;  and  iA 
Das  Stetnsalt  dr*^.,  4. 

t  Riise  impanHschm  GMrgt^  ii.  304-505. 

X  Monaisbcr.  BtrKn^  Akai*  Wiss.^  1S4S,  jvo-i* 


Pleistocene  477 

cordata  (n.  sp.),  S.  Itbrula ;  Stauroneis  lineolata. 
Lithodantium  furcatum ;  Lithostyltdium  amphiodon^ 
L.  clepsammidium,  L.  quadratum^  L.  rude^  L. 
sinuosum  (n.  sp.). 

All  these  are  freshwater  forms,  the  majority  being 
worldwide  at  the  present  day.  Pinnularia  esox  and 
Stauroneis  lineolata  have  hitherto  been  known  only 
in  America. 

According  to  Abich*  (corroborated  by  my  own 
observations,  pp.  43-5)  the  succession  in  the  W.  part  of 
the  Erzerum  plain  is  as  follows : — 

X.    White,  lacostxine  mads  and  limestones   with  Drdsiemsia 

9.  A  oonddexable  thic^ess  of  nnfossiliferoos,  [blue]  days  and 
light  grey,  loamy  maris. 

3.  Compact  coaise  psammltes  with  elements  of  the  older, 
non-volcanic  rocks  siurounding  the  plain  (cp.  my  observations  at 
Maimansor  ^  ajo). 

The  W.  limit  of  the  old  Erzerum  lake  is  marked 
hy  the  Miocene  limestone  barrier  of  Kardarich,  now 
pierced  by  the  Frat. 

My  observations  have  greatly  extended  the  known 
area  c^  these  recent  lake-deposits.  It  will  avoid 
needless  repetition  to  briefly  r^er  to  the  lake-deposits 
(with  minute  shells  of  Spherium  (Cyclas)  and  Bitkinia) 
of  Khedonun  (pp.  55,  56),  on  the  upper  Araxes,  and  on 
the  northern  and   western  outskirts  of  the  Bingol 

Dagh  (pp.  57,  58,  233-4,  ^^7)'  T^  ^^  ^-  ^^  *^^ 
Bingol  Dagh  I  found  the  lake-deposits  to  be  well 
developed  m  the  plain  of  Khinis  (pp.  65-70),  and 
remains  of  Elephas  armeniacus  have  been  found  at 
Sharvoon,  near  Khinis,  in  beds  of  this  series  associated 
with  Dreissensia  polymorpha  and  Bitkinia.  The  lake- 
deposits  extend  still  further  E.  into  the  neighbouring 
plain  of  Gopal.    To  the  S.  of  the  Bingol  Dagh  the 

^  Geal.  Forsch.  in  dm  tank.  Umdim,  iL  lOO. 


fe^ 


478  Geology  of  Arm§nia 

lake-deposits  reach  a  considerable  thickness,  with 
interbedded  sheets  of  lava  (pp.  197-201).  At  Gun- 
demir  I  found  in  this  series  a  pipeclay  with  carbonized 
reed-like  leaves,  which  occasionally  form  thin  seams 
of  lignite  (p.  20i\  To  the  E.  of  the  Bingol  Dagh  the 
great  mass  of  Khamur  is  essentially  compo^  of 
these  lake-deposits  with  interbedded  lavas  (185-6).  I 
was  also  able  to  trace  this  series  further  up  the  Murad 
valley  into  the  broad  Melaskert  plain  (pp.  180-5, 173- 
7,  92-5).  Between  Gop  and  Shakhberat  Viviparus 
viviparus  and  Unto  tumidus  are  associated  with  the 
numerous  individuals  of  Dreissensia  polymorpha  (pp. 
180-5).  In  the  heights  bounding  the  Melaskert  plain 
on  the  S.  the  succession  at  Demian  is  given  on  pp.  94- 
173-49    and   here   I   found   small   individuals  of 

elanopsis  and  Bithinia  associated  with  the  abundant 
Dreissensia  polymorpha. 

It  seems  probable  that  the  Dmssensia-limestone  is 
indicated  in  the  Dersim  by  Taylor's  statement  •  that 

**al  the  top  of  the  hill  between  Khosat  and  Avzoonik  [Arsonik]  there 
was  a  mast  of  limestone  full  of  shells  of  every  description  Ae  majority 
being  small  bivalves  of  the  mussel  spedesi  some  o^  whidi  seemed 
as  if  onlj  then  taken  from  the  water/' 

Lacustrine  deposits  have  been  recorded  to  the  N.W. 
of  Erzingan,  between  the  Chimen  Dagh  and  Orchil 
Dagh.  Here,  in  the  valley  of  the  Chak  Suyu,  a 
left  tributary  of  the  Kelkid  river,  Hommaire  de  Hell 
found  "  a  lake-deposit  with  Cyclas^  large  Viviparus 
and  an  Anodonta^^  according  to  Viquesnel,t  who 
identified  the  fossils.    They  were  found  to  occur  in 

^'a  marly  limestone  containing  greenish,  argillaceons  portions;  it 
rests  on  conglomerates  and  is  overlain  by  ft  rather  ^idL  bed  of 
limestone/' 

The  brackish- water  deposits  of  Meliksherif  are  only 
separated  from  the  Chak  Suyu  valley  by  the  Ordiil 

*  Jcum.  Roy,  Gtogr,  Soc.  xxzviiL  341. 
t  BuUn  Soc,  gioL  Frana^  sdr.  t,  vii.  jio. 


Pleistocene  479 

Dagh.  At  a  spot  about  3^  leagues  W.  of  Meliksherif 
TchihatchefiE  *  found  horizontal,  white  limestones, 
alternating  with  a  conglomerate  of  limestone  and 
serpentine.    The  conglomerate  dominates  to  the  W., 

^  at  fixst  inooherent;  afterwards  compacted  into  a  fine-grained  rocki 
passing  into  a  pinkishi  cellular  limestone,  full  of  Uttle  individuals 
of  Cardium  tvatum^  Desh.,  and  other  broken  shells  recalling  the 
Aialo-Caspian  type.'* 

In  the  region  of  the  Tortum-Olti  depression  lakes 
seem  to  have  occurred,  of  which  the  Tortum  lake 
is  the  only  existing  representative.  Hamilton,!  in 
riding  up  the  Nariman  valley  to  Id,  considered  that 

'*  the  plain  above  Nariman  for  6  or  7  miles  to  S.S.W.  was  evidently 
once  a  succession  of  3  lakes,  widi  horizontal  deposits  of  great 
Aickness,  sepa^ted  by  narrow  passes  now  broken  through.  In  many 
places  horizontal,  parallel  roads  occur  in  the  upper  basins." 

To  the  £.,  on  the  Armenian  plateau,  the  distridl 
of  Goleh  indicates  by  its  name  that  the  upper  Kur 
valley  must  have  been  a  lake  even  in  historic  times, 
and  legends  are  yet  exitant  to  this  e£Eedl;  it  is  still 
very  marshy  near  Unit. 

To  the  N.E.,   on  the  shores  of   Lake  Khozapin, 

"the  same  Dreissemsia  poiyttufrpha  marls  occur  as  at  Erivan,  the 
foot  of  Alagoz  and  in  Shor^l  [p.  473] ;  here  the  lavas  flowed  direcdy 
into  tite  shell-bearing  marb,  which  alternate  with  tcavertinous  lime- 
stones. The  fossiliferous  beds  have  been  burnt  red  by  the  lavas 
at  the  point  of  contact ."% 

Further  E.,  in  Somketia,  the  present  basalt-plateau 
of  Lori  seems  also  to  have  been  once  a  lake-basin. 
Here  Abich§  has  recorded  "a  fine,  calcareous  shell- 
conglomerate,  in  which  Dreissensia  polymorpha  and 
Cardium  catillus  (of  the  Aralo-Caspian  formation) 
predominate/' 

Remains  of  a  mammoth's  tusks  have  been  found 

^  Asu  Mimiun^  GhUgie,  iiL  i63*4. 
t  JtoMTcAit  m  Asia  Minor^  L  six. 
X  Abich,  GioL  Fpn^  im  dm  kauk.  Landim^  iiL  ss. 


480  Geology  of  Armenia 

in  this  neighbourhood — aocording  to  Karakash*-Ha 
the  Tiflis-Kars  railway,  at  115  versts  from  Tiflis,t.f., 
near  the  confluence  of  the  Lori  and  Debeda  rivers, "  io 
a  brown,  ferruginous  conglomerate/' 

An  important  deposit  of  tripolite  in  recent  lake- 
deposits  was  foimd  by  Abichy  at  Kissatip  in  the 
Uravel  valley  {sl  left  tributary  of  the  Kur,  a  litt*e 
above  AJdialzik).  It  occurs  in  beds  (the  thiclest 
about  8  feet),  alternating  with  trachytic  tuffs.  Tlie 
analysis  of  the  rock  is  very  similar  to  that  of  Bilin. 

Lower  down  the  Kur — according  to  FoumierJ— 

**the  whole  part  of  the  Kur  basin  between  Suiam  and  Mtskhet 
has  been  occupied  since  the  banning  of  the  Pliocene  by  lakaj 
in  which  considerable  masses  of  conglomerate  accumulated.  These 
lakes  probably  persisted  down  to  the  historic  period,  for  a  Geot^an 
tradition  relates  that  the  castle  of  queen  Tamari  (twelfth  centuxy)  was 
constructed  on  an  island,  now  the  dtadel  of  Gori.  These  lakes  were 
only  completely  drained  away  by  Uie  erosion  of  the  present  valley 
of  die  Kur  between  Kaspi  and  Mtskhet  Another  lake  extended  S.E. 
of  Tiflis,  now  mai^ked  by  the  old  alluvium  of  the  plateau  of  Avlabar, 
Ag  Taklia  and  Lake  Kumiskoe.  For  a  long  time  the  lake  of 
Gori  and  Mukhran  emptied  into  that  of  Ag  Taklia." 

Although  there  are  numerous  traces  of  recent  beach- 
deposits  along  the  Black  Sea  littoral,  yet  only  one 
fossiliferous  instance  has  been  recorded.  According 
to  Hamilton  §  "just  E.  of  Platana  the  low  hills 
consist  of  a  soft,  shelly  limestone,  with  many  frag- 
ments and  casts  of  recent  or  post-Tertiary  shells  "  [not 
specified]. 

In  the  Border- Ranges  of  Armenia  the  recent  deposits 
seem  to  be  confined  to  the  travertines  of  calcareous 
springs.    On  the  S.  side  of  the  Taurus,  however, 

*  Jaum.   X.   Congress  0/  Itussbm  NaiKratisis  6^  No.  7»  H^- 
Kicw,  1897.    [In  Russian]. 

t  GW.  Farsch.  Ar  *»  kank.  ZAubm,  fii.  77. 

%  DiseripiUm  gsol.  dm  Qmass  imttmi,  tot-]. 

I  JUsiorcAes  in  Asia  IfSftor^  I  146. 


Glacial  A£tum  481 

there  appear  to  be  indications  that  a  large  lake 
existed  in  the  open  region  between  the  Tigris  and  the 
Taurus,  to  the  W.  of  the  Bitlis  valley.  In  the 
ravine  at  2^kh  (2k>rg)  Hommaire  de  Hell*  observed 

"horizontal  beds  of  marls,  compact  or  friable,  overlain  by  white 
limestone,  rather  thick  and  often  veined  with  pink,  similar  to  the 
marble  used  at  Diarbekr.  At  Zorg  it  also  encloses  pebbles  and 
becomes  a  breccia.  It  is  vtxz  hard  and  white,  and  encloses  fragments 
of  fossil  bones." 

It  will  at  once  be  evident  that  this  description 
strongly  recalls  the  marls  and  tu£fs  of  Maragha  (p. 
47o\  enclosing  Mammalian  bones,  and  associated 
witn  beds  of  travertinous  marble.  It  is  therefore 
not  improbable  that  2k>kh  may  yet  be  found  to  yield 
a  similar  Pliocene  fauna  to  that  of  Maragha. 

Further  W.  Hommaire  de  Hell  (Lc.)  found  that 
the  same  **  white  limestone,  overlying  conglomerate,  is 
revealed  in  the  gorge  of  the  Yezid  river,  and  that 
nearer  Scorsey  (^rsy)  a  plateau-height — the  Dasht- 
i*Kiri — is  formed  by  a  sheet  of  columnar  basalt, 
resting  on  horizontal  clays,  sands  and  coarse  con- 
glomerates, which  continue  as  far  as  Diarbekr/' 

To  the  N.W.  of  the  Dasht-i-Kiri  lies  Miafarkin, 
situated — according  to  Chantref — "at  the  foot  of  a 
limestone  mountain,  with  enormous  numbers  of  shells, 
Helix  being  present  amongst  others.  The  soil  is  of 
morainic  origin  and  very  fertile.'*  Major  Maunsell 
has  also  informed  me  that  at  Miafarkin  he  noticed 
more  fossil  shells  than  anywhere  else  in  Armenia. 

n;.     GLACUL  AcnoN. 

Reference  may  here  be  made  to  pp.  216*219  for 
m^  discovery  of  tiie  former  existence  of  glaciers  on  the 
Bmgol  Dagh. 

^  Vcyagim  T^trptUetc.^  CrUiagU^  iv.  87-^ 
t  B^ffouti  TyiiSt  Tour  du  Monde  Iviii.  250. 

sx 


483  Geology  of  Armenia 

On  the  N.  side  of  Ararat,  however,  a  small  glarier 
still  exists  at  the  head  of  the  great  Akhury  chasm, 
descending  to  as  low  an  altitude  as  8000  feet— "a 
notable  fadl  when  we  consider  that  the  line  of  per- 
petual snow  on  this  side  of  Ararat  is  as  high  as 
14,000  feet."*  According  to  Abichf  the  glacier  60 
years  ago  only  extended  down  to  9172  feet,  but 
"large,  longitudinal  ice-hills,  slightly  covered  with 
d^bns,  lie  in  uninterrupted  connexion  with  it."  They 
had  only  partially  melted  to  2/3  their  size  in  20  years 
since  the  great  debacle,  which  overwhelmed  the 
Akhury  (Arguri)  monastery  in  1840 ;  in  Abich*s  words  J 

**  the  colossal  erratic  block  (aSa  feet  in  drcumference  and  4a  feet  hi^ 
in  1844)  must  have  travelled  at  least  7  miles  (11  vexsts).  The 
mud-stream,  which  only  began  to  flow  7  a  hours  alter  the  earthquake, 
must  have  had  a  breadth  of  half  a  mile  and  a  height  of  at  least  50  feet 
at  a  distance  of  z  z/3  mile  (a  versts)  below  Arguri  [Akhuryl." 

On  the  W.  side  of  Ararat — according  to  Abich§— 

^<a  broad  ravine  cuts  deeply  into  the  flanks  of  the  Kippgol  cone 
and  is  occupied  by  a  quite  imposing  glader,  with  very  blue,  banded 
ice,  issuing  between  high  moraines." 

Chantre  ||  passed  near  the  same  locality  in  crossing 
the  Karabula  col  between  Bayazid  and  Igdir,  finding 

**  fresh  moraines,  still  clayey,  at  the  foot  of  one  of  the  numerous 
glaciers  of  Ararat.    Great  n^v^  occur  here." 

In  descending  to  the  plain  he  noticed  numerous 
erratic  blocks  and  traversed  moraine-hillocks,  between 
which  are  innumerable  Idtkes  with  black  water.  In 
several  places  lavas  alternated  with  morainic  debris. 
In  riding  from  Igdir  to  Erivan  he  crossed  numerous 

*  Lynch,  Armema,  i.  196. 

t  Geoi.  Beob,  mw,  Kur  6r*  Araxts^  ss ;  and  Us  Vgr^ldchadi  gtsl 
Grundsuge  ^^.,  161,  and  GeaL  ForsA.  in  den  kauk.  ZamterM^  iL  A^** 

t  GioL  Beob.  ^c,  24. 

$  GeoL  Forsch,  &»c,  iL  455. 

II  Btynmt  d  Tiftis,  Tour  du  Monde,  Iviii.  S98. 


Glacial  AAiom  483 

moraines  full  of  obsidian  fragments. 

Alagdz  on  the  N.  side  of  tfie  great  mid-Araxes 
depression  has  also  been  the  birthplace  of  glaciers. 
Even  at  the  present  day — according  to  Pastuldioil  * — 

"  in  the  gorge  between  the  N.  and  W.  somnrfts  a  small  glacier  still 
oocnrs.  The  snifioe  of  the  snow-maf  aei  V  Alsgos  amounts  to  5.  i  sq. 
versts.  Fonnerly  dx  large  main  gladclv .  existed  besides  some 
secondaiy  gladeis,  which  hare  now  ,qaite  finished,  leaving  only 
mocaines.  The  4a  little  freshwater  lakes  on  the  slopes  and  at  the  foot 
of  Alagoa  probaUjr  stand  in  some  relation  to  this  circumstance." 

Pastukhoff  has  also  calculated  that  the  former  size 
of  the  Alagdz  glacier  was  8  kilom.»  with  a  thickness  of 
900  feet. 

Moraines  were  noticed  by  Freeh  and  Arthaberf 
in  the  valle]^  of  the  N.  slope  of  tiie  Gokcha  ranges, 
e^.,  at  Delejan  (Delishan),  N.  of  the  Semenovka  pass. 
Ine  heights  of  Eshak  Meidan  also  show  morainic 
slopes  according  to  Chantre4 

Palgrave  §  asserts  that  he  found  numerous  traces  of 
glacial  adtion  in  the  region  of  the  Chorokh,  but  his 
statements  are  vague  and  too  few  of  his  localities 
can  be  identified. 

On  the  other  hand,  Major  Maunsell  kindly  informs 
me  that  moraines  and  glacial  lakelets  occur  on  the 
broad  Keshish  Dagh  (E.  of  Erzingan),  and  on  the 
Shaitan  Dagh  (also  about  12.000  feet),  W.  of  Bingol. 

In  the  Taurus  short  glaciers  are  known  on  the  Jelu 
Dagh. 

^  Asami  of  Alai^^  M6n.  Cane.  Sec  Imp.  Rnss.  Geol.  Soc,  1896. 
[In  Russianj. 

t  I>as  PatSotafcmm  in  ffaeharmimin  6f  Persim^  BeiU.  Pal.  (Estr.- 
Uog-  3dL  ^69,  i86. 

%  Bij^rmii  i  ftjBst  Toor  dn  Monde,  Iviii.  298. 


APPENDIX 

KOTBS  BT  PROFESSOR  T.G.  BONNBT,  F.R.S.,  ON  ROCKS 
COLLBCTED  BY  MR.  H.F.B.  LYNCH  IN  1 893 

I^OFBSSOR  Bonnev  has  kindly  allowed  me  to  publish 
hb  descriptions  of  these  specimens  with  the  exception 
of  those  n:om  Ararat,  which  have  already  ap(>eared  in 
the  Geological  Magazine  (Dec.  v.,  vol.  ii.  52-58).  I 
have  quoted  the  descriptions  of  the  Cretaceous  and 
Miocene  limestones  on  pp.  339,  447-8.  The  remaining 
notes  are  as  follows : — 

X.  **  On  the  track  leadmg  behind  the  Ak-kol  [Ahkiul]  Dagh,  on 
the  S.  coast  of  Lake  Van  [cp.  pp.  x  x  i,  375].  A  piece  of  vein-quarts, 
a  little  impure,  slightly  stained  with  limonite.  Also  a  dark,  lead- 
coloured,  very  finegrained  schist  of  rather  fibrous  aspect  Thisi  under 
the  microscope^  consists  of  a  dear  mineral  studded  with  opadte,  filmy 
brown  mica,  and  minute  films  of  the  same ;  it  contains  also  some 
grains  of  iron  oxide,  crystals  of  rutile  and  numerous  spots  or  ill-formed 
prisms,  with  dark  granular  centres  but  dear  towards  the  edges,  the 
extinction  is  probably  oblique,  and  perhaps  the  mineral  is  a  felspar. 
The  rock  has  undeigone  severe  pressure,  and  its  structure  is  suggestive 
of  strain-slip  cleavage,  but  probably  it  never  was  very  coarseiy 
ciystalline. 

s.  On  the  road  behind  Akhtamar,  on  the  S.  coast  of  Lake  Van 
[q>.  pp.  I XX,  175].  A  dark  green  serpentine  with  some  lighter 
streaks ;  with  oblique  slickensided  surfaces,  sometimes  pairtially  coated 
with  cream-coloured  or  green  steatite,  giving  to  the  spedm-^n  a  rather 
lenticular  form.  Under  die  microscope  the  rock  shews  the  usual  signs 
of  severe  crushing,  with  the  remains  of  bastite  crystals  an^  powdered 
/    magnetite  more  or  less  in  streaks.     It  must  originally^  have  betn 


488  Geology  of  Armenia 


9» 


ft 


AWKB9.  H.    Das  mesldichc  oder  karthfi-ineieCin.  Grinxgebiige  fa. 

Bull.  i>hjt.-iiiath.  Acad.  Sd.  ix.  29-45.    St  Fetenb.  1S51. 
Notice  e]q>licative  d'one  Toe  du  cooe  de  FAnnt    BoO. 

Soc  G^ogr.y  s6r.  i|  ir.  515-15. 
(Joumejr  from  Len£dran  to  Savalan).    PpggendoffTi 

AnnaL  d.  Phys.  &  Chem.,  Izzv.  157.    Leipz^,  185 1. 
Pt  Observations  stir  le  mont  Ararat,  B.S.G.F.|  sdr.  a,  m. 

s65-7z.     Paris,  1851. 
H  Der  Salzsee  v.  Urmiah.    Z.D.G.G.,  tL  356.    Berlin, 

1854, 
„  Demiers  tremblements  de  terre  dans  la  Pene  septentr. 

&C.    Bttll.  phys. -math.  Acad.  ScL  ziv.  49-7s.    St  Petenb., 

1856. 
ff  Schwefelreiches  Tufgestein  in  d.  Thakbene  v.  Diadin. 

Bull.  ph3rs.-math.  Acad.  Sd.  xiv.  142-4;   and  Mfi.  phys. 

&  chim.  ii.  412-6.     St.  Petersb.,  1856. 
fp  Man^anerze  in  Transkauk.    BuU.   ph7S.-matL  Acad. 

ScL  xvL  303-25 ;    and  MdL  phys.  &  chim*  iii  317.   St 

Petersb.,  1858. 
9,  Tremblements  de  terre  1  T^bris  en  sept  1856.    BolL 

phys. -math.  Acad.  Sd.  xvi.  337-52;  and  li£L  phys.  & 

chim.  iii.  349-68.     St  Petersb.,  1858. 
If  Vergleichu  chem.  Untersuch.  d.  Wasser  d.  casp.  Meere^ 

Urmia  &  Wan-Sees.     M^m.  Acad.  Imp.,  s6r.  6,  ScL  math. 

&  phys.,  vii  (ix),  1-57.     St.  Petersb.,  1859. 
,,  Das  Stdnsalz  8c  sdne  geol.  Stellung  im  mss.  Armenico- 

M^.  Acad.  Imp.,  s6r.  6,  Sd.  miuh.  &  phys.,  vii  (ix),  59* 

150.     St  Petersb.,  1859. 
>f  Vergleich.  geoL  GrundzCige  d.  kauk.,  armen.  &  nord- 

pers.  Gebirge,   8cc    (Prodromus).     Mdm.  Acad.  Imp*! 

sir.  6,  Sd.  math.  &  phys.,  vii  (ix),  301-534.     St  Petersb., 

1859. 
„  (Letters  to  C.  Ritter  ;&  G.  Rose).    Z.D.G.&  xL  4^0^* 

Berlin,  1859. 
„  Recherches  g^ol.  en  Transcauc    Bull.  Acad.  Sd.,  L 

449-52.     St  Petersb.,  i860, 
ft  Voyages  en  Geoi|;ie,  en  Turquie  ic  en  Peise  en  i86s. 

B.S.G.S.,  s^.  2,  XXI.  2x5-20.     Paris,  1864. 
tt  Apercu  de  mes  voyages  en  Transcauc  en  x864«    Bull. 

Acad.  Sd.  v.  17.    St.  Petersb.,  1865;  and  in  Bull.  Soc. 

Imp.  Natural.  Moscou,  500.    Moscow,  1865, 
I,  Zur  Geologic  d.  siidostl.  Kauk.  (Gletsdierende  in  d. 

Jakobschlucht  am  grossen  Ararat).    Bull.  AouL  Sd.  vL 

21-40.    St.  Petersb.,  1866. 


Bibliography  489 

Abich,  H«    Tdlcan  de  Tandourek ;   gtologie  du  lac  d'Ourmiah. 

Bull.  Acad.  Sci.  vii.  119-24.    St.  Petersb.,  1867. 
„  GeoL   Beobacht.   auf  d.   Gebiigslandem  zw.   Kur  & 

Araxes.    Tiflis,  1867. 
99  Eine  im  casp.  Meere  enchienene  Insel.     Mdm.  Acad. 

ScLy  six,  7,  vi.  No.  5,  45.     St.  Petersb.,  1868. 
„  Die  armen.-geoig.  Traidiyte.     Verb.  k.-k.  geol.  Reichs- 

anst,  932-3.    Yiennat  1860. 
9,  Erdbeben,  vulk.  Erschein.,  &c.,  in  d.  kauk.  Landern. 

Mitth.  k.-k.  geogr.  Ges.,  iii.  166-75.     Vienna,  1869. 
„  Minend  springs  of  Tiflis.     [In  Russian].     Tiflis,  1870. 

y,  Die  Reihen-Vnlkangruppe  d..  Abul  &c.    Jahrb.  k.-k. 

geol.  Reicbsanst.,  xx.  275-8.    Vienna,  1870. 
99  Fulguriten  im  Andesit  d.  kL  Ararats  &c.      Sitzungsber. 

k.-k.  Akzd.  Wiss.9  Ix.  153-62.    Vienna,  1870. 
99  Der  Ararat  in  genet.  Beziehung  betrachtet.    Z.D.G.G.9 

xxii.  69-91.    Berlin,  1870. 
„  Ein  venneintl.  thatiger  Vulkan  an  den  Quellen  des 

Eupbrats.     Bull.  Soc«  Imp.  Natural.  M0SCOU9  xliii.  x-17. 

Moscow,  1870 ;  and  in  Peterm.  Mittb.  xvii.  71-3.    Gotha, 

1871. 
99  RrystalUn.  Hagel  im  tbrialetb.  Geb.  &c    TUSa,  187X9 

and  Vienna,  1879. 
99  Mineralog.  Forsch.  im  tbrialetb.   Geb.    M^m.  Acad. 

ScL    St.  Petersb.,  1871. 
,9  GeoL  Beob.  auf  Rdsen  im  Kauk.  im  Jahre  1873.    Bull. 

Soc  Imp.  NaturaL  Moscou,  xlvii.,  I.  278  &  II.  639  243. 

M0SCOW9  1874. 
9.  Das  tiirialetb.  Tbermalquellen-System  &c.    Z.D.G.G.9 

xzix.  820-9.     Berlin,  1877. 
,,  Mitth.  aus  d.  Kauk.    Verb.  k.-k.  geoL  Reichsanst,  29- 

35.    Vienna,  1877. 
99  Huge!  bei  Digala 'am  Urmia-See.    Verb.  k.-L   geol. 

Reicbsanst.,  67-9.    Vienna,  1877. 
9,  Geol.  Forsch.  in  d.  kauk.  Landera.    '>nenna9  1878-87. 

99  Die  Lage  d.  Schneegranze  &  die  Gletscher  d.  Gegen- 

wart  in  d.  Kauk.    Bull.  Acad.  Sd.  xxiv.   958-82.    St. 

Petersb.  9  1878. 
ti  Geology  of  Western  Armenia.    Translated  from  bis 

Geol.  Forsch.  &c  into  Russian  by  B.  Kolenko.    M^m. 

Sec.  Cauc  Soc  Imp.  Russe  de  G^ogr.9  xxL    Piatigor8k9 

1899. 
A1NSWORTH9  W.F.    Researches  in  Assyria9  Babylonia  &  Chaldsea. 
LoDdon,  1838. 


490  Geology  of  Armenia 

AiN8wosTB»  W.F.    Journey  from  Kusazydi  by  MalatU  to  BirepL 

J.ILG.S.,  X.  31 19  489.    London,  1840. 
„  Visit  to  the  ChaMrans,    J.R.G.S.t  xL  ai-75. 

London,  1841. 
„  Travels  Be  Researches  in  Asia  Minor,  Me90> 

potamia,  Chald«a  &  Armenia.    London,  184^. 
„  Personal  Narrative  of  the  Enphrates  Eaq;)editton. 

London,  1843. 
„  Sources  of  the  Euphrates.    Geogr.  Joutn.,  vL 

X73-7.     London,  1895. 
Albbrti,  F.  von.    Halurgische  Geologic,  L  893.    Stiittgart,  1853. 
AusHANiAN,  G,    Political  Geography  &,  Topography  of  Greater 
Armenia.    [In  Annenianl    Venice,  1853. 

AxTHULA,  D  J.    Kreidefossilien  d.  Kauk«    Beitr.  PaL  (Estr.-Ung. 
xit.  5S-I59*    Vienna,  189^. 

Arghiac,  a.  d'.    Histoure  des  progits  de  la  g6ologie,  5L  193,  8t 
V*  39X9  401-     Paris,  1850. 
Artbabsb,  G.  von.    See  I^ech. 

Abzruni,  a.    Reise  |nach  Siid-Kaukasien.    Verh.  Ges.  Erdkunde, 
xzii.  6oa-xi.     Berlin,  1895. 

Arzrunx,  G.    Economic  posiiton  of  the  Armenians  in  Turkey.    [In 
Armenian].    Tiffis,  1894. 

B1.RB0T  DS  Marny,  N.    Rock-salt  at  Kulpi.    [In  Russian].    Mat 
Geol.  Cauc,  sit.  2,  ii.  51-75.    Tiflis,  1888. 

Barth,  H.    Reise  v.  Trapezunt  nach  Skutari  &c.    Peterm.  MitdL 
Erganzungsheft.    Gotha,  i860. 
Batsrvitob,  L.     Geol.  description  of  the  district  of  Sharopan.    [In 

Russian].     Mat  Geol.  Cauc    Tiffis,  1876, 
„  Geol.  descr.  of  part  of  the  disticts  of  .Signakh  {& 

Tiflis.    [In    Russian].     Mat    Geol.   Cauc.    Tiffis, 
1878. 
„  Geol.  Researches  in  the  Batum  &  Artvin  districts, 

rin  Russian!    Mat  Geol.  Cane.,  1884,  and  set.  % 
1.  73-162.    Tiffis,  1887. 
Bklck,  W.     Die  Niveau-Schwankungen  d.  Goktschai-Sees.     Glo- 
bus, Ixv.  30X-3.    Brunswidc,  1894. 
,,  Reisebriefe   .    •    armen.  Expediticm*    Mittfa.  geogr. 

Ges.,  XV.  1-23,  &  zvi.  16-70.     Hamburg,  1899,  1900^ 
Belintaiytb,  L»L.    Journey  through  Russia,  1897.    iftoc   Ubrr. 
Durham  Phil.  Soc.  i  103-19.     Durham,  1898. 

Bbll,  cm.    Geol  notes  on  part  of  Mazanderan.     Trans.  Ge<^ 
Soc.,  ser.  2,  v.  577.    London,  1840. 

Bell,  M.S.    Around  &  about  Annenia.    Scott  Geogr..  Mag.,  vi, 
113-35.     Edinbuigh,  1890. 


Bibliography  491 

Slav,  O.    Vom  Unma  See  nach  d  Wan  See.    Pe^erm.  Mitth.,  ix. 
floi-xo«    Godia,  x863. 

BooDAHOvrxcB,  K.  Die  Orpgraphie  &  Geologie  d.  Nord-Persien. 
Petem.  Mitth.,  xiiL  183.    Gotha,  1867. 

B000SLOV8KT.  A.    The  Suzam  tunneL    Fin   Russian].    Joum.  of 
the  Ministiy  of  Communications,  376-96.    St.  Petersburg,  1888. 
^  Borannr,  T.G.    Some  rocks  from  Ararat    Geol.   Mag.,  dec   5, 
S.  51-8.    London,  x$D5« 

BoBHi^  G.  Ton  dem.  Der  Jura  am  Ostnfer  d.  Urmia-Sees.  Halle, 
1891. 

Bralt,  a.  Lei  mines  d'Allabyerdi.  Tchamlouk,  8c  Aktala.  Paris, 
1897. 

Bravdt,  J.F.  Die  von  Gobd  bd  Maragha  entdeckte  Saugethiere- 
reste.    Denkschr.  Natnrf.  Ver.    Riga,  1870. 

Brant,  J.    Tourney  through  a  part  of  Armenia  &  Asia  Minor. 
J.R.G.S.,  tL  187-293.    London,  1836. 
,9  Notes  of  a  joumej  through  a  part  of  Kurdistan.    J.R. 

G.S.,  X.  34i-43a-    London,  1840. 

CHANOOUMOia^  &  de.    Exploration  gfioL  d'une  partie  xAn  peu  con- 

nue  de  la  Turquie.    Comptes-rendus,  Acad. 
ScL,  xvifi.  897.    Parisi  1844. 
„  La  nature  des  eaux  du  lac  de  Van  &c 

CR«,  Acad.  ScL,  xxL  iizi.    Pari^  1847. 
Chajtibs,  E.    De  Beyxout  ^  Tiflis.    Tour  du  Monde,  IviiL  909- 
304.    Palis,  1889. 
„  I/Aiarat    Annales  de  gtegr.,  iiL  81^94.    Paris, 

1893. 
Chapxr,  M.    Un  ^Ite  cumenx  d'origine  rote  du  Cauc  mftidion- 
ale.    B.S.G.F.,  s6r.  3,  xxi.  No.  9,  101-9.    Pferis,  1893. 

Cbolst,  A.P.  de.  Voyage  en  Turquie  en  Asie^  Armdnic^  Kurdi- 
stan 8e  Mdsopotamie.    Puis,  xSja. 

Clayton,  EL  Mountains  of  Kurdistan.  Alpine  Journal^  xiiL  993- 
300.    London,  1887. 

CoRFi,  P.M.  Catastrophe  of  Kantsorik.  Q.J.G.S.,  xItL  39-5 ;  Ac 
GeoL  Mag.,  vii.  44.    London,  1890. 

CsARNoiTA,  J.  Sahend.  Jahib.  k«-k.  geoL  Reicbsanst,  iii«  No.  a, 
105.    Vienna,  1859. 

Dalykx,  R.A.O.  Earthquake  of  Eizerum.  J.R.G.S.,  xxxiiL  934- 
y.    London,  1863. 

Damsky,  a.  Analyses  of  andente  from  Borjom.  [In  Russian]. 
Mem.  Soc  Techn.,  No.  7-8,  4.    St  Petersb.,  1896. 

Datidofp,  p.  Gold  &  ooppor  in  the  Batum  province.  [In 
Russian].    Kutais  Govt.  Gasette,  L  149-73.    Kutats,  1896* 


ceotnlc  de  la  diaine  do 
S^i^^     Jot 

-_,,„jne  *i  Ciocue  centtal. 

i^3||;i^.  acn.    839-4a.    Put*. 

^Bi^cenAl.    Mttwfll^  1S96. 
'"''3^  die  PenUa  pronnces. 

ft    Notdaftika.     HeocB 

Jfi«*^J,  t94,  3fe,  385,  395- 

3g»ife'P>Ko»olcom   ID   Hoch- 
<iBMiB'  lii,  161.    Vleniu,  1900. 

^_, ft  Siudna. 

^SS^^<^  it.     %  Petenb.,  1894. 
^'^^UCCand  Lebedev. 


.jtft-.js.g'.-.jp^^.^jjl^^^^jl^ajt.  fetenb..  1897- 


n6o.    St.  Fetenb.,  1899. 


494  Geology  of  Armenia 

jAUKBAt  — w  (Warm  springi  ft  natural  faridlge  of  DiadBn).  Das 
AoslanS,  576,    Aiigrtniig,  1839. 

KlAiuKASHy  N.  New  finds  of  mammoth  remains  in  the  little 
CaucasQS.  pn  Russian].  Jonrn.  X  Congress  of  Russian  Naturalists, 
^ec.  No.  7,  94I-S.    XJew,  1897.    See  also  IiiosTRANznrr. 

RHANTKorr,  N.  Tremblement  de  terre  a  T^bris.  BuIL  phys.- 
math.  Aoid.  ScL,  xiii.  252-4,  &  zvi.  337-52 ;  ii6L  phjs.  &  dum.  iL 
sio-iy  &  iii.  348-68.    St.  Peteisb.,  1855,  1858. 

Rhodsko,  J.    Uascension  de  TArarat  en  1850.     BuIL  Soa  gfogr., 

sdr.  4f  i-  55'^S*    P<^>  '^S'* 

RocH,  C.H.E.    Reise  im  pontischen  GeUige  &e.    Weimar,  x846. 

„  GeoL    RMte  Ton  der  kankas.   Isdtfhus  &  von 

Armenien.    Berlin,  18^4. 

RoLfMLo,  B.    See  Abich. 

RoNSBiK,  A    Copper  ores   of  Zangesur.    [In   Russian].     Mat 

GeoL  Cauc,  ser.  s,  iv.  109-244,    TiBis,  1890* 

„  Mineral  .waters  of  Borjom  and  Abastuman.    [la 

Russian].     Mat.  GeoL  Cane,  ser.  2,  m  i-ix  Ac  1-104. 

Tiflis,  1893. 

„  (Palsogene  of  Borjom,   Akhalak    &   Abastuman^ 

Guide  des  excursions  du  VII  congrte  gtel.  intemat, 

xviii  &  zxvL    Sl  Petersb.,  1897. 

RoTSCHT,  T.    Reise  von  TYapetunt  nach  dem  Wan-See.    Petecm. 

Mitth.,  vL  68-77.    Gotha,  x86o. 

RuRKAKOv,  N.    Salt  industry  of  Transcaucasia.    [In   Russian]. 

Tifliski  Listok,  No.  10.    Tiilis,  1899. 

Lacroix,  a.    I.  Les  roches  a  leudte  de  Tr^bisonde ;  II.  Qndqnet 
roches  d'Arm^nie.    B.S.G.F.,  s^.  3,  zix.  732-48;  and  in  Comptes- 
rendus  Acad.  Sci.,  ex.  302.    Paris,  x89r. 
LsBBDKV,  N.I.     Uebersicht  d.  geoL  Sammlungen  d.  kaukasischen 

Museums.    Mitth.  kauk.  Mus.,  I.  iL    Tiflis,  1897. 
,•  Gold  im  Rauk.  8c  im  Flussthale  v.  Tscfacxodu 

Mat.  GeoL  Cauc,  s6r.  ^,  L    Tiflis,  1898. 
„  Infusorial  Earth  of  Rissatib.    ]ln  Russian].    Msi:L 

Geol.  Cauc.,  sfr.  4,  iii.  42.    Tiflis^  1895^ 
„  Die  Sammlungen  d.  kauk.  Museums,  in.  Geologie. 

(GeoL  CataL  Tiflis  Museum).    T^ts,  190  r. 

„  E]q>loration   gfoL    d'une   partie   du   district    de 

Bortchalinsk   (Somkh6tie).      [In   Russian,    with    a 

French  summary].    Mat.  GeoL  Cauc,  s£r.   3^    in. 

Tiflis,  1902. 

LnsovsKV,  B.    Transcaucasia  [a  chapter  on  the  minerals].     [In 

Russian].    Mem.  Cauc  Sec  Imp.  Russ.  Geogr.  Soc,  sx.    x-sjs* 


Bibliographyt  495 

St.  Petenb.,  1896. 
Lcxwoisoiif-LAsaxMO,  F.     La  poiphyrite  ancKsitique  a  amphibole  de 

D€w€bofpm!  en  Turquie.    Bull.   Soc  beige 
g&>L,  L  ZI6.    Brussels,  z888. 
,»  Fu^surit-Andesit   vom   kl.    Ararat    Sits- 

ungsber.    Naturf.    Ges.   Univ.    JmjeWy   zi. 
Protoc,  ciiL    Dorpat,  1897. 
Ldftos,  W.K.    Geology  of  portions  of  the  Turco-Persian  frontier. 
QJ.G.S.t  X.  464,  &  zi  a47-344.    London,  1854,  1855. 
Ltmch,  H.F.B«    Armenia :  travels  &  studies.    London,  1901. 

Maxsoft,  E.    E3q>^tion  sctentifique  au  Caucase ;  ascension  du 
Grand  Ararat    Bull  Soc.  roy.  beige  de  g^ogr.,  xii. 
577-99.    Brussels,  1888. 
M  Geophysik  d.  Goktscfaai-Sees.     Freiburg,  1896. 

Makxovxikov,  B.  a  deposit  of  thenardite  St  hydrothenardite  in 
Russia.  [In  Russian!  Joum.  Russ.  Pbys.  &  Chem.  Soc.,  six.  No. 
5,  S45-54.    St  PetersD.,  1889. 

Mavlrr,  p.  de.    Quelques  observations  g^ogenic|ttes  sur  le  bassin 
de  la  mer  noire,  le  Caucase,  &  TAnn^nie.    Rev.  umv.  d.  Mines,  zv, 
•40.    Paris,  1891. 
Mauhsell,  F.R.    Kurdistan.    Geogr.  Joum.,  iiL  8x-95.    London, 

1894. 
,9  Eastern   Turkey  in  Asia   &  Armenia.    Scott. 

Geogr.  Mag.,  zii.  295.    Edinbuigh,  1896. 
„  Central  Kurdistan.    Geogr.  Joum,  xviii.  xsi-44. 

London,  1901. 
Mat,  W.    Die   beigbauliche   Verhaltnisse   in  d.   TiirkeL    CEstr. 
2Seitscfar.  fiir  Berg-  &  Hiittenwesen.    Vienna,  May,  1896. 

Mim,  M.  The  basin  of  Lake  Gokduu  [In  Russian].  Joum. 
Russ.  Mines,  908-48.    St  Petersb.,  1891. 

MnnRiCATKR,  K.  Beitr.  z.  Kenntniss  d.  Mikro&una  d.  oberen 
Kreideschicfaten  v.  Transkaukasien.     Erlangen,  1896. 

Mojsisovics,  E.  voiL    Zur  Altersbestimmun^  der  Sedimentar-For- 
mationen  d.  Arazes-Enge  bet  Djo^a  Ih  Armemen.    Verb.  k.-k.  geol. 
Reichsanst.,  171.    '^enna,  1879. 
MaLLiR,  W.    Report  on  the  mines  of  the  Caucasus  &  Transcauc. 

in  1886.    [In  Russian].    Tifiis,  1887. 
„  Useful  minerals  &  mineral  waters  of  the  Caucasus. 

[In  Russian].    Mat   Geol.  Cauc,  ser.   s,   iii.   i-vi^ 
x-430.    Tiflis,  1889. 
„  &  M.  Dbnissow.    Useful  minerals  of  the  Caucasus. 

[In  Russian].    St  Petersb.,  190a 
MoLDKNHAxnBR,  F.    Analysis  of  the  waters  <tf  Bor  jom.    [In  Russian]. 
St  Petersb.,  1896. 


49^  Geology  of  Armenia 

MoHTBiTH,  W.    Toor  fhrou^  Asabidjan  te.    J.ILG.S.,  SL  1-58. 
Londo&i  1835. 
MoROAH.  J.  de.    Mitti<m  tdentifiqiie  aa  CaocAse.    Buis,  1889. 
MuLLBi-SaiOMiay  P.    Du  Cancaae  ao  golfe  peniqoe.    Pazis,  i89t« 

NAmcAHH,  E.  Die  GnindUnien  AnatbKms  &  Centnlaaieiis.  Hdt- 
ner^s  geogr.  Zeitadir.,  ii.  Jahig.,  1896,  7-25 ;  &  in  Report  of  Intemat 
G«ogr.  CoQgren,  661-670.    Londooi,  1895. 

NxsTEsoYSKry  N.  Copper  nunef  bf  Kedabdu  [In  RiiMan]. 
Jonm.  Rnss.  Mines,  125-48.    St  Petenb.,  1893. 

NiUMATR,  M.  8e  V.  Uhuo.  Ud>^  die  von  H.  Abicfa  im  Katik. 
getanundte  JmafoMiIien.  Deokidu  1/Kener  Akad.,  lisL  z-xsj. 
Vienna,  1891. 

NxwTOK,  R.B.  Marine  Tertiaiy  [Miocene]  moDasca  of  Lake  Unn^ 
8c  Note  on  a  Pabeos.  limestone  of  Lake  UnnL  Jonm.  Linn.  SoCf 
jczviL  43o*453*    London,  1900^ 

Omvaxnc,  P.  Die  Priabonaschichten  &  ibze  Fanna,  Pateonto* 
gnuphica,  zlviL  305-7.    Munidi,  1900. 

Oswald,  A.  Besteigui^^  d.  Aiant,  1897.  Jahib.  adiweis.  Alpen- 
dab,  zxnr.  157-83.    Berne,  x899-i9oa 

Oswald,  F.  Notes  on  Uie  geoksgjr  of  Central  Kmdistan.  Geogr. 
Jonm*,  xviiL  140-1.    London,  z9oi« 

Palobati,  W.G.    Vestiges  of  gbuial  action  in   N.E.   Anatolia. 
Nature,  yL  536.     London,  1879. 
Parrot,  F.    Retse  nach  dem  AiacaL    Berlin,  1834. 
pASTDKHOFr,^P.    Ascent  of  Ararat.    Qn  Russian].    MeuL  Cane 

Sec.  Imp.  Ross.  Gepgr,  Soc,  xvi.  43s.    Sc  Petehb.| 

1894. 
,,  Ascent  of  Alagos.    [In  Russian].    Mem.  Cane 

Se&  Imp.  Russ.  Geogr.  Soc.,  zz.  Mardi  15.    St. 
Petenb.,  1896. 
PiLiXAM,  A.     Petrograph.  Untersucfaungen  Empthr-Gesteine  ans  d. 
Kaukasus-Landem.     Beitr.  Pal.  QBstr.-Ung.,  iz.  83.    Vienna,  1894. 
FkacY,  Earl.     JEIighlands  of  Asiatic  Turkey.    London,  1901. 
Perkim», — .    (Visit   to   Enerum,   Ararat,   Are.).    Tmns.    Assoc. 
Amer.  Geol.  383.    Boston,  1840-a. 
PoHUG,  H.    Geol.  Untersttchungen  in  Persien.     Verb.  k.-k.  geoL 
Reichsanst,  281-4.  /Vienna,  1884.  _.. 
„  Weitere  Reise-Eigebnisse  in  Persien.     Verb.  Naturb.- 

Ver.  Rbcdnl.,  Siturngsber.,  173.    Bonn,  1884. 
„  I.    Ammoniten  aus  Mexico  8c  Persien;   II.   Eisen- 

arragonit  vom  Urmiahsee ;   III.  Watere  Ausgrabnngcn 
piioc.    Saiq;ethiere  su   Maragha.    Veriu   Ni2turb.*Ver. 


Bibliography  497 

Itheinl,  Siuungsber.,  ^a,  1419326.    Bonn,  188^. 
i>  Ueber  men  Hipparionen-Faonft  v.  Maragha  in  Nord- 

penien  &c»    Z.D.G.G.t  zxxvii.  1022-7.     Berlin^  1S85. 
I,  I.  Entstehung^eschichte  des  Urmiasees;    II.   Ueber 

das  nordper^.  Miocan;  III.   Ueber  den  Sevanga  See. 
Verb.  Naturb.-Ver  .Rheinl«y  Siuungsber.,  29,  49,  174-6. 
Bonn,  1886. 
»,  Ffiocene  of  Maragba.    Q.J.G.S.,  liL  179.     London, 

r886 ;  &  in  BulL  Soc.  Geol.  Ital.,  v.  411.    Rome,  1886. 
9,  UrmiabmanBOr  aus  Persien.    Z.D.G.G.y  sdv.  163-4. 

Berlin,  1895. 
PoLUNGTOM,  Viscount  'Journey  from  Erz-Rum  to  Aleppo  (1838). 
J.R.G.S.,  z*  445;54-    London,  1841, 

PoxpicKJi  J.F.  Palaont  &  stratigr.  Nodzen  aus  AnatoHen.  Z.D. 
G.C.,  xUx.  7x3-829.    Berlin,  1897. 

Raddk,  G.    Reisen  im  armen.  Hocfaland  (187 1).    Peterm.  Mitth. 
zviii.  367,  &  xix.  X74*83.    Gotba,  1872,  1873. 
„  Reise  an  d.  persiscb-mss.  Grenxe.    Tal]^(cb  &  seine 

Bewobner.    Ldpsig,  1875. 
„  Der  Bin-gol-dagb  &c.    Peterm.  Mittb.,  zziiL  41X-22. 

Gotba,  1877. 
„  Reise  im  russ.  ICarabagb*    Peterm.  Mittb.    Eigiina- 

ungsbeft  No.  loa    Gotba,  18^-90. 
„  Die  Sammlungen  d.  kalik.  Museum,  Bd.  iii.  Geologie 

von  Prof.  N.I.  Lebeder.     Tiflu,  1901. 
Rawumson,  ILC    Journey  from  Tabr|Hbrot^  Persian  Kurdistan. 
J.R.G.S.,  z«  X.     London,  1841. 

RsDUCH,  K.A.  Der  Jura  der  Umgebung  von  Alt-Acbtala.  Bdtr. 
Pal.  GEstr.-Ung.,  iz.  55.    Vienna,  1894.  „ 

Rbulbaux,  J.  Les  gisements  de  mangan^  de  la  Transcauc, 
Ann.  Soc  G^ol.  Btlge,  ix.  X37-42*    Bnissds,  X882. 

RiCKMXR-RiCKif  BR8,  W.  Annt  Zeitscbr.  deutscb.  &  osterreicb. 
Alpenver.,  xzvL  3x5-26.    Gras,  x895. 

RoDLRR,  A.  I.  Einige  Bemerkunj[en  xur  Geol.  Nord-Persiens ; 
II.  Zur  Entstebungsgescbicbte  d.  Urmia  Sees ;  III.  Eine  geoL  Rebe 
in  W.  Persien.  Sitzungsber.  k-k.  Akad.  Wiss.,  Matbem.-naturw.  CL, 
xcvii.  204,  tk  zcviiL  28.    Vienna,  1888,  x889. 

ScHUTT,  O.  Aus  dem  N.W.-licben  Persien.  Verb.  k.-k.  geol. 
Reichsanst,  30X-5.    Vienna,  x882. 

Sbidltxz,  N.  von.  Pastucbow's  Besteigung  des  Ahgos.  Globi^ 
bat.  85-90.    Brunswidc,  X896. 

SsMiATCHBNSiLv,  P.  Desmiue  crystals  from  Bofjom»  [In  Russian]. 
Jtev.  Nat,  335-344«    St  Petersb.,  x893. 


2K 


^ 


498  Geology  of  Armenia 

SMMUxaamKft  t.    Rock-nk   from   Kagiiman,      [In    Rnssiaii]. 

Piog:    Soc   Imp.   Nmt.   xxnL    io9-xi4.     St. 
PMenb.,  1896. 
»t  Lanmontite  in  canties  of  kva  in  the  Pambak 

TRiky.    [In  Rnariiui].    P10&  Soc  Imp.  Nat, 
SOL  IS*    St  Petenb.,  1899. 
SxyBnouim^  B.    Let  pierret  alomineiiaca  dcs  monta  Aiamta.    Mem. 
Acad.  ScL,  aer.  5,  iiL  109-14.    St.  Peterri).,  18x0. 

SusLv  J.    Joamqr  from  Tabrii  through  Koidtitan  Ac.    J.ILG.S., 
vitL  I.    London,  i^S^ 

SmoxB,  R.    Die  Schwankungen  d.  hocharmen.  Seen  adt  1800. 
Mitth.  gfpgr.  Gea.,  xxmL  95,  159^  390.    Yknna,  a888;  &  Petenn. 
Mitth.,  Na  3.    Gotha,  x888. 
SufONonroH,  S.    GooL  obaenrationa  on  the  NJE.  dope  of  the 

Tkialeti  Mta.    [In  Ruasian].    Mkt  GeoL   Cane. 
TUia,  1878. 
„  Exointon  It  Tknbooli ;  de  Sooram  It  KontaiL 

Guide  dea  escuraiona  du  VII  congiAa  gfc^  intenat, 
aocv.    St.  Petenh.^  1897. 
,^  &  A.  Sonouv.    Geology  of  the  Sharopan  district 

[In   Ruasianl     BCat.    GeoL    Cane 

Tiflis,  i88s. 
„  9,  Gedlogy  of  the  Cherimda  TaDcy. 

pn   Ruasian].     Mat.   GeoL    Cane 
Tiilis,  188S-6. 
,9  „  Geology  of  the  Tertiazy  basm  of 

Akhalsik.    [InRuaaianl    Mat.  GeoL 
Cauc.    Tlffia,  1886. 
,y  „  JMaepgene  of  the  Kntaia  dntiict. 

,  un   Ruasian].     Mat.    GeoL   Cauc 
"    Tiflia,  1886,  1887. 
SjteEKN,  H.    Beitr.  a.  GeoL  d.  Bogea  Smlan  im  N.  Persien. 
Trans.  Russ.  MineiaL  Soc,  ser.  a,  ladr.  36-66.    St«  Petexsb.,  1888. 

Smyth,  W.W.  GeoL  features  of  the  country  round  the  mines  of 
the  Taurus.    Q.J.G.S.,  L  330-40.    London,  1845. 

SOkOLOv,  N.  I.  Manganese  8c  sine  ores  of  the  Kutais  district ; 
IL  Infusorial  earth  of  Kisoatib  in  the  Akhalsik  district  [In  Russian]. 
Mem.  Imp.  Soc.  Russ.  Tedm.,  i-io.    Sl  Petersb.,  1897. 

SoKOKiN,  A.  I.  Eocene  of  Mtskhet ;  II.  Geology  of  the  districti 
of  Gori,  Dushet  &  Sharopan.  [In  Russian].  Mat.  GeoL  Cauc  L 
187.    Tifiis,  1879.    See  also  Simohovitch. 

STACHovdKi,  O.  Analyses  of  the  waten  oi  some  lakes  in  the  Tifiis 
8r  Kars  districts.  [In  Russian].  Westn.  RybopromysdiL,  150-5. 
1897. 


I 


Bibliography  499 

SmNiTZKr,  J.    Das  pontische  Gebiige.    Peterm.  Mitth.»  xzviii. 

339.     Gotha,  xSSa. 
y,  Catastrophe  in  Asia  Minor  near  the  frontier  of 

the  Kars  province  on  July  37,  1889.    [^^  Russian]. 
BulL  Geogr.  Soc.,  xxv.  430-3.     St,  Petersb.,  1889. 
Srsvns,  C«    Manganese-ore  industry  of  Transcaucasia.    ColUery 
Guazdian,  Izxv.  809.    LondoUi  1898. 

Strxcur,  W.    Das  obere  Zab-Ala  Gebiet  &c.    Peterm.  Mitth., 

357-63.    Gotha,  18^3. 
•9  Die  wahrscheinl.  altere  Form  d.  Wan-Sees.    Zeitsdi. 

Ges.  Erdk.)  iv.  549-53.    Berlin,  1869. 
STRXsuKNvsav,  see  Inostranzbff. 
SuisSy  E.    Das  AntliU  d.  Erde.    Leipzig^  1885,  &c. 

Taylor,  J.G.    Travels  in  Kurdistan  &c    J.KG.S.,  zxxv.  31-58. 

London,  1865. 
„  Journal  of  a  tour  in  Armenia  8cc    J.R.G.S.,  zzxviii. 

381-361.    London,  i868. 
ToBXRNiK,  G.    Gold  from  the  Batum  province.      [In  Russian]. 
Joum.  Russ.  Phys.  &  Chem.  Soc.,  zxvii.  384-5,  493-6.    Sl  Petersb., 

1895- 
TcHiHATCHXPFy  P.  de.    Asic  Mineure.    G^logie  &  Paltentologie. 

Paris,  1853-69. 

„  La  gtelogie  8e  Torographie  d'une  paitie  de 

TArm^e.    Comptes-rendus^  Acad.  ScL,  xlvii. 

44^-3,  515-7-     Pftria,  1858. 

TaxtXR,  C.    Description  de  TArm^nie  &c    Vol.  i.    Paris,  1843. 

Thost,  C.     Mikrosk.  Studien  ail  Gesteinen  d.  Karabagh-Gaus  8cc. 

A^bhandl.  Senckenberg.  Naturforsch.  Ges.,  xviii.  {ii)^  1-373.     Frank- 

fiart  a/M.,  1894. 

TiRTZB,  E.    Einige  Bildungen  d.  jiingeren  Epochen  im  N.  Persien. 

Jahib.  k.-k.  geoL  Reichsanst.,  zxzi.  67-130.    Vienna,  1881. 

TozER,  H.F.    Turkish  Armenia  8c  £.  Asia  Minor.     London,  x88x. 

TsxTLUKiDBB,  G.     Researches  in  the  districts  of  the  Alget  &  Khram 

rivers.    [In  Russian].    Mat  Geol.  Cauc.,  ser.  3,  i.  x-70.    Tiflis,  1887. 

Ubuo,  V,  see  Nkumatr. 

Valshtim,  J.  Bericht  uber  meine  Reise  nach  Tiflis  k  die  TeQnahme 
in  dem  Radde'schen  Expedition  in  den  Karabagh  Gau.  Beficht. 
Senckenbeig.  Naturforsch.  Ges.,  159-339.     Frankfurt  a/M.,  1891. 

VARBirrzov,  A.  Mines  of  Kedabek.  [In  Russian].  Joum.  Russ. 
Mines,  394-419.    St  Petersb.,  1894. 

ViQURSNXL,  A.  Sur  la  collection  des  roches  recueillies  en  Asie  par 
feu  Hommaire  de  Hell.    B.S.G.F.,  s6r.  3,  vii.  491.    Paris,  1850. 


500  Geology  of  Armenia 

VotKOBomxor.  ICajor.    (GeoL  obserratioiis  in  die  Talidi  nngitt) 

Joam.  Rnss.  Mines,  x8a8,  ziL;    18309  i,; 
183a.  IT,    St.  Petenb. 
9,  Official   account   of  the   earthquake  of 

Akhiny  in  1840.    Athensmn,  157,    Loo* 
don,  1841. 

Waomxe,  M.    Heise  nach  dem  Ararat  &c.    Stattgart,  1848. 

,y  Reiae  nach  Pecsien  &c    Leq»i^,  1859. 

Wnn,  W.  IVemUements  de  terre  en  Azm6ue.  La  Nature,  axx. 
Parii,  x888. 

WnTHonn,  K.A«  Jura  &  Kreide  aua  d.  N.W.-licfaen  Perrien. 
Sitaungsber.  k.-k«  Akad  Wiis.,  Mathem.-naturw.  CL,  xcvuL  (fX  756* 
Vienna,  x889. 

WiLBRAHAM,  Sb  R.  Traveb  in  the  IVanicancaaan  provinces  of 
Russia  ftc  London,  1839. 

WiLic,  T.    Palladium   in   auriferous   alluvittm  near  Batum.    [bk 
Russian].     Joum.  Russ.  Phys.  8t  Chem.  Soc,  zzviL  105-6,  505-7, 
St  Petersb.,  1893. 
WuNscH,  J.    Die  Flusslaufe  d.  Komur  8cc.    Mitth.  L-k.  gcoff. 
Ges.,  ser.  a,  zzvii.  aox*x9.    Vienna,  1884. 
„  Die  Quelle  d.  westL  Ttgrisarmes  &&    ICitth.  k.Jc 

geogr.  Ges.,  ser.  a,  xzviiL  x^ai.    Vienna,  1885. 

YoRKB,  V.  A  journey  in  the  Tdlejr  of  the  iqyper  Euphrates. 
Geogr.  Joum.,  viii.  317.    London,  1896. 


I 


INDEX 


Ab,  390-1 
Abiiiap  Rher,  164 
Abastaman,  394 
AddjivaSy  161-4,  i^Ti  4A6 
Adenk,  465 
Ag  Dagh,  357 
Aghbolakh,  375 
A^  D.,  380 
A^erov  i>.,  976 
Aghkentt  375 
Aghmexer  D.,  476 
Aghpur,  Khan,  367 
Agki  (Shatin)  D.|  J64-5,  4x2-3 
Agluikala  (Bingol  D.),  9x7-8,  993 
Aftlagan  D.,  7,  963,  375,  473 
AsTaUia,48o 
Ahkitd  V.  Ak-kul  D. 
Aiama,  9X 

Aiger  (Eighri)  Golg  171 
Ajaris  R.,  S,  499 
Aji  plain,  405 
„  R.,  460 
Ak  (Chekmah)  D.,  xx,  6x,  64,  68* 

70,  79,  X87,  98X-9 

Ak  R.,  4x7 
Akarak  R.,  364,  4x9 
Akdaban  (Kidd)  D.,  409 
Akera  R.,  3x9-3 
Akh  D.,  349 
Akha,  376 


Akhalkalaki  (W.  of  Abul  D.),  379 
,f  (in  the  vallqr  of  the 

Postkoff  RiX  339-90 
Akhalak  (Akhaltsykh)  baain,  8, 

379f  3^-9S»  4«6.9 
Akhbaba  D.,  7,  93  x 
Akhbulakh,  379 
Akhgachi  370 
Akhkent,  379 
Akhl^,  98,  X45,  XS9,  X56,  x66, 

446 
Akhoti,  397 
AkhtaD.,  386 
-AUitala,  315-90 
Akhtamar  ifSand,  448 
Akhury  (Aiguri),  4S2 
Akhveran  (of  the  (Siorokb  ba9m), 

«3«f  a34-5t  340 
Akhveran  (N.  of  Khinu),  9*89 


(in  Darachichdc),  964 
Akhviran  (near  Sipan),  95,  179-3, 

446 
AkRdpriD.»449 
Ak-ku)  (Ahkiul)  D.,  xxx,  4^5 
Akstala  K,  3451  349»  S^'*  374f 

39« 
Akiuraf  ^Axaial),  447 

Aktahtt-dereh,  393 

Alagaya  D.,  963,  33X 

360 


J 


5oa 


Geology  of 


Ak8$s  D.,  S,  6,  lo,  66,  47 <»  4^3  \ 
,,     (&  of  the  Bu«ol  D.X  X99 

Akja,  44 

Alaahkectt  S»  xa 

AiAiAKy  34S-50 

Albon  (Elbon)  Mti.,  4f  ^^4$  S97» 
306,309 

Alexftndropol  (Gumri),  375,  473-4 

A]^  lU,  7>  383*  397 
Ab,  35a 

Ali  D.  (near  Aighana),  384*  4»^ 

,,     (near  Bitlis),  a  78 
All ja^krak,  463 
Ali  Knlek,  75 
Alikolikent,  344-5f  3So-Xf  355^, 

3^t  3^* 
Ali  Meaer  Vezerli,  a43-4 

Ali  Mur  (Karaba)|   59-61,  aa4, 

a8a 

Alinja  R.,  408 

Al  Khes,  8x-a,  45a 

Almaly,  v.  Elmali 

Altun,  M2J 

Altiin-bulak,  a66 

Amamly,  375 

Anabat,  407 

Andamidi,  407 

An|;ag,  870 

Am,  474 

Antiuunis,  4f  5»  7f  «i  »fi7f  «73f 

a8i,  285 
Anzek,  v.  Entakh 
Anzub,  449 
Apaud  (Aipaout).  420 
Aptian,  346-7 
Aquitanian,  427-8;  433-3 
Axabkev  D.,  405 
Arabkir,  272-3,  457 
Ararat,  5,  10,  267*  284,  286,  364, 

43*1  473»  48a 
„      ,  Littlci  289,  296,  298 
Aras  Oghlu,  474 

Araxes   depresaion,    xa,    289-90, 
«98f  3S7»  4"f  43«f  4631  47a 


Axazes  R.,  4-6,  54-7,  993,  996, 
227,  286-9,  291-3,  a9S-6, 300-5, 

337-8»  359.   3^9.   37«t   405-7t 
43a-3i    4S«-3t   4S9-60p    469-3, 

47«.  47St  477 
Archag,  Lak^  448-9 

Ardanach,  8,  353,  365 

ft         R*f  S35t  s^Sf  34^  353f 
380-1 

Axdasa,  25 

Ardebil,  5 

Ardevi  D.,  374 

Arenjik,  184 

Argachi,  432-3 

Arghana,  384,  421 

»      R-t  384 

Argovan,  476 
Arikhe,  392-3 
Arjevaa  D^  7,  377 

Arjish,447 
Arkhava,  8 

Ariodha,    cpoclia   of  moontaia- 

folding,  9,  xxo,  26X,  267, 

286-7,   300,   383,   435f 

458-9,  469 

former  lakea  of,  441 

469 
linesofftactaxein.  xo, 

xxo,  X5X,  197,  469 

main     mountain-foldi 

of,  4-? 

position  in  the  plateau- 
belt  of  Asia,  X,  469 
Armkniah  Bordkb  Rahgxs,  B., 

V.  Gokcha  &  Tafish  Ranges 
Armsnian  Border  Rangbs,  N., 

V.  Thxialetic  &  Pondc  Raxiges 
Armxnian  Bordxr  Rangb,  &• 

▼.  Tanius 
Armsnian  ULXXSt  xx~6,  469 

„        PLAOfS,  origin  of,  ix, 
X2,  44.  46-71 62,  xxo,  X48,  175, 

X97f  «33  ^    , 

Armenian  FLATXAxr,  folds  of,  9, 


If 


ft 


ff 


t» 


tndex 


503 


sS6 

AraxmAif  WLLTEkV,  scrnctdfe  of, 

II,  3o6f  3Mt  375f  S7Sf  S^tt  469 
Armihum  riybbs,  cfaandcriftici 

rff  S4f  4«f  ««« 
Armbmun  yowAirosSi  pootum  olf 

»«f  4Tt  «S» 
Amot  D.,  t7tf 
Aipa,37o 

M     R.  (of  Alezandropol),  473*4 

M     R.  (of  Daxalagos),  264,  s86» 

990*1,  §94-5,  370,  411 
Afpaout,  ▼•  Apaiid 
Anakent,  964 
Ammik  (Avioonik),  478 
Artyia,  8,  aSSt  381 
Anian  D.,  8,  lo,  498 
Asam  R.,  #99 
Aahaga  Bunin,  t.  Bumii,  Ashaga 

„      Danaik,  ▼•  Danaikt  Aahaga 
AaUala,  4a,  464 
AaiA  Mdtok,  devfttioii  of  north 

ooaal^  14 
Asuam  vlate au-bbi^i  b>4 
Aapmja,  499 
Anvreti,  396 
AalBiB  R.,  165 

Aatttji,  449 
Avlabar,  4B0 

ATioonik,  v.  Axannik 

Azaraf,  ▼•  Aksuraf 

Amaburt,  337,  347i  857-8 

Baarem  hilbjQebd  Tor),  411 

Babduchan  D.,  315 

Baghir  D.,  5,  349 

Ba&urt,  56,  J37,  340,  415^,  486 

BaiBdir«  87- 

BajociAir,  Lower,  310*5 

ft  .    f  pPP«f  8«S-«» 
Bakirem,  467 

B'^''^^'^  iSt  *40 
Bani,  88 

Bar,  464 


Baidak,  V.  Taadi 

Bardiia  R«,  464 

BAaBBMiAM,  343.5 

Baiaar  (Kaq^ar)  R.,  373 

Bwtaa,337 

Babtovxar,  390 

Baiarkent,  314 

Baahkcnt,  la,  aSa,  451^  466-7 

„         D.,  444 
Baah  Nnrashin,  v.  Norashin,  Bash 
Baahtttbel  R.,  361,  398 
Baakan,  aoo 
Baatok,  aaa-4 
Basyrgan,  4ii>444 

BATHONIAMy  318,  321^ 

Batman  R.,  458 
Bayaad,  41a,  44^,  444-5»  4<o 
BajBs  Tnala,  v.  Tuzla,  Bayas 
Baar.R.,^13,  397,  399 
B^chinakh  (Kannalinovka)  paaa  ft 

Rf  3^9.  409 
Begi],386 

Belen,  973 

Bdey,  ▼.  Bidava  or  Bdev 

Bell  D.,  986 

Beliaor-ideresai,  464 

Benara,  499 

Bendimahi  R.,  116 

Beraatik,  457 

Beigoshet  R.,  344-5*  355-^t  3^, 

37* 
Bbbbzas  SBRiBa,  335-6 

Beah  KUiaaa,  98 

Bezobdal  D.,  6,  7 

Bielava  (Belev,  Bidav,  of  Darala- 

gDs),337t  3581370,411 
Bidogore,  378 
BidyUdiadi,  7,  351,  363,  376-7, 

383 
Bilejan  D.,  175-6, 178,  i8o-i,  449 

Bilev  (Belev,  of  W.  Kaxabagh), 

407-8 

D.,  II,  sSt  5«>  ^«^3»  ^5» 


103-95,  477-8 


5<H 


Geology  of  Armenia 


BiQCol  Dm  diffi  of,  58^  i^-7f 
199.901.  M^  tas 
.,         cmidfonn    inicnptioii 

of,  914*5 
,y         foffinar  gjUden  of,  916- 

90,  481 
„         N.  (cirqiie)  region,  9x5- 

90,994 
„         N.W.  pUteau,  990-7 
„         &  region,  903-10,  994- 

5         .        , 
,9         suinimt-region,    910-^ 

«a4 

„         volcanic  gaccesnon  of, 

994.5 
BingoDuda,  9x4 
Bingol  Sii  (of  Chaxbahiir),   199, 

n         (of  Khinig),  69-4,  68, 
70,  74,  180-1,  919^  994, 

4S»»  467 
Biolis-Kebi,  459 
Bir,  568 
Bishi,  83 

Bishiweb,  350,  354 
Bitlis,  977-C  486 

„     R.,  977,  467 
Bitlu,  183 
Bogvi,  396 
Bohtan  D.,  494 

Bojan  R.,  3x1 
Bolnis  iL,  363 
BoNNXT,  Prof,  T.G.,  descnptiong 

of  Armenian  rocks,  339,  447-8, 

485.6 
Borchal  IL,  ▼.  Pambak  R. 
Borjom,  8,  383 
B02  Tepeh,  x  7 
Bulama   (Kesan,    Shailu),   Lake, 

175.7 
feulanik,  173*7,  '80-x 

BUKDIGALIAN,  439 

Bunin,  Ashaga,  990.x,  994 


BnmskOi  83 

Cauxhtux^  319,  399-5 
Caneqoa,^.  gh^na^ 
CABBomfiBODS,  Lower,  999-8 
„  ,  Upper,  998-9 

Caucasus,  10,  949,  961,  300,  306, 

309.  3«7.  33 1»  334f  33^ 

7»  346,  343,  351*  355« 

373i  33a,  339f  4i9f  453* 

9 
„       ,  Litde,  ▼•  Gokdia  Ban* 

ges 

CwoiciifiAN,  350-4,  371 

Ch (guttnral),  ▼.  Kli 

Cha  D.,  X96,  450 

Cbaganly  D.,  357 

Chagbelik,  sox 

Chauimi  R^  333 

Chak  Suyu,  478 

Cbaldir,  Lake,  7,  xx6,  469 

Chalkhan  D.,  370 

Cl^uiaia,  v.  Slhanaga 

Chanakhcfai,  445 

Changeli,  ▼.  Karapet,  Surp 

Cbarbahnr,  X99,  198 

Chashlu  Gedik,  364 

Chatiin.kdpri,  93 

ChanndurR.,  398 

Chaush,  465 

Cbehrik  R.,  v.  Zaba  R. 

Cbekmab  D.,  t.  Ak  D« 

Cbelifteb  D.,  977 

Cbellek,  v.  GneQek 

Cbembek  D.,  973 

Cbemisbgezdk:,  457 

Cberimela  R.,  961,  390-9,  343-9 

Cbermuk,  384 

Cbevirmeb,  68 

Cbiltlik  ^on  Arszes),  463 

„      (on  Kum  K.),  389 

Cbikhnis,  453,  465 

Cbilgyas  D.,  331 

Cbimen  D.,  478 


I 

/ 


Index 


S05 


Chishevaa,  388 
Chishv^ran,  465 
XHdt  Rm  ts 
Choban-kopri,  475 
Chorak  Khan,  948,  4x8 
'  Chorokh  IL,  7,  8,  36,  936-7,  965, 

338-41  f  SSSf  3^S»  a*!.  4^4*  4«3 
Chustop  D.,  34 < 

COBALUAN»  v.  SlQUANUJr 

CsETACBOUSy  LoweTi  33W7 
ft         f  Upper,  348-8X 

Dagdagaa,  371 
Dagras  (Dagtaz),  330,  379 
Daheha,  987 
Daly  D.,  314,  361 
Daljkh-taDa,  4x0 
Dammty  D.,  5 
Dampal,  38X 
Dam^ah,  369,  376,  379 
Daniel  Begamish,  996 
Daniel,  Suip,  x8o 
Dansik,  Ashaga,  990,  994*5 

„     ,  Omaga,  370 

„    ,  Yukhary,  998 
Dara,  499-3 


Darachichak  D.^  ^  964 
Daradys  (Klutf«ui' 
Darakol,  970 


)  D«,  360 


Daialagoi^  5,  964,  967,  984,  986, 
989^5,  998,  337,  347,  356^, 

S^4f  369-7X1  409-"i  43<^«t44S 
Darosham,  309,  404,  433.4 

^     »t         Evaghlii,»97 

Daroshinsky,  309 

Daiy  D.,  495-6,  460 

Dashalti  R,,  313 

Dashkent,  313,  36X 

Dashkessan,  394,  369 

Dasht-i-Kiri,  48X 

Davachukhur,  398-9 

Davalu,  990 

Debeda  R.,  374,  480 

Dedeviren,  69 


Degma  Danga,  986 
Deir-i-Yakub  hills,  499 
Deinnendereh,  19 
Ddejan  (Delishan),  347, 349,  36J1, 

381.  483 
Demian,  94,  173,  479,  478 

Demirkala  (Bingol  D.),  913,  9x7*8 

Derek,  499 

Deiik,  190 

Derkashen,  449 

Denim,  478 

DsvoMiAN,    general    distzibation, 

983-7 

„       ,    Middle,  287-90 

„       ,   Upper,  990-9 
Dewil  D.,  46a 
Diadin  (near  Bayadd),  444-5 

„     (S.  of  Bin^oIX   900 
Dibeneh  R.,  v«  Zibeneh  R. 
Dibija,  456 
Diduwe  D.,  8 
Digom,  7 
Dilman,  46  x 

Dj ,v:j 

Dodan,  900 

Dokhuspunar  D.,  8,  380 
Dort-kiUssa,  966 

Dsch— — ,  ▼.  J 

Dsynlerly  D.,  5,  990,  994 
Duchu  R,,  3x4,  361,  383,  398 
Dukhan,  467 
Dtuda,  467 
Dudak,  46  X 
Dtuyurt,  55 

Egin,  5»  457f  465-6 
Egri  R-,  55f  a«7 
Eighri  (Ajger)  Gdl,  171 
Elbura  v.  Albuxz  Mts. 
Elenendorf,  373 
Elisabethinskoi,  396 
Elisabetpol  Mts.,  311, 373 
Elmali  (Almaly,  in  Ae  Fxat  ba^), 

365*  41 5»  456 


5o8 


Geology  of  Armenia 


Jda  D.,  979-81, 483 

Jenapa,  134 

Jcniiik,47S 
Jevizbky  ao»  957 

Jezirehy  9,  499 
Jeziroki  178 
Jiatura,  969 
ja  Gdl,  165.7 
Jinis  R.»  43 
Jtidi  D.9  9,  385-6 

Julfa,  984,  30O-S.  337,  433r4f  459* 
60 

Jnranlyy  4x3 

Jurassic,  Lower,  306-10 

,,      ,  Middle,  3io-9a 

„      •  Upper,  399.34 
Kabagh  D.,  419 
Kabagtappa,  394-5 
Kachevan,  r.  Varxahan 
Kageba  D.,  8 
Kagizman,  4x9.3,  463 
Kajurin,  Khan,  367 
Rata  R.,  4x3 
Rala'at  D.,  491 
Rala'at  Marah,  499 
Raladara,  1x3 

Kalajik,  3X 

Kalarach  v.  Kelarik 

R*H,  83,  45a 

Rameaka  R.,  964 

Ramshut  D.,  375 

Ramsor,  v.  RomenUor 

Ramiishlii  D.,  aoo-x 

Ramzakendi  (Hamzakend),  386* 

Randsassar  (Yank),  314 

Rangly  D.,  416 

Ranjean,  80,  453 

Rantar  D.,  904 

Ranzag,  447 

Rapujik  D.,  5,  965 

Rara  All,  94 

Rarabagh,  5,  6,  306,  356 

n       f  E-i  6,  319-4,  397,  399- 1 


^  33.  337»  3So»  3«  V3«9»  371-3 
Rarabagh,  N.,  963,  350,  361 

„       ,  Central,  6,   xi,  399^ 

3^f  407 
„       ,  W.  &  &,  5,  943t  «<4t 

3*7-9.  333.  344-S.  3So-«f  355. 

359.  ?68-9.  39«»  407,  409 
Rarabairkh,  43,  454 

Rarabula,  489 

Rarabulakh  R,,  376 

Rarachoban,  70 

Slaradagh,  4,  6 

Karadagly,  337 

Rara  Dereh,  947 

Raragash,  370 

Raragaya  D.,  3x5 

Raraghor,  X79 

Raiagol  Toda,  v,  Tuda,  RaxagH 

Raiaja  D.,  9.x  x,  499 

Rarajdiennem,  364 

Rarajurcn,  387 

Raratvixen,  8x,  459 

Rarakala,  9x4.5,  9x7,  9x9 

Rarakaya  D.,  5,  61-9,  989 

„        Sanislo,  7 
Rarakent,  379 
KftTakili^ffa,  449*3 
Rarangu  R,  308,  471 
Rara  Ogti,  x8x 
Rarapet,  Surb  (Changtli,  Tdianla- 

kihssa),  X96.7,  495,  450 
Rara  Seid,  X97.8,  994 
Rarasu  (Tanms),  977 
Rara  Su  (Axerbei janX  S 
Raratuban,  394,  498 
Karayazi  steppe,  79,  80 
Rarchkal  D.,  8,  4x9.90,  499 
Rardarich,  49,  454-6,  4^.  477 
Raigabazar,  43,  47S 
Raigapunar  D.,  8 
Rargar  R,  v.  Barsar  R. 
KarmalinoTka,  v.  BecfainaUi  ' 
Rarmenik,  464 
Rarmirvank,  984, 999,  »94*5^  i9h 


Index 


509 


Karmach,  145-Ot  151 
Kaminsht  vA  xS^i  ^54 
Kan  &  Kan  IL»  7,  474 

Kartevin  D.,  88^9 

Kasanyaila  D.;  358,  369-70,  411 

Kashlrinakf  265 

Kaspi^  480 

Katzkhi,  307-8 

Kayaldshlar,  4x9 

Kaxakh,  375 

Kazbd  GtdUL.  73,  75,  45  < 

Kazikly  D.»  948-59 

Keban  Maden,  971-9,  38X1  490-1, 

4S7 
BLechal  D.,  6,  409 

Keinik  (Koinuk),  408 

Kelarik  (Kalarach),  414,  454 

Kdcdi,  17s 

„     D.,  147,  178 
Kdkid  R.,  7,  8,  4x6 

„     — Chorokh  watenhefl,  4x5 
Kolus  R.,  379 
Kcmakh,  4x5,  465-6 
„       Boghaz,  349 
Kemler,  ▼•  Komentzor 
Kendalen  (Kjunddjaxi)  IL,  379 
Kensun,  X7^  4x7 
Kerkur  D.,  xx5.7 
Kcnik  R.,  8x-9,  459 
Kend,  939,  4x6 
Kesan,  88 

„    ,  Lake,  v.  Bnlama 
Keser  R.,  424,  468 
Keshish  D.,  5,  349,  483 
Koshishkiant,  41  x 
Keshishkoi,  947 
Ketam,  359 
KctukD.,373 

Kenmur  (Komur),  N.  of  Kemakh, 

388.9,  466 
„       D.,  S.  of  Bit&s, 

385 
Khaa,  S.W.  of  Khaiput, 


»i 


M 


384 

Khach  D.,  5 
Khachik,  358,  4x1 
Khachin  R.,  314,  330-1 
Khachkar  D.,  266 
Khachmach,  379 
Khadrak,  35 
KbaGl  Divana,  475 
KhamBhujur,  74,  451 
Khamur  D.,  61,  185-7,  478 
Khan  — ^,  ▼.  the  second  name 
Khanaga  (Canequa,  Chanaia),  393- 

4 
Khanik,  ^6,  226 

Kharaguh,  378 

Khard&or,  970 

Kharput,  420 

Khanhut  R.,  25-32,  342-3,  4x7-8 

Kharzan  D.,  v.   Daradys  D.  or 

Itushem  D. 

Khaskoiy  270 

Khedonun,  56,  224,  997,  477 

Kherbesor,  57,  993-4 

Khero  (OkheroX  393 
Kheveh,  83 
Khidistav,  396 
Khingemek  R.,  966 
Khinis,  65-7,  994 

„      plain,   X9,  69,  64,  68-9, 

477 
Khizan,  977 

Khodudjur,  966 

Khodulia,  464 

Khoi,  984,  44x-'9  46x-9 

KhojaU  R.,  330 

Khopa,8 

Khorasan,  474-S 

Khoros,  977 

Khofidrab,  989-90,  996 

Khoshabpunar  D.,  j^  >3s-39  339- 

41 
Khotanan,  333 

Rhozapin,  Lak^  7,  469,  479 

Khozat,  478 


5IO 


Geology  of  Armenia 


KhnMienif  7 

„      (Kada)  R.,  t6j,  351^  J7« 
Khreiti  dome,  307 
KhriseU,  310 
KiAki  D..  333 
Kieti  D.,  375 

KigwA,  473^ 

KilisM-dereh  (Medjingett),  64 

Kiiita,359 

KiMMiKiDcuy,  326-33 

Kippgol,  43t,  48« 

Kirchu,  457 

Kiregan»  213 

Kirklar  Gdl,  170 

Kirkat,  v.  Kukurtlu 

Kirrind,  384 

Kishishkenty  361,  373 

Kisil  iU,  ▼.  Koiur  R. 

Kiskin,  966 

Kissatipy  480 

Kitowa  D.»  245-6 

Kivrag,  459 

Kiyau  IL,  367 

Rizil  Bogas  D.,  6 

Kizilkaya  D.,  432 

Kiziltash  D.,  v.  Kiziljreri  D. 

Kizil  Usen  R.,  462 

Kizilyeri  (Kiziltash)  D.,  5,  433 

Kiz  Kalassiy  330 

Kizvag,  X4i-3f  151 

Kizzuljah  D.,  v.  Kus  Ije  D. 

Kjundeljan  R.,  v.  Kendaien  R. 

Kobliao  R,  429 

Koinuk,  v.  Keinik 

Koli  Baba  D..  18$.  188-9 

Komentzor    (Kamsor^    Kemler), 

453.  465 
Kompor,  353 

Komur,  v.  Keumur 

Konakti,  424 

Rop  D.,  7.  39,  40,  232,  235j  338. 

41,  4^6 

„    Khan,  M.^  38,  339 


Kop  R.,  37-S 

KarR.,  180 

Kocbo-oH,  t6a 

Kofchevan,  359 

Kor  0|^,  387 

Koieh,  416 

Kotur  (in  Fxat  basin),  433 
tt  on  Azefbdjan),  387 
„     (KisQ)  R.,  302^  386,  46X 

Koyundagbi  uland,  436-^8 

Rbjkssknbbrg  aoRizoKy  395, 405- 

Krimizi  Ttiala,  v.  Tiuda,  Krimiat 

Ktutzi^and,  447 

Kubadu,  373 

Kttdiasheny  464 

KuU  D.,  V.  Akdaban  D. 

Rukiirtlu  (Kxri^ut),  388 

Kulidaia,  409 

Kulpif  Sf  364.  433»  462.3.  473 
Kumiskoey  Lake,  480 
Rupchivaot  460 

Kur  depression,  8,  10,  330*  373 
„    R.,  7,  8,  262,  266,  380,  396, 

4*91  459i  479-80 
„  — Araxes  watershed,  313 
„  — Chorokh  watershed,  266 

Kurbittu  R.,  44a 

Kurdkoi  (Elpkisdkoi),  465 

Ruriki  D.,  279,  366-7 

Kuro  R^,  425 

Rurreh  Hassan  D.,  58,  282 

Kuit  R.,  236 

Rum  R.,  388-9,  4X5r  466  | 

Rnseh  D.,  10,  87  / 

Ruzulje  (Kizziiljah)  D.,  449 

Kvirila  R.,  261-2,  310-ix,  320-9, 

383.  430»  458-9 
Ryabyrli  D.,  409 

Ksia^R.,  V.  Rhxam  R. 

Lala  D.,  74,  187,  4^t 
Larkhana  D.,  236 
Lasheh,  335 


5^1 


Z^tCTf  174*5 

Iialvar  D.,  «,  7, 314*10,  j6j,  375 
Lias,  306*10 

ft    — Xhindm  vaO^,  004 
LidgaS;  935,  341 
LiGURUV,  4971  429 
lijeh,  J71 
Lao^7 

tUi449 
Lori,47^ 

Losavonch,  Smpy  (Vaak),  3S8 

LuTKTiAM,  390 

• 

Machakbelisrttkhali  (R.^  4x9-90 
Machka-dereh,  90-i»  418 
Madavants,  156 
Maden  Khan,  36,  340 

19     R.  (Pontic  Raoge),  at 

„      R.  (Taurus),  S77 
Ma^havaUy  v.  Madravank 
Madis,  980 

Madrak,  ss-jf  994,  997*8,  453 
Madrayank  f  Madghavank),  447 
Magarah  valley,  933 
Mahmiit,  Sheikh,  409 
><«»^uisor,  930,  477 
Maku,  443-4 

„      Mts.,  996 

„     R,,  984,  449 
Malatia,  9 
Malishkent  (Malishke,  Ma^ka), 

410,  431,  44S 
Mamakhaton,  387, 4x3*4,  465 
Maman,  469 
Ifamkodi,  8    ^ 

M^nglWf  7»  39* 
Maragha,  399,  470-x 

Maiand,  984,  999,  460 

Mardin,  9,  350,  367,  385,  4««-3 

Marelisi,  969,  344,  459 

Martiros,  410 

Masca  UaDobeni,.  976 

Masra,  409 


ICadiaTeia  R.,  376 
Mass,  T.  Moss 

Matrtikan  D.,  S7j 

Matsalakhet,  353,  380 ' 

Mauhsbll,  Colonel  F.R.,  geo- 
logical observations  by,  976-7, 
384-6,  465,  468,  481,  483 

Maurek,  474 

Medjingert,  v.  Kiltssa-dereb 
Mehmedik  R.,  449 ' 
Meirinuun  Dmh,  959-6,  4x9 
MelaSkert,  99-4 

w  .V:  ^    .?***"•  94.  17s.  47» 
Meukuitnf,  478-9 

Me}t>ert  D.,  5,  349 

Madosh,  9<5,  4x9 

Mexge  Khel,  79 

Meigemir  D.,  v.  Sharian  D. 

Merikent,  371 

Mcqan  Bc^^ias,  388 

w"  .^^-^  St  34«,  3M 
Memud  R.,  449 

Merov  D.,  984 

Meschic  Mts.,  v<  Zinila  dome 

Meseidx  Khan,  943 

„      f  Upper,  943 

Miaiarkin,  481 

Mianeh,  469 

Migrif  359 

Mihiab  D.,  384,  49X 

Mikhailov,  383 
Mileyian,  467 
Miocm,  Lower,  435*5* 

J,      ,  Upper,  435i  45«-** 
Mtnam  D.,  5,  387-8 
Modoti^pa,  Lake,  7,  376 
M<ddie,  499 
Molita,  344 
Mofkh,  149 
Moss  (Mass),  4x1 
Mukhet,  8,  379,  459f  4*0 
Mttkhoxtioiian  D.,  6 
Mukhian,  480 
Mukns,  976 


5" 


Geology  of  Armenia 


liund  Khan,  30,  343 

„      R.,  85-94, 177,  x8o-a,  184- 
S,  188-95,  444i  449-SOi  47S 
li  R.,  311 

Murow  D.,  zo,  $^0 

Mush  plain,  za,  117,  148-50,  267, 
969 

MiisUu  (Tttzla)  R.,  4x3,  453 

Muakhana  R.,  964 

MiistttT-deressi,  453 

Mtbor  D.,  5,  7,  34a 

Nagashir  D.,  405 

Nakhicfaevan,  300,  459 
„  R.,  408 

Nakhichevanik,  361,  379-3 

Nariman  R.,  464,  479 

Naaik,  Lak&,  177.9,  449»  469 

Negram,  301-9,  305 

Nbocomian^  Lower,  335-6 
„        ,  Middle,  336-43 
„        ,  Upper,  343-5 

Nepiskaro  D.,  8 

Norjiki,  970 

Nermit  D.,  386 

Nimnid  D.,  xo-19,  1x3-55,  '79 
„        ,  chronology  of,  150-5 
„        ,  crater  of,  1 13-4,  15 1-5 
„        ,  external  features  of, 

114.8,  130-55.  179 
„        (near  Samsat),  368 
Nisham-kuh,  v.  Itushem  D. 
Norbert,  381 
Noreg,  970 
Norashem,  999-3,  995 
Norshen,  109,  149,  154,  270 

NUMlfULinC  LlMBSTONK,  389-495 

Nurashln,  445 

„       ,  Bash,  994.5 

Oannes,  Surp,  349 

Okhchi  R.,  965,  397-8 

Okhero,  ▼.  Khero 

Oktaberd,  411,  430- z  ( 


Okys  D„  376 
Olek  (Ulek),  978 
Oliqocsns,  390, 396,  4^6-34,  439 
„        ,  Lower,  497,  499 
„        ,  Upper,  497-8, 430^ 
OW,  463 

„    depression,  7,  456,  479 

„   R.,  381,  463-4 
Omaga  Dansik,  v.  Dansik,  Onu^ 
Ora,  Sheikh,  103-4,  974 
Oramar  D.,  98^ 
Orchil  D.,  478 
Ordubad^  359,  369, 406-7 
Oshora  D.,  8 

„      R.,  499 
Osluk  EJian,  35,  349-3,  4x6 

n     R-.  35 
Ostuk,  940 

Otiuk  D.,  7 

OXFORDIAN,  399,  395-6 


,68 
Pais,  370 
Pakhund,  978 

Palandoken,  5,  xz,  46-5x,  998,36$ 
Palu,  9,  384 
Pambak  D.,  6,  964,  374-5,  i^h 

397-«      , 
„       (Bordial)  R.  369,  374-S> 

398 
Panaskert  R,  380 

Paiaga,  407 

Pardi  (Bardak),  354,  4S3 

Parly  D.,  v.  PerH  D. 

PartskhnaU,  459 

Pashalii  R.,  410 

Pasha  Punar  plateaa,  453,  4^ 

Pasin,  5,  19,  4x3,  453,  475 

Patnots  R.,  99 

Pdo,  975.6 

Pennek  R.,  464 

Pcpis,  375 

Peril  (Parly)  D.,  364,  4x9 

PuoaAN,  Upper,  309-5 


k 


Index 


513 


PftUCO-TlUABSICy  300*5 

Pemaut  R.9  4^*3 
Pcnar,  417 
Pertekrek,  166,  464 
PimmafCTin,  89 
Pimakapon,  41 ,  464,  486 
Platanai  4S0 
Plustocenb,  469^  47x*8x 

PuocBNS,  469-7x1  473 

Plur,  4x6 

Podkasapork,  33  x 

Poni,  s6t)  344,  34«-9f  SS«f  S^Si 

379 
P011TIAN9  458 

Pontic  Range,  8,  X9»i5,  140-57, 

2^6,  306^^41 7-<o,  499 

Postkoff  (Poskhor,  Poskbor)  IL, 

389»-4a7 
Poyas  IL,  370 

PlJC-DxVOIflAM,  t6t'S9 

Priabona  borxion,  39-),  395,  4x1 
Pullarik,  415 
Putkiehi  Khana,  9j6 
Pyrjamal,  399,  350 
Pynrily,  474 

Rahraiidatn,  1x7 

Bashan,  75,  451 

Redwan,  385 

Itddu,88 

ReaEli,  JS4 

Rion  basm,  383,  416,  4301  458 

Rowandns,  xo 

Sadagjm-Sfrtjr  pan,  415 
Siu^alaUo  D.,  8 
Sanend,  3^9,  46a,  47o*x 
Sttgrama  D.,  6 
SMU^xa,459 
SaTmas,  984 

&lvar^  D.»  St  3«9i  409 

Saljrardak,  4x7 
Samjd-iobat  R.,  499 
SmMU,  9,  368^  458 


Samiim,  x6 
StaoMlio  D.|  377 
Saidarak,  9M,  288 
Sank,  975 

SARMATiAir,  435,  458-68 
SazprR.,  964 

Sax7da9h,  457 
Saxykaya,  453 
Sa9an,  970 
Satchkheri,  378 
Satld,  499 
Savalan,  4-5 
Scortey  (Seny),  481 
Sonnn,  393 
Seifi  Rud,  308 
Segirik,386 

Sor,  440*x 

„    (Scyr)  D.,  440-x,  469 
Sciva,  333 
Sdvan,  xxa,  386 
Sdin  R.,  398 
Semenovka  pa99,  483 
Semi  R.,  398 
Sbkoxiah,  368«8i 
Sbquamam  (Coraijlian),  396 
Serai,  449 
Serokh  D.,  900 
Seny,  r.  Scorsqr 
Scrt,  9,  385,  467-S 
Sevakiar  ^Sevaker),  333 
Sevanga,  Lake,  v.  Gokcha 
Seyr  D.,  ▼•  Sdr  D. 
Shagadi  D.,  445 
Shaker  R.,  440 
Shahl,  984,  436-8,  471 
Shahtek,  ▼•  Shatakh 
ShaiMi,  398 

Shailtt,  Lake,  ▼•  Bolama 
Sbaitan  D.,  904,  989,  483 
Shakhberat,  185,  x88,  478 
Sham  R.,  368 

Shamkhor  R.,  394,  369,  374 
Shamlughi  3x6*90 

9J. 


*      V 


5H 


Cmi^j  &f  Armenia 


1^477 
<Staiick)l  IIS.  aYf 


iM.  179-3.44^ 


to,  It 
IX.M4 


SUrraa,  S76-7 
Sliokb.394 


Sbacr-«ba,  460 


47J 

Sbup,  ▼.  Sheep 

Shnnil^  408 

Shwha,  6,  t6^  313,  331-3 

Shnsluikcnty  313,  361 

Signakh,  8,  333 

Sil]er.3S8 

SiiDoned,  459 

Sinkot,  490 

Sinyerly  D.,  164 

Sipan  D.,  10,  16S-71 

Sipikdr  D.,  7,  34i-»,  456 

Sivmn,  271 

Soganty  D.,  10,  475 

Sogmaiuiva  D.,  444 

Sogutly,  475 

Sokman,  416 

Somketian   Mtt.^  a6$,  351^ 

^  374-6,  397 

Sor,  158 


S<3» 


459 

t4te 

«5»-4 


409 
Svciiemdi  R.^  999 


&0P»  iia,  9tS 

tedi  D.,  9.  I64-5,  446 
Syr-er-ayr^aly  D..  6 
%ikh-Labena,  a6a 

S6a 


I 


Tidns,46o 
TMldcui.  84 

Tadvai^'ioi*^  iiXp  967-8 
TagmaoriD.,  8 
TUcht-i-SaEnaa,  98,  144^5 
Talgraka  (Tak jiOta)  O^  $,  364^ 

4IS-3,  4^3,  473 
Talakketl  D^  380 

Talidi  Mts.,  <p  1^5  | 

Tafcci  lis, '971 

Tambali  pa«,  1S9 

Tanali  R.,  449-3 

Taodmek  D.  (S.  of  Aiant)*  to-\ 

.t  (I 

Taatm  IL,  439 

Tapavnk,  t.  Tcgluiit 

Tanqano,  367 

Tarkhankala,  314 

Tathkenen,  54 

T>iriik6pri,  aco-i 

Tiatev  (TalieiX  3*9 

Tknroa,  8^  9, 11,^100,  io»,  108-0^ 
9664u  30^  S5o»  3^5-^  38*4, 
4a«-4f  4S7f  4^7-^  48p.i^  4^ 

4«5-« 
Taivah|468 

TaaehlKnd»  308-9 


Takyalt^364 


Index 


5^5 


'9  V,  Ch- 

Tdumlii-kiltssa,  ▼.  Kaimpet,  Sorb 
Tcghmt  (Tapavmnk),  146,  154 
TddK  ^in  Kluushut  vallqr).  30 

„     (S.  of  Lake  Van),  ty7 
Tefanaa,  5,  $1,  54 
TdO  D.,  4a8 
Tericesliy  370-1 

Tartar  R.»  163, 3ta»  397,  361. 398 
Thiialetk-Iinmup  Rani^i  7,  8, 

««x-«i  30«.  SSSf  J^Ji  JM.  J77- 
9i  3«3f  396,  4M 
Tlflit,  7,  396,  4a6^  4a9p  480 

Tigipa  D.,  380 

Jksn^  *•#  9»  3«7f  3*St  4«4f  457 
TifeboS,4i7 

TiTHONiAH,  333 

Tkvibufi  dome,  306*  3x0,  346 

ToKOBXAM.  4^7 

Toporavan^  Lake,  469 

Toprak-kaUi,  448 

Tmrltt  steppe,  8t,  187,  45a 

Torosoe,  §41 

Tortia  D.,  8 

ToRTORUH,  161,  436a  443i  44«t 

4SOf  455 
Tortam,  Lake,  464,  479 

u      R^f  S35f  341 »  4S^f  463-4 
Trapawr  D.,  333^  335,  3451  35<f 

TreDuond,  x8 
TlUA%  Lower,  305 
Tsipa,  262 
T«ia,38o 

Tag,  333 

Ttikh,  Z08,  III,  X4S 
Tanua,  156 

Tur,  Jebd,  t.  Baaiem  hOb 
Tura  DatiO,  a 8t 
„    Diiri,  367 
Tor  Bnrftsllah,  a79  . 
Turoo-Peniaa  ftoBtkr-iaqfea,  a8x 
TcmoHXAN,  354-68, 381 
T^takb,  85 


Tndafeear  Geijaiiifl)^  465 

„  teiear  Taahkcssen),  466 

„  (in  Torcma  iFaBey),  464 

„  >Ba7ai,i87 

„  ,  Kasa^l,  466 

„  ,  Kximui,  467 
u     IL,  ▼.  IfatUu  R 

Udhdaia,  #(3,  313 
Ulek,T.  Olek 
Unaof,  175 
Uiaa  Gaa,  iM 
UraYel  R.,  480 

H^  9f  3Vt  45» 
Ur^iav^  343-5 
Unpif  440-1 

„  ,  Lake,  ix«,  184,  «96-7»  «99» 
308-9,  38a-4,  33«f  436*40^ 
469.71 

„    pli^taui,  999,  435-4*1 468 
Uxta^,  109,  ixx^  874 
Unit,  479 
Ustnpi  R.,  407 
OtnnfciKwa,  474 
Umnlyai^  374 

Vai|diir,4o8 
VALAMOiiriAV,  335-6 
Van,  448,  47s 
„  ,  Lake,  la,  98,  loa,  104.6, 

116,    15^,   267,   174.5,  446-9> 

469,  47i.a,  485-6 
Vaai  dome,  307 
Vank  (E.   Karahagh),  v.  Kand- 


,.    (W.  KaiahaghX  333 

„    (near   Erxingan),   v.   lAtfa- 

▼orich,  Smp 
Vaiak  D^  449 
Vaxisma,  41s 

Varkhan  (Werkhants),  4*4 
Vanrarin,  s68 
Vanahan  (Kadsevaa),  7,  36,  938- 

9f  4x5-6 


5«6 


Geology  of 


Vavuk  Dm  8»  3S-4,  342*3 
Vcdi  R.,  356-7 
VeimIL,  7 
Vostan,  976 

WerUiaiiti»  t.  VaiUum 

Yaghyn-Aly,  439 

Yagluja  D.p  7,  429 

Yaiji,  289,  SOI,  295,  S98,  406,  460 

Yakab,  ^il^  180 

Yenikoi,  68 

Yenikowy  393 

Yethil  Innkk,  16 

Yead  R.,  4St 

Yowully  Khan,  247 

Ypresian,  393       ^ 

Yukhaiy  Dansik,  4.  Dantikt  Ya- 

kharj 
Yungali,  177,  180 

2ab  R.,  280 

ZMg&  (ScgUk).  ^6»,  374 

Zagbt,  464 

Zagros  Mt2.|  9, .  281,  285,  306, 

3SOi  3^5-6,  iSi,  384,  4«4 
Zanga  R.,  264,  398 
Zangimar  R,  442-3 
Zaottk,  ir.  Sovuk 

Zfl^fllflT  l^^ft",  234 


fi 

99 


ZxcBsixiN,  Lower,  302-4 
„       ,  Uppec,  305. 
ZerdigeD.,  5  . 

Zemu  D.,  T.  2niek  o. 
Ziaret,  467 

Zibeneh  (Dibeneh)  R.,  467 
ana,  24 

I>.f  «3-4 
Maden,  24 

Sghag,  102,  144, 131 

Zilbi2  (Zunnt)  R.,  461 

Zunaia  D.,  466 

Zirndk  (Zernak)  D.,  70,  74*   zBj^ 

4S«»  4^7 
Zirovani,  310 

Zinila  dome  (Meschie  Mti.)*    <o, 

261.3,  306-7,  320-2,  smS^J, 

335A  343-4f   34<-9f  IS5«# 
3^3*    37«-9f    39^ 

4S8-9 
H     R-ff  262 
ZIniK,  307.8 
Zimra  D.,  270 
ZohalL,  462 

Zokh  (ZorgX  4J7,  467,  482 
Zoma  Snwam,  279 
Zoig,  V.  Zokh 

iZronk,  196,  450 
Zuba  (ChditQuR.,  46a 
Zontts  K,  V.  ZiDaa  ]L 


Hand-printed  bf  tfis  atttfaor  at  Oak  Cottage,  I>aiiiidi|!S.S.,  r903^|. 


¥" 


llAI 


II  ■ 


1