Skip to main content

Full text of "Report of the North Carolina geological survey. together with descriptions of the fossils of the marl beds"

See other formats


REPORT 


NORTH-CAROLINA   GEOLOGICAL  SURVEY. 


AGRICULTURE  OF  THE  EASTERN  COUNTIES; 


TOGETHER    WITH 


DESCRIPTIONS    OF   THE    FOSSILS    OF    THE    MAEL    BEDS 


3IIu«trati& 


or  THE 
UNIVERSITY 

r»r 


EBENEZER    E  M  M  0  N  S 


EALEIGH: 
HENRY    D.     TURNER, 

1858. 


HOLDEN    AND    WILSON, 

Printers  to  the  State. 


To  His  EXCELLENCY,  THOMAS  BKAGG, 

Governor  of  North- Carolina: 
SIR: 

I  am  gratified  that  another  opportunity  is  furnished  me  to 
express  my  obligations  to  your  Excellency  for  the  interest 
you  still  entertain  for  the  Geological  Survey  of  North-Caro- 
lina. This  fact,  while  it  has  been  extremely  gratifying,  serves 
at  the  same  time  to  impress  me  with  the  importance  of  the 
work,  and  to  excite  a  fear,  also,  that  it  may  fall  short  of  your 
expectations,  and  thus  disappoint,  not  only  yourself,  but  many 
others  who  feel  and  manifest  an  interest  in  its  success.  No 
one,  however,  could  feel  a  greater  disappointment  at  such  a 
result  than  myself;  and  fearing  that  my  labors,  together 
with  the  labors  of  those  who  assist  me,  might  fail  to  be  satis- 
factory, I  have  certainly  lost  no  time,  nor  spared  any  work, 
which  I  deemed  necessary  to  secure  the  wished-for  result. 

With  the  consciousness,  then,  of  having  done  this  much 
for  its  success,  I  submit  with  cheerfulness  this  second  report 
to  your  Excellency's  consideration. 

I  am,  Sir, 

Your  obedient  servant, 

EBENEZEK  EMMONS. 

KALEIGH,  March  1,  1858. 


PREFACE. 


THE  subjects  which  are  treated  of  in  this  Keport,  are  mostly 
practical,  and  it  has  been  my  aim  so  to  treat  them,  that  the 
matter  shall  be  useful.  The  agricultural  part  embraces  de- 
scriptions and  statements  of  the  composition  of  many  of  the 
soils  of  the  Eastern  counties.  These  samples  of  soils  which 
have  been  analyzed,  are  preserved  in  the  Geological  collec- 
tion for  future  reference.  I  have  sought  to  obtain  all  the 
practical  information  respecting  them  which  I  could,  and  for 
this  end,  the  analyses  have  been  usually  carried  as  far  as  was 
necessary.  The  number  of  soils  which  have  been  thus  sub- 
mitted to  analysis,  are  sufficient,  probably,  for  the  purposes 
intended  by  the  projectors  of  the  survey.  I  think  they  em- 
brace all  the  classes  of  soils  which  exist  in  this  section  of  the 
State.  But  there  are,  no  doubt,  many  additional  analyses, 
which  would  be  useful  where  they  appear  to  be  special  in 
their  composition,  and  exhibit  certain  peculiarities.  A  class 
of  soils  of  great  interest  exists  in  several  of  the  eastern  coun- 
ties, of  which  a  type  is  well  known  in  the  county  of  Hyde. 
I  felt  that  it  was  an  object  to  determine  the  composition  of 
this  class  with  accuracy,  and  to  see  it  in  place  with  the  bur- 
then of  its  crops  still  standing.  In  my  researches,  I  have  dis- 
covered that  this  peculiar  soil  exists  in  a  greater  or  less  degree 
of  perfection  in  several  other  counties.  In  some  instances, 
the  soil  is  the  same,  but  is  less  deep ;  in  others,  it  is  fully 
equal  to  the  Hyde  county  or  the  Mattamuskeet  lands,  both  in 
depth  and  rjchness.  It  seemed  to  be  a  prevailing  impression 
that  Hyde  county  soils  existed  no  where  else5  and  were  con- 


vi 


PREFACE. 


fined  to  that  county.  But  Onslow,  Jones,  Hanover,  Bruns- 
wick, Beaufort,  and  others,  still  possess  equally  rich  swamp 
lands. 

The  Gallberry  lands,  which  occupy  a  middle  position  be- 
tween these  rich  swamp  lands  and  the  sandy  rolling  uplands, 
are  usually  very  poor ;  but  there  are  many  tracts  which  rank 
under  this  class,  which  may  be  cultivated  profitably.  'There 
are  two  kinds  of  Gallberry  lands :  one  which  is  black  or  black- 
ish, which  consists  mainly  of  vegetable  matter,  and  a  white 
marine  sand.  This  variety  of  this  class  is  generally  too  poor 
to  pay  the  expense  of  reclaiming.  It  may  produce  a  few  tol- 
erably fair  crops  of  corn,  but  it  is  soon  exhausted,  for  it  con- 
sists only  of  sand  and  vegetable  matter.  It  may  graduate 
into  a  better  kind,  as  the  white  sand  is  exchanged  for  a  drab 
colored  one,  and  which  becomes  fine.  The  other  variety  of 
this  class,  is  clay-colored,  and  is  very  stiff,  and  mixed  with 
coarse  particles  of  flint.  It  is  almost  impervious  to  water.  It 
is  naturally  cold,  and  is  not  productive,  prior  to  draining  and 
the  employment  of  fertilizers.  It  has  a  body,  and  is  better 
than  the  black  soil  with  the  usual  admixture  of  white  sand. 

.  In  the  examination  of  soils,  the  physical  properties  require  as 
much  attention  as  the  chemical ;  for,  in  order  that  a  good  chem- 
ical mixture  of  elements  may  be  fertile,  they  should  possess  a 
certain  degree  of  adhesiveness  or  closeness,  which  will  retain 
water.  Those  which  are  porous  and  coarse,  permit  water  to 
pass  through  almost  immediately.  The  result  which  follows, 
is  fatal  to  plants,  or  crops  of  value ;  chemical  action  under 
those  circumstances  is  too  feeble  to  furnish  it  with  sufficient 
nutriment.  The  fertilizers  of  the  eastern  and  south-eastern 

counties  have  received  all  the  attention  which  could  be  be- 
* 

stowed  upon  them.  The  great  defect  which  I  find  in  their 
composition  is,  the  great  excess  of  sand.  This  element  being 
in  excess,  gives  them  only  a  local  value ;  that  is,  they  are  not 


PREFACE.  Vll 

rich  enough  to  permit  of  transportation  to  neighboring  coun- 
ties. 

In  order  to  increase  their  value,  I  have  been  led  to  enter- 
tain the  opinion,  that  they  may  be  washed.  In  this  opera- 
tion the  sand  may  be  separated  from  the  valuable  parts.  This 
opinion,  however,  requires  a  confirmation  by  experiment. 
The  material  which  remains  after  the  sand  is  separated,  con- 
tains phosphate  of  lime,  carbonate  of  lime  and  magnesia,  pot- 
ash and  soda ;  those  elements  which  make  the  marl  the  most 
valuable.  If  any  cheap  process  for  washing  the  marls  could 
be  employed,  the  material  could  be  transported  to  most  of  the 
midland  counties  with  profit. 

The  cultivation  of  the  grasses  to  a  much  greater  extent  than 
has  hitherto  been  done  in  this  State,  has  seemed  to  me  very 
desirable.  I  have  given  considerable  attention  to  the  subject, 
and  for  the  purpose  of  aiding,  as  far  as  possible,  a  measure  of 
this  kind,  I  have  selected  several  of  the  most  valuable  for  de- 
scription, that  information  respecting  their  value,  may  be 
more  widely  spread.  I  am  confident  that  many  of  them  will 
succeed.  Very  few  efforts  have  hitherto  been  made  for  their 
cultivation, — most  planters  entertaining  the  belief  that  it  is 
impracticable,  or  else  the  labors  of  the  plantation  are  supposed 
to  be  much  more  profitably  directed  to  the  raising  of  cotton. 
Under  the  present  system  of  curing  the  grasses  for  winter  fod- 
der, the  labor  is  so  much  cheapened  that  it  seems  to  me  that 
the  raising  of  cotton  or  any  other  of  the  great  staples,  will  not 
interfere  with  the  project  of  keeping  more  stock,  and  in  a 
better  condition  than  has  hitherto  been  attempted  in  the  State, 

In  connexion  with  the  marls  of  the  eastern  counties,  I  have 
given  a  brief  sketch  of  the  fossils  of  the  different  kinds  of  beds. 
Those  who  will  take,  the  trouble  to  examine  the  figures  of  the 
fossils  which  belong  to  the  different  beds,  will  not  fail  to  per- 
ceive the  striking  differences  which  prevail.  It  is,  for  instance. 


PREFACE. 

exceedingly  rare  to  find  a  species  common  to  two  beds,  al- 
though they  lie  in  juxtaposition ;  or  one  may  repose  upon  the 
other.  Hence,  the  utility  of  the  presence  of  fossils  to  distin- 
guish beds  belonging  to  the  different  epochs  from  each  other. 
Another  object  which  I  had  in  view  in  occupying  so  much 
space  upon  this  subject,  was,  to  aid  those  who  wish  to  become 
acquainted  with  this  interesting  subject.  Geology  is*  now 
commanding  the  attention  of  some  of  the  best  minds  in  this 
country  and  Europe.  It  is  invested  with  great  importance 
and  interest,  as  it  is  through  the  discoveries  in  this  depart- 
ment of  science,  that  we  obtain  a  knowledge  of  the  ancient 
history  of  the  globe.  This  pursuit  is  especially  recommended 
to  the  attention  of  the  young.  It  will  be  found  extremely 
interesting  and  useful,  and  no  one  will  regret  afterwards  that 
he  devoted  a  portion  of  his  leisure  hours  to  its  study. 


TABLE  OF  CONTENTS. 


PRELIMINARY  REMARKS.    1 — 1. 

CHAPTER  I. 

Reference  to  a  former  report. — Dependence  of  seed  on  the  perfection  of  the  soil. — 
Nutrient  matters  necessary  to  animal  life  traced  to  the  soil. — Essential  elements  of 
a  good  soil. — Character  and  classification  of  the  soils  of  the  Eastern  counties. — 
Importance  of  determining  the  smallest  percentage  of  earthy  matter  in  a  vegetable 
soil,  which  is  compatible  with  a  remunerating  crop. — Some  elements  more  essential 
to  form  a  good  soil  than  others. — The  organs  of  a  plant  are  composed  of  different 
elements. — The  extremes  of  certain  kinds  of  soil. — Remarks  on  the  adaptation, 
together  with  a  statement  of  their  composition.— Soil  of  the  Open  Ground  Prairie 
in  Carteret  county. — Pocoson  and  Swamp  Lands. — Soils  of  Hyde  county.  8 — 22. 

CHAPTER  II. 

The  best  soil  of  Dr.  Long,  of  Hyde  county — its  composition — its  yield  of  corn  per 
acre. — Mr.  Burroughs'  soil  of  the  north  side  of  Mattamuskeet  Lake. — Amount  of 
inorganic  matter  which  a  crop  of  corn  removes  from  the  soil. — Each  organ  to  be 
furnished  with  appropriate  nutriment.— Maize  an  exhausting  crop. — Soils  from  the 
plantation  of  Gen.  Blount. — Gen.  Blount's  letter.  22—  36. 

CHAPTER  III. 

Topography  of  the  Eastern  counties,  from  Wake  eastward  to  Onslow.— Character  of 
the  soil  of  the  White  Oak  Desert. — Mr.  Francke's  Pocoson  and  Swamp  Lands. — 
The  better  kind  of  Gallberry  Swamp  Lands. — Mr.  McNeil's  soil;  will  pay  for 
drainage. — Barren  soil  of  Bogue  Sound,  furnished  by  D.  A.  Humphrey,  Esq. — 
Cause  of  barrenness  of  these  soils.  36 — 48. 

CHAPTER  IV. 

Soils  of  Jones  county,  taken  from  the  plantation  of  J.  H.  Haughton,  Esq. — Composi- 
tion of  a  brown  earth  overlying  the  marl. — Recapitulation.  48 — 59. 

CHAPTER  Y. 

FERTILIZERS.— What  constitutes  a  Fertilizer.— Sources  of  Fertilizers.— Those  from 
the  vegetable  kingdom  consist  of  the  ash. — Ash  of  plants  resembles  in  composition 
the  inorganic  matter  of  soils. — Quantity  of  fertilizing  matter  removed  from  the 
soil  by  different  plants. — Methods  to  be  adopted  in  order  to  prevent  waste  of 
fertilizing  matter. — How  restored.  S9 — 78. 

CHAPTER  VI. 

FBRTILIZERS  (CONTINUED.) — Marl  beds. — The  different  periods  to  which  they  belong, 
and  their  relation  to  each  other.  78—89. 

CHAPTER  VII. 

FERTILIZERS  (CONTINUED.) — Stone  Marl — its  economical  value. — Composition  of  the 
Green  Sand  of  Cape  Fear  River.  89—101. 


X  TABLE   OF   CONTENTS. 

CHAPTER  Till. 

Eocene,  or  White  Marl.— Quantity  of  lime  variable,  but  greater  than  the  average  of 
other  varieties.— The  Wadsworth  beds.— His  letter.— Beds  upon  the  Neus«— 
Haugh ton's  marl. — Composition,  101 — 107. 

CHAPTER  IX. 

Shell  marl. — Heterogenous  in  its  composition  and  arrangement  of  materials. — Chemi- 
cal constitution. — Application  of  marl. — Poisonous  marl. — How  corrected. — Theories 
respecting  the  operation  of  marl.  107 — 126. 

CHAPTER  X. 

Animal  manures. — Fish, — Crabs. — Cancorine. — Compost  of  crabs. — Preservation  of 
the  offal  of  fish.  126—132. 

CHAPTER  XL 

Clay. — Characteristics  of  a  good  clay. — Composition  of  fine  clays. — Composition  of  a 
clay  upon  Bogue  Sound.  132 — 135. 

CHAPTER  XII. 

The  grasses  and  their  functions. — Different  objects  attained  by  their  cultivation. — 
Chemical  constitution  of  the  grasses. — Elementary  organs  and  parts  of  the  blos- 
som.— Division  of  grasses. — Southern  genera. — Cultivated  species  with  their  de- 
scriptions and  properties.  135 — 181. 

CHAPTER  XIII. 

Red  clover. — Organic  constitution. — Composition  of  the  ash. — Differs  in  composition 
from  the  true  grasses. — Failures  in  its  cultivation. — For  a  green  crop. — Lucerne. — 
Sanfoin.  181—188. 

CHAPTER  XIV. 

Methods  by  which  the  valuable  grasses  may  be  cultivated  successfully. — Soiling  and 
its  advantages.  188 — 193. 

CHAPTER  XV. 

PALAEONTOLOGY. — Fossils  of  the  green  sand  and  tertiary. — Mammals. — Horse.— Hog. — 
Mastodon  and  elephant. — Deer. — Whales,  or  celaceans.  193 — 213. 

CHAPTER  XVI. 

Description  of  reptilian  remains  of  the  marl  beds  of  North-Carolina.— Reptiles  of 
the  green  sand.  213—225. 

CHAPTER  XVII. 

PISCES.— Description  of  the  remains  of  fish  in  the  North-Carolina  marl  beds.  225— 
245. 

CHAPTER  XVIII. 

MOLLOSCA. — Description  of  the  cephalopods,  gasteropods  and  lamellibranchiata.  245 — 
303. 

CHAPTER  XIX. 
&ADIATA. — Description  of  the  echinoderns. — Sea  urchins. — Polyparia.    303 — 314, 


INTRODUCTION. 


IT  is  one  of  the  distinguishing  characteristics  of  the  day  to 
attempt  to  utilize  science.  The  leading  minds  of  the  age 
seem  to  be  as  intensely  engaged  in  diffusing  knowledge  and 
disseminating  it  as  common  stock,  as  they  are  in  acquiring  it 
for  themselves.  The  consequences  which  have  already  flowed 
from  their  efforts,  are,  to  have  already  made  knowledge  re- 
lating to  many  departments  the  common  property  of  the 
masses.  This  knowledge  is  not  probably  exact  in  many  indi- 
viduals, perhaps  in  none,  excepting  those  who  make  those 
subjects  objects  of  special  study;  but  then,  they  know  the 
nature  of  the  subjects  treated  of,  as  well  as  many  of  the  con- 
clusions which  have  been  obtained.  They  know  enough 
to  make  intelligent  inquiries,  and  a  subject  matter  for  con- 
versation ;  their  minds  are  sufficiently  informed  to  lead  them 
upon  the  proper  road  of  inquiry.  More  than  this  has  been 
gained  in  many  instances  in  common  life.  The  way  is  al- 
ready prepared  for  a  general  diffusion  of  knowledge.  Of 
those  subjects  which  are  the  most  useful  to  society,  none 
occupy  a  higher  rank  than  those  which  are  related  to  agri- 
culture. Thus,  the  chemistry  of  agriculture  is  of  the  highest 
importance.  The  mechanics  of  agriculture  are  also  impor- 
tant, and  more  attention  has  been  paid  to  this  branch  than 
the  former.  Indeed,  one  of  the  first  evidences  that  agricul- 
ture was  really  upon  the  road  of  improvement,  was  the  ap- 
preciation of  better  implements  of  husbandry.  Their  im- 
provement was  first  attempted.  It  was  right  that  it  should 
be  so,  for  to  make  chemical  principles  available  at  all,  it  was 
necessary  to  change  by  mechanical  means  the  condition  of  the 
soil.  Improvements,  then,  in  agriculture,  began  at  the  right 
end.  The  more  abstruse  principles  of  the  business  have 
become  subjects  of  investigation  since,  and  now  there  are 
but  few  farmers  who  are  entirely  ignorant  of  the  chemistry 
and  other  collateral  branches  of  the  philosophy  of  agriculture. 


xii  INTRODUCTION. 

But  still,  these  important  hand-maids  to  this  indispensable 
calling  have  only  just  begun  to  exercise  an  influence  over 
old  modes  and  old  practices.  But  two  great  obstacles  to  the 
introduction  of  rational  methods  in  agriculture  are  being  rap- 
idjy  removed ;  that  is,  prejudice  in  favor  of  the  old  methods 
pursued  by  the  fathers,  and  prejudice  against  innovation. 
Whatever  is  good  in  the  old  methods  will  be  retained ;  and 
ultimately,  what  is  erroneous  and  worthless  will  be  rejected. 
Improvements,  however,  in  agriculture,  are  necessarily  slow 
in  getting  a  foothold ;  much  more  so  than  in  the  mechanic 
arts  \  for  there  are  stronger  prejudices  to  be  overcome,  and 
in  the  former  it  seems  there  is  a  ready  appreciation  of  value 
in  every  improvement  which  is  made,  while  in  the  latter,  a 
prejudice  is  to  be  first  overcome  by  ample  experience.  But 
we  ma}^  be  assured  that,  sooner,  or  later,  the  benefits  of  a 
change  will  appear,  as  the  improvements  address  themselves 
to  men's  pockets^  which  is  one  of  the  most  influential  of  mo- 
tives in  common  life. 

The  principles  which  control  industrial  pursuits  are  per- 
fectly simple ;  and  being  simple,  have  been  and  still  will  be 
liable  to  be  overlooked.  Who  among  the  merchants  of  a 
village,  acquires  most  rapidly,  ease  and  independence  for 
himself?  It  is  the  one  who,  from  a  more  extensive  acquaint- 
ance with  his  occupation,  a  more  attentive  observation  of  the 
markets^  and  a  more  careful  application  of  his  judgment,  un- 
tiring energy  and  prudent  industry,  buys  the  best  article  and 
sells  it  the  cheapest. 

Who,  among  the  mechanics  of  the  town,  commands  the 
business  in  his  special  line  of  production?  It  is  that  one  who 
has  been  thoroughly  instructed  in  the  principles  of  his  handi- 
craft, applies  his  mind  and  judgment  to  his  labor,  and  by  that 
means  improves  the  articles  he  makes,  or  the  modes  of  its 
manufacture,  and  can  thereby  outstrip  his  competitors  by 
manufacturing  more,  as  well  as  better,  and  selling  cheaper, 
ft  is  a  natural  result — a  simple  law  of  trade  and  commerce. 

But  who  among  the  agriculturists  of  the  land  are  the  most 
prosperous?  It  is  he  who  is  not  content  to  follow  the  beaten 
track  of  his  forefathers,  or  pursue  olie  course  which  they  have 
pursued,  and  because  they  pursued  and  beat  it,  but  he  who 


INTRODUCTION.  Xlll 

thoroughly  imbues  his  mind  with  sound  principles,  who 
studies  into  the  nature  of  his  processes,  and  the  reason  why 
he  does  this  in  preference  to  that ;  who  investigates  the  na- 
ture of  his  soils,  and  fits  them  most  perfectly  for  his  crops, 
and  is  moreover  seasonable  in  his  preparation.  He  will  raise 
the  most  to  the  acre,  and  have  more  to  sell,  and  can  sell  the 
cheapest,  and  make  the  most  money.  The  greatest  production, 
coupled  with  the  best,  controls  the  pockets  of  the  purchasers, 
and  insures  to  him,  what  is  ever  sought  after,  the  earliest  in- 
dependence and  the  first  honors  in  the  line  of  a  profession. 

What  lies  at  the  foundations  of  commerce  ?  What  spreads 
her  sails,  or  generates  the  steam  of  our  floating  castles  which 
ply  from  port  to  port  and  from  country  to  country  ?  It  is  ag- 
ricultural production.  There  is  no  other  substratum  upon 
which  the  business  of  the  world  can  rest.  Nothing  else  can 
impel  the  mighty  engines  of  commerce,  or  set  in  motion  the 
locomotive,  with  its  heavy  train  of  cars.  It  is  not  because 
the  merchant  buys  and  sells. again.  That  is  not  production. 
But  it  is  because  the  farmer  produces.  The  other  is  but  a 
transfer,  and  is  only  an  incident  in  trade.  The  production  is 
the  ruling  cause.  It  is  that  which  supports,  which  moves. 
Put  a  stop  to  production,  and  the  wheels  cease  to  move,  the 
paddle  ceases  to  turn,  the  locomotive  stands  still,  and  the 
whistle  is  no  longer  heard.  Production  is  the  great  element 
of  life  in  commerce  and  manufactures.  It  is  because  agri- 
culture exists,  that  commerce  thrives,  that  the  merchant  can 
buy  and  sell.  The  earth  is  properly  called  the  common 
mother  ot  all.  Her  fruits  nourish  us,  and  supply  the  mate- 
rials for  the  arts  and  manufactures,  and  the  articles  for  trade 
and  commerce.  The  earth  is  the  mother  of  all,  but  that  does 
not  justify  the  agriculturist  in  waiting  for  her  fruits  with 
folded  arms,  and  to  neglect  to  store  his  mind  with  the  ele- 
ments and  principles  of  agricultural  knowledge,  or  hope,  in 
inactivity,  on  a  good  Providence,  or  good  fortune.  If  mother 
earth  is  rightly  depended  upon,  it  will  be  accompanied  by 
works  and  the  study  of  principles  as  connected  with  what  he 
is  to  do  for  his  soils.  He  cannot  ask  much  of  mother  earth, 
who  neglects  to  study  elements  and  principles  in  this  connec- 
tion. I  say  elements  and  principles,  for  it  is  not  enough  to 


XIV  INTRODUCTION. 

know  the  mechanical  part.  It  is  not  enough  to  know  how  to 
plow,  and  reap,  and  mow ;  these  are  a  part  of  an  education, 
but  it  is  not  all  of  it. 

Thus,  we  see,  the  commanding  position  of  agriculture.  Its 
position  is  commanding,  independent  of  the  mode  in  which 
a  community  of  individuals  conduct  it.  As  it  regards  this 
section  of  the  Union,  its  importance  increases  with  the  popu- 
lation of  our  country.  The  Agriculturalist  is  not  restricted 
to  the  production  of  bread.  While  her  granaries  are  over- 
flowing with  corn  and  wheat,  she  has  still  two  other  great 
staples  of  trade  to  arouse  her  energy:  cotton  and  tobacco. 
These  have  been  and  are  increasing  in  importance  from  the  day 
the  first  seed  germinated  in  her  soil.  These  are  money  crops. 
In  all  these  great  staples,  industry  need  not  be  paralyzed,  nor 
the  spirits  be  made  to  sink  for  want  of  a  market.  No  one 
needs  fear  that  a  surplus  will  be  left  on  his  hands ;  that  his 
toils  will  be  unrewarded  or  his  industry  avail  him  nothing. 
Such  is  the  condition  of  the  world,  that  the  great  staples  are 
sought  for  from  necessity.  Cotton  must  be  had  at  any  price 
to  satisfy  the  imperative  wants  of  the  world.  The  loom  can- 
not stand  still.  The  necessities  of  thousands  now  demand  it.  The 
force  of  habit  in  the  use  of  tobacco  is  so  strong  and  so  general, 
that  its  price  can  never  be  less  than  it  is  now.  It  is  rather  prob- 
able that  it  will  be  higher.  Its  production  may  be  cheapened, 
its  cost  may  be  diminished,  but  its  price  in  market  will  never 
be  less.  The  advantage  will  ever  be  on  the  side  of  the  produ- 
cer. Farming,  then,  has  an  advantage  over  all  professions. 
There  may  be  too  many  lawyers,  physicians  and  merchants, 
but  never  too  many  farmers.  This  is  so,  because  the  seaports 
of  the  world  are  their  markets,  and  because  there  is  a  world  of 
human  families  which  are  not  producers,  and  hence  have  to 
be  fed,  their  looms  kept  running,  and  their  habits  gratified. 

It  is  not,  therefore,  the  domestic  market  which  is  to  be  sup- 
plied. The  products  of  the  soil  of  North-Carolina  are  con- 
sumed far  away ;  some,  in  the  cities  of  the  north,  but  a  far 
greater  amount  by  the  population  of  the  Old  World.  Impor- 
tant measures  are  being  taken  abroad  to  supply  cotton  for 
English  manufactories  from  India  and  Africa,  and  no  doubt 
with  the  hope  that,  ultimately,  this  nation  may  make  herself 


INTRODUCTION.  XV 

independent  of  this  country  with  respect  to  this  indispensable 
article.  A  project  of  this  nature  must  be  regarded  with  some 
concern.  It  cannot  succeed  immediately,  and  it  is  doubtful 
whether  cotton  can  be  produced  in  those  countries,  so  as  to 
compete  successfully  in  market  with  our  own.  In  the  first 
place,  the  husbandry  of  cotton  is  fully  understood  in  the  Sou- 
thern States;  and  in  the  second  place,  the  adaptation  of  climate 
and  soil  is  perfect,  and  the  means  for  supplying  fertilizers  to 
sustain  its  continued  production  are  equally  well  established. 
Marl  is  the  true  fertilizer  for  cotton.  This  is  fully  established 
by  experience  and  chemical  analysis.  All  these  facts  put  it 
in  the  power  of  the  South  to  sustain  vigorously,  for  an  inde- 
finite term  of  years,  its  production.  From  the  Roanoke  to 
Florida,  this  fertilizer  in  numberless  forms  is  inexhaustible. 
Hitherto,  it  has  been  almost  impossible  to  be  satisfied  that 
there  has  been  a  systematic  and  sustained  effort  to  carry  this 
production  to  the  limit  which  the  want  of  it  abroad  demands. 
The  time,  however,  has  come,  when  its  production  has  be- 
come doubly  important.  The  hopes  of  foreigners  for  success 
in  supplying  themselves  with  cotton  from  India  and  Africa, 
are  based  in  a  good  degree  upon  its  failure  here,  through  some 
misfortune,  such  as  political  revulsion,  exhaustion  of  the  soil 
and  other  casualties  which  may  occur,  but  which  cannot  now  be 
foreseen.  As  it  regards  the  exhaustion  of  the  soil,  there  need 
be  no  fear,  with  the  resources  at  command.  It  is  true  that  large 
tracts  have  been  exhausted,  but  agriculture  is  understood 
better  now  than  formerly ;  and  hence,  the  planter  is  abundant- 
ly able  to  forestall  such  an  event  and  prevent  its  occurrence. 
But  in  any  event,  the  principles  stated  in  the  foregoing 
paragraphs,  will  govern  the  market.  The  best  and  cheapest 
article  will  be  bought,  and  that  will  insure  its  sale  in  any 
quarter  of  the  globe,  in  spite  of  the  combination  of  Cotton 
Associations  to  produce  it  in  India  and  Africa.  If  American 
planters  can  produce  the  best  at  a  lower  rate  than  it  can  be 
produced  in  India,  then  American  cotton  will  find  a  market 
in  Liverpool.  It  is  a  simple  question  of  production ;  and  for- 
eign efforts  to  secure  a  market  and  exclude  the  American 
cotton,  will  result  simply  in  arousing  the  cotton  planter  to 
make  a  successful  effort  to  retain  his  foothold  in  the  market 


XVI  rNTKODUCTIOlT. 

which  he  now  supplies.  "When  the  cotton  planting  States  have 
once  fully  taken  into  consideration  their  immense  advantages 
for  production,  it  seems  impossible  that  they  should  sleep  over 
them.  Cotton,  Indian  corn,  wheat  and  tobacco,  four  great 
staples  on  their  hands,  for  which  the  markets  of  the  world  are 
open.  These  minor  productions  of  the  homestead  furnish  busi- 
ness for  all.  The  Alleghanies  and  their  slopes  are  well  adapted 
to  grazing,  and  hence  the  raising  of  stock  will  become  an  item 
of  immense  importance  to  planters.  Intercourse  with  the  ex- 
tremes, the  east  and  the  west,  will  soon  be  made  easy.  It  will  be 
cheap,  if  an  enlightened  policy  controls  the  fare  upon  railroads. 
If  an  opposite  policy  should  unfortunately  prevail,  the  hopes 
of  the  planter  and  graizer  will  be  partially  disappointed. 

The  encouragement  for  pursuing  agriculture  may  be  found 
in  the  certain  prospect  of  the  mining  resources  of  the  State. 
In  the  various  branches  of  this  business,  it  will  ultimately  be 
found,  that  a  large  population  will  have  to  be  fed.  A  popu- 
lation devoted  to  this  interest  are  not  producers  of  bread, 
meat  or  fruits.  They  are  necessarily  dependent  for  all  these 
and  more ;  and  hence,  a  home  market  is  furnished,  which,  as 
far  as  it  goes,  is  as  important  as  the  foreign. 

But  I  need  not  dwell  on  the  importance  of  agriculture;  its 
importance  is  felt.  I  was  more  anxious  in  this  connection,  to 
state  my  views  of  an  improved  agriculture ;  one  which  is  un- 
derstood, or  one  founded  upon  established  principles, — one 
which  leaves  a  beaten  road  and  inquires  into  the  why  and 
wherefore.  This  is  the  only  kind  of  agriculture  which  will 
elevate  the  masses,  and  give  laborers  a  status  or  standing 
beside  professional  men,  and  enable  them  to  exercise  an 
influence  as  wide  as  theirs.  Regarded  in  this  light,  it  is  not 
simply  an  extraordinary  crop,  which  is  to  be  produced,  but  it 
is  a  development  of  the  mental  faculties.  These  are  compati- 
ble objects.  Indeed,  they  go  almost  necessarily  together,  be- 
cause they  are  the  result  of  an  exercise  of  the  mind.  The 
labor  of  thinking  is  involved, — a  labor  which  is  not  at  first 
performed  without  effort, — for  that  reason  many  prefer  to  let 
others  think  for  them ;  and  hence,  they  continue  in  that  un- 
enviable condition  which  is  properly  called  a  statu  quo* 

RALEIGH,  March  1, 1858. 


REPORT 

<  »  ' 

OF  THE 

NORTH-CAROLINA  GEOLOGICAL  SURVEY. 

t 

AGRICULTURE  OF  THE  EASTERN  COUNTIES. 


PKELIMINAKY  EEMAKKS. 

FOR  any  thing  we  know  to  the  contrary,  there  is  such  an 
ample  provision  in  the  economy  of  nature,  that  the  produc- 
tion of  food  shall  not  depend  upon  skill,  or  a  deep  acquaint- 
ance with  the  laws  of  life. 

Seeds  are  sown  broadcast,  the  winds  waft  them  from  their 
parent  stocks,  or  they  fall  unheeded  to  their  roots ;  yet  such 
is  the  relation  of  seed  to  earth,  air  and  moisture,  that  they 
germinate  and  become  new  individual  plants  which,  in  due 
time,  contain  the  appropriate  nutriment  for  some  existing 
organism.  It  may  be  it  is  food  only  for  the  insect  tribes,  the 
beast  of  the  field,  or  it  may  serve  the  table  of  the  Prince. 

The  simple  growth  and  nutriment  of  plants  is  independent 
of  science,  high  culture,  or  skill  in  the  ordinary  round  of 
nature. 

There  is  a  provision  to  meet  a  certain  amount  of  the  wants 
of  life,  so  far  as  food  is  concerned,  which  may  be  obtained 
without  tillage.  It  is,  however,  limited.  When  the  habita- 
tions of  men  become  concentrated  upon  a  comparatively 
small  area,  or  a  dense  population  fills  the  land,  the  natural 
magazine  which  furnishes  the  ordinary  or  regular  supply  of 
nutriment  to  the  vegetable,  especially  the  cereals,  then 
becomes  insufficient  to  supply  the  increased  demands  of  nun> 


2  NORTH-CAROLINA    GEOLOGICAL   SURVEY. 

bers,  and  hence  the  natural  resources  fail,  and  there  ever 
afterwards  exists  a  demand  for  skill  and  science  to  meet  these 
artificial  wants. 

The  first  efforts  to  supply  the  meat  and  bread  of  a  dense 
population,  in  the  earliest  stages  of  society,  are  those  which 
belong  to  the  simplest  kinds.  They  consist  mainly  in  provid- 
ing more  room,  light  and  air,  or  providing  for  the  free  pene- 
tration of  roots  through  the  soil,  and  the  exclusion  of -weeds 
or  unnutritive  plants.  But,  inasmuch  as  nutritive  matter  is 
measured  out  and  limited,  and  as  there  is  no  special  provision 
to  create  a  new  supply,  constant  removal  will,  in  the  course 
of  years,  so  far  diminish  the  original  stock,  that  the  plant 
ceases  to  grow  or  perfect  its  fruit,  or  does  so  under  circum- 
stances less  favorable  for  its  perfection. 

At  this  period  it  becomes  necessary  to  inquire  how  fertility, 
when  lost,  may  be  restored ;  and  this  inquiry  becomes  more 
pressing  in  the  direct  ratio  that  the  population  has  increased. 

Experience  does,  or  may  step  in  and  postpone  the  period 
of  exhaustion,  and  partially  supply,  for  a  time,  the  nutritive 
elements.  But  generally  these  shifts  to  postpone  the  period 
of  exhaustion  fail,  for  they  are  merely  the  efforts  of  the  em- 
pyric.  Empyricism  in  no  business  is  likely  to  lead  to  the 
discovery  of  sound  principles ;  indeed,  it  cannot  inform  us  of 
the  fact  of  exhaustion  at  all ;  and  hence,  empyricism  is  not  in 
the  direct  road  to  improvement.  In  one  instance  it  may  prove 
successful,  but  in  the  many  it  fails ;  as  it  cannot  assign  a  cause 
or  state  a  reason. 

The  perfection  of  cultivation,  or  the  perfection  of  agricul- 
ture, demands  a  reason ;  and  the  period  when  a  reason  can 
be  assigned  may  be  regarded  as  the  third  stage  of  improve- 
ment. It  is  at  this  stage  that  agriculture  requires  a  direct 
inquiry  respecting  cause  and  effect ;  or,  in  other  words,  into 
antecedents  and  consequents,  in  order  that  it  may  make  pro- 
gress when  the  rules  of  empyricism  fail.  Agriculture,  in  some 
of  its  scientific  aspects,  has  obscurities,  because  it  has  en- 
quiries to  make  which  are  closely  related  to  those  of  life ;  and 
life,  whether  regarded  as  a  mysterious  principle,  or  a  force 
dependent  upon  chemical  relations,  or  chemical  actions,  ia 


NOKTH-CAKOLINA   GEOLOGICAL   SURVEY.  3 

profoundly  mysterious.  Calling  this  force  life,  without  at- 
tempting to  tell  what  it  is,  we  know  that  it  controls  all  the 
results  effected  in  and  by  the  vegetable  tissues.  An  organ, 
as  a  whole,  possesses  no  force :  the  leaf  has  no  force,  neither 
have  the  steins,  bark  or  kernel.  The  force  alluded  to  resides 
in  the  cell ;  and  hence  it  is  sometimes  called  a  cell  force,  and 
the  sum  or  aggregate  force  of  all  the  cells  of  an  organ  secures 
all  the  results  in  their  proper  season.  The  matured  fruit  is 
the  result  then  of  the  combined  force  of  all  the  cells  which 
compose  it,  acting  under  the  influence  of  outward  forces,  as 
air,  light  and  heat. 

The  sum  or  aggregate  of  these  changes,  however,  from 
germination  to  the  consumation  of  the  mature  fruit,  is  con- 
cealed from  view.  We  know  only  the  simple  fact,  that  of 
change  from  day  to  day.  Of  the  effective  agency  residing 
in  the  cell  we  know  nothing.  But  fortunately  the  questions 
which  belong  to  scientific  agriculture  have  only  a  slight  re- 
lation to  these ;  they  are  not  questions  relating  to  cell  force, 
or  to  life  in  the  abstract.  These  are  one  step  farther  back 
than  it  is  necessary  to  carry  them.  We  need  make  no  in- 
terrogatories respecting  cell  force,  or  life,  in  order  to  till  the 
soil  in  the  best  modes,  or  to  grow  large  crops  of  wheat.  But 
still  these  obscure  questions  bear  a  relation,,  sufficiently  close 
to  darken  or  cioud  those  which  must  be  answered,  and  we 
almost  instinctively  pass  from  those  investigations  which  lie 
in  the  field  of  research  to  those  which  are  a  step  farther  back, 
and  lie  beyond  the  limits  of  legitimate  enquiry. 

§  2.  The  field  of  investigation  is  really  much  nearer  to  us 
and  more  within  the  scope  of  legitimate  inquiry.  If  we 
wish  to  know  what  is  the  appropriate  food  of  the  wheat  plant, 
we  have  only  to  analyse  it,  or  to  determine  the  elements 
which  compose  the  kernel.  It  is  not  how  it  is  made,  how  the 
cell  power  operates,  but  simply  what  the  constituents  of  the 
wheat  or  corn  plant  are. 

In  practice,  then,  the  farmer  is  merely  required  to  sow  his 
wheat  upon  grounds  which  contain  enough  of  the  elements 
it  wants.  It  is  true,  certain  collateral  questions  of  great  im- 
portance have  to  be  answered,  such  as  those  which  relate  to 


4  NORTH-CAROLINA   GEOLOGICAL   SURVEY. 

the  physical  condition  of  the  soil,  the  measures  which  ought 
to  be  adopted  to  prevent  the  operation  of  injurious  agents, — 
as  frost,  drought,  depredations  of  insects,  etc. 

"When  experiments  and  observation  have  satisfied  the  far- 
mer respecting  the  composition  of  wheat,  corn,  and  of  the 
soil  in  which  they  are  to  be  planted,  he  has  only  to  secure  the 
proper  mechanical  condition  of  the  soil,  and  put  it  into  that 
state  which  is  best  adapted  to  their  constitution.  From  the 
foregoing  statement,  it  is  evident  that  the  range  of  scientific 
enquiry  is  limited  to  an  experimental  circle.  The  farmer  is 
not  required  to  go  out  of  that  area  to  determine  the  true 
theory  of  agriculture,  to  perfect  the  art  or  the  practical  part 
of  the  business. 

§  3.  The  following  report  is  based  on  the  preceding  views 
relative  to  the  scope  or  range  of  agricultural  enquiry.  The 
planter  or  farmer  may  speculate  on  vital  or  chemical  forces, 
and  form  such  theories  upon  those  recondite  forces  as  best 
comport  with  his  knowledge  of  facts  and  principles ;  yet,  as 
has  been  said  already,  practical  enquiries  do  not  extend  to 
them ;  it  only  demands  a  range  of  knowledge  which  is 
bounded  by  experimental  researches,  and  the  deductions 
which  legitimately  follow  therefrom. 

It  is  therefore  f«rue,  that  enquiries  into  the  nature  of  the 
ceil  force  or  vital  ibrcQ  are  not  excluded  from  the  philosophy 
of  vegetation,  but  these  ultimate  interrogatories  have  no  prac- 
tical utility,  so  far  certainly  as  the  principles  of  culture  are 
concerned.  From  these  remarks,  however,  it  should  not  be 
inferred  that  agriculture  requires  only  an  extremely  limited 
range  of  knowledge — that  its  connections  with  other  sciences 
are  distant  and  doubtful.  So  far  is  this  from  being  true,  that 
it  may  be  shown  that  it  is  intimately  related  to,  and  de- 
pendent upon,  several  of  the  important  branches  of  knowl- 
edge. We  have  seen,  for  example,  how  important  chemistry 
is  to  agriculture.  To  this  it  is  wholly  indebted  for  its  won- 
derful progress  in  modern  times.  Climatology  also  is  closely 
related  to  agriculture,  inasmuch  as  a  knowledge  of  the  influ- 
ence of  light  and  heat,  air  and  winds,  height  and  depth,  must 
influence  the  farmer  in  his  selection  of  crops  for  tillage,  and 


NORTH-CAROLINA   GEOLOGICAL   SURVEY.  O 

the  modes  by  which  they  should  be  treated.  Soils  too,  being 
derived  from  rocks  of  different  periods  and  constitutions,  in- 
fluence their  composition  and  capabilities  more  or  less.  Close 
observation  relative  to  those  influences  frequently  establish 
important  generalizations;  and  hence,  geology  may  be  re- 
garded as  a  department  very  intimately  connected  with  agri- 
culture, and  whose  principles  are  capable  of  advancing  its 
interests. 

It  is  scarcely  necessary  to  refer  to  botany,  as  an  allied 
branch  of  science.  A  practical  knowledge  of  soils  may  be 
derived  from  it.  Nature  rarely  errs  in  collocation.  Plants, 
without  selecting  soils  in  truth,  do  really  flourish  best  on  cer- 
tain tracts  whose  soil  is  found  to  be  adapted  to  their  special 
wants.  Some  are  lime,  others  are  potash  plants ;  and  hence, 
the  farmer  may  be  satisfied  where  certain  plants  abound,  that 
certain  important  constituents  of  soils  are  present. 

Animals,  however,  form  a  large  part  of  his  care  and  over- 
sight. Often  his  chief  wealth  consists  in  cattle.  The  rearing 
of  stock  of  favorite  breeds,  their  improvement  in  general,  and 
often  in  special  points,  demands  a  knowledge  of  physiology 
and  anatomy.  There  is  property  in  a  knowledge  of  the  foot 
of  the  horse,  the  joints  of  the  bullock  and  the  structure  of  the 
hoof.  There  is  property  in  a  knowledge  of  the  skull  and  the 
physiognomy  of  the  horse  and  the  kine ;  and  there  is  pro- 
perty in  the  knowledge  of  habits  and  best  food  for  cattle  and 
flocks,  and  in  the  organization  of  the  stomach  and  its  depen- 
dencies. 

The  farmer  and  planter,  therefore,  may  say  that  they  have 
not  only  property  in  lands  and  in  cattle,  but  also  in  the  phe- 
nomena of  nature,  as  they  may  make  those  phenomena  sub- 
servient to  their  interests;  the  sunbeam  and  shade  add 
golden  dust  to  their  stores,  when  seed  times  and  tillage  are 
chosen  under  the  guidance  of  philosophy. 

§  4:.  "While  agriculture  in  all  its  aspects  presents  a  wide 
field  for  investigation,  it  still  has  very  clearly  such  subdivi- 
sions of  labor,  that  in  practice,  it  may  reach  a  high  degree  of 
perfection.  We  find,  for  example,  that  climate  frequently 
restricts  the  most  profitable  productions  to  one  or  two  staples. 


6 


NORTH-CAROLINA   GEOLOGICAL   SURVEY. 


Cotton  cannot  be  grown  with  profit  north  of  Yirginia.  The 
sugar  cane  and  coffee  return  profits  only  on  our  most  southern 
border.  Tobacco,  though  not  so  strictly  limited  by  parallels 
of  latitude,  still  requires  certain  peculiarities  of  climate  and 
soil,  which  greatly  restricts  its  cultivation.  Tea  requires  a 
peculiar  climate.  In  some  parts  of  the  world  it  rarely  or 
never  rains;  in  others,  rains  are  frequent;  in  others  still, 
there  are  seasons  of  rain  followed  by  others  which  are  rain- 
less. These  peculiarities  favor  the  growth  and  perfection  of 
a  class  or  a  family  of  plants,  while,  at  the  same  time,  others 
are  excluded.  Hence,  the  cultivation  being  limited,  perfec- 
tion in  the  culture  of  a  few,  necessarily  reaches  a  better  and 
higher  grade  of  perfection,  than  if  the  attention  of  the  planter 
was  divided  among  many.  Profit  depends,  in  a  great  degree, 
upon  the  adaptedness  of  climate  to  a  particular  crop.  The 
difference  arising  from  the  cultivation  of  a  variety  of  cotton, 
which  is  perfectly  matured  in  this  climate,  and  one  that  does 
not  attain  perfectly  that  perfection,  except  under  the  most 
favorable  circumstances,  is  very  great  in  the  long  run.  The 
rearing  of  cattle  is  much  more  profitable  where  they  are  at 
home,  than  where  they  require  much  attention  and  care  to 
make  them  thrifty. 

The  cereals  have  the  widest  range,  while  plants  of  little 
value  to  man  are  often  very  restricted  in  their  ranges.  We 
recognize  in  this  important  fact,  a  prospective  provision  de- 
signed expressly  for  the  benefit  of  man. 

If  the  foregoing  remarks  are  true,  the  education  which  ag- 
riculture demands,  in  order  to  improve  its  condition,  requires 
that  of  the  highest  grade.  Agriculture,  while  it  is  not  to  lose 
its  place  as  an  art,  must,  in  order  to  advance,  demand  of  its 
cultivators  more  knowledge  of  the  collaterals.  Some  call  this 
mere  book  learning  which  is  of  no  account  in  practice ;  arid 
in  support  of  this  view,  say  that  agriculture  has  got  along 
very  well  without  them.  Indeed  none  of  our  fathers  had  the 
benefit  of  the  collateral  or  direct  lights ;  and  yet  they  made 
money  by  their  simple  modes  of  culture.  This  is  no  doubt 
true.  The  planters  of  North- Carolina  found  a  rich  virgin 
soil.  The  crops  of  maize  required  but  little  attention.  Cot- 


NORTH-CAROLINA   GEOLOGICAL   SURVEY.  7 

ton  at  a  later  day  became  a  profitable  staple,  its  importance 
increased  with  the  return  of  every  year.  But  what  have 
been  the  results  upon  the  soil  from  the  midland  counties 
of  North-Carolina  to  Alabama  ?  Let  one  pass  along  the  rail- 
road from  Kaleigh  to  Columbia,  and  then  through  Georgia 
to  Montgomery.  The  exhaustion  of  the  soil  by  its  culture  is 
too  palpable  and  plain  to  be  overlooked.  Exhaustion  on  the 
whole  route  is  the  prominent  feature.  It  took  place  slowly 
but  surely.  What  were  rich  lands  under  the  simple  culture 
of  the  fathers,  have  now  become  the  poor  and  worn  out  lands 
of  their  sons.  It  is  at  this  stage  that  education  or  knowledge  is 
demanded.  The  fathers  got  along  very  well,  and  made 
money;  but  the  sons,  though  they  may  inherit  money  al- 
ready made,  must  be  content  with  that,  or  move  away,  or 
else  seek  by  superior  knowledge  to  replenish  the  worn  out 
inheritance.  New  modes  of  culture  must  be  devised,  and  a 
much  greater  amount  of  knowledge  and  skill  will  be  required 
than  the  fathers  possessed. 


8  NORTH-OAKOLINA  GEOLOGICAL  SIJKVEY. 

. 

CHAPTEE  I. 


Reference  to  a  former  report.  The  perfection  of  seed  depends  on  the  char- 
acter of  the  soil.  Nutrient  matters  necessary  to  animal  life  traced  to  the 
soil.  Essential  elements  of  a  good.  The  soil  the  reservoir  of  all  these 
elements.  Character  and  classification  of  the  soils  in  the  Eastern  coun- 
ties. Importance  of  determining  the  smallest  per  centage  of  earthy 
matter  in  a  vegetable  soil,  which  is  compatible  with  a  remunerating  crop. 
Some  elements  are  more  essential  to  form  a  good  soil  than  others.  The 
organs  of  a  plant  are  composed  of  different  elements.  The  extreme  of 
certain  kinds  of  soil.  Remarks  on  the  adaption  of  soils,  together  with 
a  statement  of  their  composition.  Soil  of  the  open  ground  prairie  in 
Carteret  county.  Pocosin  and  swamp  lands.  Soils  of  Hyde  county. 

§  5.  In  a  former  report,  that  of  1852,  I  deemed  it  neces- 
sary to  point  out  certain  facts  which  have  a  direct  bearing 
upon  the  principles  of  agriculture,  and  which  indeed  appear 
to  constitute  the  foundation  upon  which  it  is  based ;  and  as 
the  present  report  may  fall  into  the  hands  of  those  who  may 
not  have  seriously  reflected  upon ,  those  principles,  I  now  pro- 
pose to  recapitulate  them  very  briefly. 

Soils  must  contain  a  sufficiency  of  certain  inorganic  ele- 
ments, otherwise  no  seed  can  be  perfected.  The  elements 
which  support  animal  life  may  be  traced  to  those  which  exist 
in  the  vegetable,  especially  the  seed  and  fruit.  Hence,  the 
important  products  of  life  are  derived  from  the  soil,  it  being 
possible  to  trace  them  back  through  the  vegetable,  and 
the  reverse,  from  the  soil  through  the  vegetable  to  the  animal. 
Those  products  of  life  then,  which  can  be  traced  to  no  other 
source  than  the  soil,  must  be  regarded  as  essential  elements 
of  the  soil,  and  as  designed  to  sustain  and  support  life.  The 
office  of  the  vegetable  tissue  through  which  they  pass  to  fit 
them  for  sustaining  animal  life,  are  to  simply  modify,  or  to 
form  new  combinations,  and  not  new  substances  or  elements. 

Those  which  I  regard  as  essential  to  animal  life,  and  all 
of  which  exist  in  the  soil,  are,  phosphorus,  sulphur,  potash, 
soda,  lime,  magnesia,  iron,  silica,  nitrogen,  oxygen,  hydrogen 


NORTH-CAROLINA  GEOLOGICAL   SURVEY.  9 

and  carbon.  They  do  not  seem,  in  any  instance,  to  enter 
into  the  composition  of  living  bodies  in  the  elementary  state, 
but  as  compounds ;  thus  hydrogen  combines  with  oxygen 
and  forms  water,  or  nitrogen  and  forms  ammonia ;  oxygen 
combines  with  phosphorus,  sulphur,  etc.,  before  they  are 
fitted  to  enter  into  the  composition  of  the  animal  tissue. 

The  soil  then,  being  the  great  reservoir  or  source  of  these 
elements  which  are  truly  essential  to  life,  and  so  far  as  nutri- 
ment is  concerned  are  dependent  upon  them,  we  cannot  over- 
estimate the  importance  of  preserving  it  in  the  best  condi- 
tion ;  and  when  the  soil  is  so  far  deprived  of  these  elements 
that  the  crops  are  imperfect,  we  see  the  importance  of  those 
fertilizers  which  contain  them.  It  appears  also,  that  sub- 
stances which  do  not  contain  them,  have  never  been  denomi- 
nated fertilizers  at  all.  Hence,  when  matters  are  added  to 
soils,  it  is  expected  that  they  contain  more  or  less  of  phos- 
phorus, sulphur,  potash,  soda,  etc.,  in  certain  states  of  combi- 
nations which  the  plant  is  able  to  obtain. 

§  6.  The  soils  of  North-Carolina  are  remarkable.  They 
belong  very  frequently  to  the  extremes  of  certain  well  dis- 
tinguished classes.  On  the  one  hand,  these  extremes  consist 
of  sand,  a  marine  product,  nearly  pure,  or  with  only  a  trace 
of  other  matters ;  on  the  other,  they  are  composed  of  nearly 
pure  vegetable  matter,  with  only  a  trace  of  earth  or  soil 
proper.  These  are  not  simply  rare  exceptions  to  the  common 
run  of  soils,  but  they  form  classes.  So  also  the  stiff  clays 
which  are  also  marine  deposits,  form  another  class.  These, 
however,  do  not  materially  differ  in  composition  from  the 
same  class  in  other  sections  of  the  State. 

The  two  former,  I  believe,  are  sectional,  and  are  confined 
to  the  lower  counties. 

Besides  the  foregoing,  where  rocks  exist  near  or  at  the  sur- 
face, we  may  clearly  recognize  other  classes  which  differ,  both 
as  to  their  origin  and  composition.  For  example,  we  may 
readily  distinguish  from  all  others  the  deep  red  soil  of  the 
argillaceous  slates  from  that  of  gneiss  or  granite,  though  the 
latter  has  a  deep  red  color  also,  or,  from  the  deep  red  soil  of 
the  sandstone  of  Orange,  Chatham,  Moore  and  Anson.  There 


10 


NORTH-CAROLINA  GEOLOGICAL   SURVEY. 


is  also  another  peculiar  soil  which  skirts  the  northern  counties, 
Granville,  Person,  Caswell  and  Rockirigham.  It  is  adapted 
to  the  growth  of  fine  tobacco.  It  is  a  light  gray  soil. 

The  soils,  however,  which  form  the  subject  of  this  report, 
occupy  the  eastern  counties  of  the  State,  and  may  all  be  re- 
garded as  marine  products  with  one  exception,  the  vegetable 
eoils,  which  occupy  the  swamps  and  pocosins  of  the  extreme 
eastern  part  of  the  State.  The  others  which  have  been  re- 
ferred to  are  derived  immediately  from  the  rocks  upon  which 
they  rest,  and  have  been  formed  by  atmospheric  agencies. 

The  vegetable  soils,  on  the  other  hand,  were  formed  by  the 
growth  of  vegetables  which  have  long  since  ceased  to  live, 
and  have  undergone  disintegration  in  a  greater  or  less  degree ; 
some  are  coarse  and  fibrous,  others  exist  as  a  close  compact 
mass  of  vegetable  matter,  perfectly  disorganized  and  in  the 
best  condition  possible  for  cultivation.  The  mass  remains  in 
situ,  frequently  homogeneous,  and  may  be  cut  into  blocks 
and  dried  like  brick. 

I  have  applied  to  these  vegetable  accumulations  the  usual 
term  soil,  for  the  reason  that  they  are  cultivated  and  frequent- 
ly productive.  Others  probably  come  more  properly  under 
the  common  name  peat,  as  the  mixed  earthy  matter  is  too 
small  to  be  cultivated  without  the  addition  of  earthy  matter, 
and  have  remained  in  situ,  and  undisturbed  since  their  seeds 
took  root. 

The  peculiarity  of  this  vegetable  soil  then  consists  in  ita 
composition,  and  the  interest  which  is  especially  attached  to 
it  arises  from  the  small  amount  of  earthy  matter  which  it 
contains.  It  gives  us,  therefore,  an  opportunity  to  determine 
the  smallest  amount  of  earthy  matter  compatible  with  re- 
munerating crops.  It  is  also  proved  by  observation  that  all 
crops  require  earthy  matter, — it  may  be  comparatively  small, 
but  if  the  inorganic  matter  is  reduced  to  a  certain  small  per- 
centage, the  crop  fails,  although  it  is  placed,  in  one  sense,  in 
a  magazine  of  food.  The  determination  of  the  smallest  per- 
centage of  inorganic  matter  which  is  compatible  with  a  good 
crop,  is  practically  important.  Large  tracts  of  land  in  North 
Carolina  consist  of  organic  matter,  with  too  little  soil  to  permit 


NORTH-CAROLINA   GEOLOGICAL    SURVEY. 


u 


of  its  cultivation.  If  inorganic  matter  is  added,  it  will  make  it 
productive,  and  possibly  valuable.  But  how  little  is  required, 
how  much  expense  may  be  required  to  bring  it  to  or  put  it  in 
a  cultivable  state,  is  a  legitimate  inquiry,  and  one  which 
may  be  productive  of  considerable  profit.  It  is  evident, 
however,  that  in  a  country  like  this,  where  there  are  vast 
areas  of  wild  land  to  be  subdued,  that  these  lands  under  con- 
sideration cannot  come  in  competition  with  good  soil  at 
government  prices,  unless  it  can  be  shown  that  the  expense 
of  reclaiming  them  is  comparatively  small ;  still,  the  question 
sought  to  be  determined  is  an  interesting  one,  and  I  have  at- 
tempted its  solution,  the  results  of  which  will  be  given  in  the 
subsequent  pages. 

§  7.  A  secondary  fact  requires  a  passing  notice.  Whilo 
all  the  elements  enumerated  are  essential  to  a  good  soil,  some 
are  more  so  than  others.  Thus,  certain  plants  require  potash, 
while  to  others  this  element  is  not  so  essential,  or  it  holds 
only  a  subordinate  place.  In  wheat  it  is  very  necessary, 
while  to  clover  it  is  less  so,  and  in  the  latter  lime  seems  to 
take  its  place.  As  a  general  law  the  most  expensive  elements, 
as  potash  and  phosphoric  acid,  abound  in  the  seed  and  fruit, 
while  lime  is  most  usually  found  in  the  wood  and  bark  or  stem. 

Silex  in  the  cereals  is  an  essential  element  in  the  stem  or 
stalk.  Its  office  is  to  give  it  strength  and  hardness. 

Each  element,  therefore,  being  destined  for  a  particular 
organ,  performs  or  fulfils  a  certain  office  or  function. 

These  specializations  wre  may  regard  as  predetermined  re- 
sults, effected  through  the  instrumentality  of  the  cell  force ; 
but  how,  it  is  impossible  to  say ;  how  the  salts  or  compounds 
of  phosphoric  acid  are  carried  up  to  form  the  seed  and  there 
remain  and  accumulate,  and  how  the  silex  is  arrested  and  ac- 
cumulates in  the  stem,  it  is  impossible  to  say. 

We  may  be  assured,  however,  that  the  machinery  of  a 
plant  will  work  right  if  it  is  fed  with  the  necessary  food. 
Knowing,  therefore,  what  a  plant  wants,  it  becomes  the 
special  business  of  the  farmer  to  supply  it.  The  perfection 
in  agriculture  will  consist  in  a  strict  application  of  the  doc- 
trine of  specialities,  and  this  specialization  will  not  be  confined 


12  NORTH-CAROLINA   GEOLOGICAL   SURVEY. 

to  a  supply  of  food  simply,  but  will  extend  to  the  mechanical 
cultivation :  each  plant  will  no  doubt  be  found  to  do  or  grow 
better  under  a  certain  mode  of  cultivation. 

§  8.  Sandy  soils  predominate  to  a  great  extent  over  all 
others  in  the  eastern  counties,  though  there  are  tracts  in 
which  clay  is  in  as  great  excess  as  sand.  The  extreme  varie- 
ties may  be  summed  up  as  follows :  1st,  sandy  soil  to  an  excess 
which  destroys  cohesion  and  becomes  blowing  sand ;  2d,  clay ; 
3d,  vegetable  soils  to  such  an  extent  as  to  exclude  earthy 
matter,  or  to  contain  merely  some  4  or  5  per  cent,  of  it. 

Between  the  extremes,  as  enumerated,  there  exist  mixtures 
in  various  proportions,  as  usual,  except  that,  as  a  general 
rule,  the  proportion  of  sand  is  somewhat  greater  than  in  the 
soils  belonging  to  other  parts  of  the  State. 

As  an  example  of  soil  in  which  sand  is  in  greater  excess,  I 
may  state  that  the  following  is  an  instance  worthy  of  note. 
The  specimen  was  taken  from  Bladen  county,  near  Elizabeth- 
town,  and  represents  a  kind  common  to  that  section.  Thus, 

Silex,    94.80 

Water,  1.20 

Organic   Matter,  1.50 

Per  oxide  of  iron  and  alumina,   65 

Lime,  01 

Magneisa,    trace, 

Potash  and  soda, traces. 

The  essential  constituents  of  a  good  soil  in  this  example 
exist  only  in  the  smallest  proportions, — and  though  it  pro- 
duces plants,  yet  the  valuable  elements  exist  in  too  small 
proportions  to  pay  for  tillage. 

The  great  excess  of  sand  is,  however,  palpable,  and  it  is 
also  evident  that  there  is  a  great  deficiency  of  clay  or  alu- 
mina, which  gives  consistency  to  soils,  and  which  forms  the 
basis  upon  which  fertilizers  may  be  profitably  applied. 

It  belongs,  it  will  be  conceded,  to  a  particular  class,  as  there 
is  a  single  element  in  great  excess.  Although  there  is  a  great 
excess  of  sand  in  these  examples,  to  which  many  more  might 
be  added,  still,  this  excess,  in  itself  considered,  does  not  dis- 
qualify them  for  the  growth  of  certain  crops,  particularly  the 


NORTH-CAKOLINA  GEOLOGICAL  SURVEY. 


13 


ground  pea,  though  it  is  possible  their  constitution  may  not 
be  fully  adapted  to  that  crop,  yet  so  far  as  the  proportion  of 
sand  is  concerned  it  is  not  in  excess.  This  fact  is  stated  for 
the  purpose  of  alluding  to  what  may  not  be  known  to  many, 
that  a  soil  which  is  really  poor  and  unsuitable  for  one  crop, 
may  be  well  suited  to  another.  The  quality  of  the  crop  may 
be  much  better  when  grown  upon  a  soil  where  sand  is  in 
great  excess  than  upon  a  rich  and  well  proportioned  soil. 

§  9.  The  contrast  between  soils,  one  of  which  is  not  well 
proportioned,  while  the  other  is,  is  strikingly  exemplified  in 
the  composition  of  another  soil  from  Halifax  county.  Thus, 
I  found: 


Silex,    74.80 

Water,  21.90 

Organic  matter,  5.40 

Alumina  and  per  oxide  of  iron, 14.00 

Phosphoric  acid,  01 

Lime,   40 

Magneisa,   20 

Potash, 05 

Soda,  03 

Another  from  Halifax  county  resembles  very  closely  the 
former ;  thus,  I  found  on  submitting  it  to  analysis : 

Silex,  94.15 

Water,  1.30 

Organic  matter, * 1.35 

Oxide  iron  and  alumina,  1.80 

Lime,    15 

Magneisa,    01 

Potash, 01 

Soda,  01 

Another  soil  from  Halifax  which  had  been  long  under  cul- 
tivation, but  whose  composition  is  somewhat  better;  thus,  it 
contained: 


Silex, 92.56 

Water,  1.20 

Organic  matter, 2.70 

Oxide  iron  and  alumina,  2.70 

Lime,  18 


14  NORTH-CAROLINA   GEOLOGICAL   SURVEY. 

Magneisa,   24 

Soluble  Silica, 10 

Potash,     trace, 

Soda, 18 

. 

The  presence  of  phosphoric  acid  was  not  determined  in 
either  of  the  foregoing,  but  as  it  is  in  combination  with  the 
small  per  centages  of  oxide  of  iron  and  alumina,  it  is  evident 
that  it  exists  in  proportions  less  than  that  of  the  alkalie.s. 

The  soils  of  Halifax,  were  originally  sandy,  yet  the  rela 
live  proportion  of  sand,  as  they  are  now  constituted,  is  con- 
siderably greater  than  when  they  were  first  brought  under 
cultivation.  The  soluble  matters,  those  consumed  by  the 
crops  which  they  have  borne,  having  been  removed  with 
them,  and  nothing  returned  to  supply  their  places,  they  are 
yet  capable  of  bearing  very  light  crops,  but  it  is  doubtful 
whether  the  cultivation  of  land  so  poor  as  these  really  pays. 
If  an  example  of  poor  soil  is  placed  side  by  side  with  a  good 
one,  the  comparison  is  much  facilitated : 


GOOD  BOTI*  POOR  SOIL. 

Silex, 74.80  94.15 

Water,    4.90  1.30 

Organic  matter,   5.40  1.35 

Alumina  and  per  oxide  of  iron,   14.00  1.80 

Phosphoric  acid, ,   . .  51 

Lime, 40  15 

Magnesia,    20  01 

Potash,  25  01 

Soda,  13  01 

In  making  a  safe  comparison  between  the  composition  of 
good  and  poor  soils,  it  should  be  stated  that  less  alumina  and 
iron  would  not  displace  the  soil  from  the  position  I  have 
placed  it.  The  silex  is  in  the  proper  proportion,  and  the  or- 
ganic matter  may  be  regarded  also  as  sufficient,  though  as 
we  shall  see  in  the  sequel,  this  element  may  be  greatly  in- 
creased to  the  advantage  of  long  cultivation.  "Where  it  is 
wholly  absent,  seed  fails  to  ripen ;  a  fact  which  shows  the 
necessity  of  its  presence.  Silex  is  the  basis  of  all  soils,  and 
where  it  is  entirely  absent,  barrenness  is  certain.  It  is  sola- 


NOKTH-CAEOLINA   GEOLOGICAL   SUHVET. 

ble  under  needful  conditions,  and  it  enters  largely  into  the 
straw  of  all  cereals. 

Alumina  never  enters  into  the  composition  of  plants  at  all ; 
but  it  performs  an  important  function  notwithstanding;  it 
holds  as  it  were  the  particles  of  earth  together.  Its  true  of- 
fice may  undoubtedly  be  shown  by  experiment.  Pour  water 
upon  a  soil  well  charged  with  clay,  and  it  remains  upon  the 
surface  ;  but  poured  upon  sand,  it  quickly  disappears.  If 
the  water  was  charged  with  fertilizing  matter,  this  also  will 
remain,  and  be  held  near  the  surface  by  the  clay,  and  within 
reach  of  the  roots  of  the  plant. 

§  10.  The  fact  is  well  known  that  sandy  soils  do  not  retain 
manures ;  while  on  the  contrary,  clay  soils  retain  all  fertiliz- 
ing matters  with  great  force.  Clay  indeed  absorbs  ammonia 
under  all  circumstances,  and  it  cannot  be  entirely  dissipated 
or  driven  oif  short  of  a  red  heat.  It  obstinately  retains  water. 
Some  of  the  functions  of  clay  are  performed  by  other  ele- 
ments. Lime  and  iron  and  organic  matter,  for  example,  give 
cohesion  to  soils,  and  aid  in  the  retention  of  water. 

"Water  exists  in  soils  in  two  conditions.  In  the  first,  it 
seems  to  adhere  to  the  surfaces  of  particles,  and  hence  is 
liable  to  constant  variation.  This  is  hygrometric  water.  In 
the  second,  it  forms  a  constituent  part  of  the  salts  in  the  soil, 
as  the  soluble  salts  of  lime  and  alkalies,  the  crenates,  etc.  In 
the  first  instance,  it  is  mostly  dissipated  by  an  exposure  of 
400  degrees  of  Fah.,  while  a  heat  near  to  redness  is  required 
to  remove  it  from  the  organic  salts. 

All  the  elements  which  have  been  enumerated,  except  alu- 
mina, enter  into  the  constitution  of  plants ;  but  as  I  have  had 
occasion  to  say,  in  different  proportions  in  different  plants, 
and  also  in  different  proportions  in  the  parts  of  plants. 

An  example  or  two  of  soils  occupying  another  extreme* 
where  the  organic  matter  is  in  great  excess,  may  be  cited 
from  localities  in  Tyrrel  and  Carteret  counties.  In  the  for- 
mer county,  large  tracts  lying  upon  Croatan  Sound,  furnish 
organic  matter  in  great  excess,  and  at  the  same  time  they  are 
deficient  in  the  earths.  Thus  in  an  uncultivated  soil  I  found 
it  composed  of 


NOKTH-CAKOLINA  GEOLOGICAL  SURVEY. 

Organic  matter, 92.70 

Sand,   6.02 

Lime,  0.02 

Phosphate  of  lime,  alumina  and  iron, 0.90 

Potash, 0.20 

Soda,   0.06 

Magnesia,   trace. 

The  silex  in  this  case  is  a  whi^te  marine  sand  which  becomes 
visible  after  rains,  or  after  a  year  or  two  of  cultivation.  It 
is  too  coarse  to  furnish  the  necessary  amount  of  soluble  silica 
for  a  succession  of  crops.  When  the  vegetable  matter  is  re- 
moved, it  remains  as  a  white  sand  still,  and  is  blown  into 
ridges. 

§  11.  The  condition  of  the  vegetable  matters,  as  in  the 
case  of  the  other  elements,  is  quite  variable.  Sometimes  it 
is  very  fine,  and  is  thoroughly  incorporated  with  them ;  in 
other  instances  it  is  coarse,  or  in  the  condition  of  fibres.  In 
the  former  state  the  sand  is  not  so  readily  exposed ;  in  the 
latter  it  is  always  visible,  and  is  indicative  of  a  poor  condi- 
tion, or  of  its  unsuitableness  for  cultivation.  It  has  not  been 
exposed  long  enough  to  change  it  to  the  condition  required 
for  crops  of  the  most  valuable  kind. 

A  still  more  remarkable  case  of  excess  of  vegetable  matter 
composes  a  tract  in  Carteret  county,  and  is  known  as  the 
open  ground  prairie.  This  tract,  or  that  portion  of  it  lying 
within  a  certain  zone  of  rich  and  productive  land,  contains  a 
growth  of  sphagnum  or  moss,  together  with  other  vegetables 
intermixed,  with  which  there  is  only  a  minute  quantity  of 
earth.  I  obtained  it  from  a  depth  of  18  inches,  and  it  gave 
only  3  per  cent,  of  inorganic  matter,  and  this  was  mostly  the 
ash  of  the  vegetable  fibre.  This  case  furnishes  an  example 
of  an  unproductive  soil,  so  far  as  the  grains  are  concerned. 
The  outer  rim  of  the  open  grounds  is  an  excellent  soil. 

Much  has  been  said  respecting  the  open  ground  prairie,  and 
enquiries  are  now  frequently  made  respecting  the  character 
of  this  tract ;  and  whether  it  is  susceptible  of  a  profitable 
cultivation.  As  the  soil  is  now  constituted,  a  kernel  of  corn 
planted  in  it  would  germinate  and  grow  well  apparently  until 


NORTH-CAEOLINA   GEOLOGICAL   SURVEY. 


it  is  about  one  foot  high,  when  it  turns  yellow  and  dies.     It 
is  then  evidently  in  an  uncultivated  condition. 

The  question  then  comes  up,  can  the  open  prairie  be  made 
cultivable  artificially,  and  if  so,  how  ?  The  question  first  put 
is  not  designed  to  inquire  strictly  into  the  possibility  of  the 
tiling,  because  all  who  have  given  some  thought  to  the  ques- 
tion, know  very  well  that  it  is  possible,  because  a  soil  can  be 
made  from  the  start,  by  putting  together  the  proper  elements, 
and  this  can  be  done  with  the  open  ground  prairie ;  but  can 
it  be  done  profitably  ?  Now,  when  we  are  assured  that  the 
soil  of  the  open  prairie  ground  is  composed  exclusively  of  vege- 
table matter,  it  is  plain,  that  the  earths  must  be  added  to  give 
it  the  composition  required  for  the  perfection  of  vegetables 
of  any  value  to  man.  The  old  practice  consisted  mainly,  in 
giving  peaty  soils  (as  this  must  be  ranked  in  that  class,)  a 
heavy  dressing  of  lime.  It  is  evident  on  reflection,  if  the 
principles  in  the  foregoing  paragraphs  are  correct,  that  this 
practice  could  not  be  relied  upon,  for  it  would  only  acquire 
a  single  element.  Something  more  is  wanted.  Not  only 
lime,  but  iron,  alumina  and  silica  are  required.  We  may 
infer  that  the  phosphates  and  alkalies  will  be  supplied  by  the 
decay  of  vegetable  matter,  and,  from  this  fact,  ii;  appears 
at  least  plausible,  that  the  treatment  which  the  open  ground 
prairie  demands,  is  the  addition  of  some  natural  soil.  It  may 
be  taken  from  the  nearest  marsh  where  mud  or  soil  may  be 
obtained,  provided  it  contains  silex,  alumina,  iron,  etc. 

Knowing,  then,  what  substances  are  wanting  in  this  soil. 
and  hence  what  must  be  added,  the  question  resolves  itself 
into  this :  how  much  does  a  soil  of  the  description  of  that 
under  consideration  require  to  make  it  productive?  We 
have  seen  that  the  soil  upon  Croatan  sound  is  at  least  tolera- 
bly productive,  which  contains  only  7.30  per  cent,  of  inor- 
ganic matter,  and  that  the  element  which  greatly  predomi- 
nates over  the  rest,  is  sand,  in  a  state  unfitted  to  furnish  solu- 
ble silica.  We  may  regard  the  Croatan  soil  as  containing 
the  smallest  quantity  of  earthy  matter,  and  at  the  same  time 
possessing  the  ability  to  grow  the  cereals.  Leaving  the  sand 
out  of  view,  we  may  infer  that  the  least  quantity  of  earth  which 
3 


NORTH-CAROLINA  GEOLOGICAL   SURVEY. 

is  required  to  the  open  ground  prairie  will  be  not  less  than 
140  to  150  tons  to  the  acre.  When  this  expense  is  added  to 
the  expense  of  drainage,  it  is  evident  that  in  a  country  where 
land  is  cheap  it  would  not  be  economical  to  expend  so  much 
money  and  labor  to  create  as  it  were  a  soil  adapted  to  the 
better  class  of  vegetables. 

§  12.  The  effect  of  cultivation  of  soils  composed  mainly  of 
vegetable  matter  and  marine  sand,  is  to  consume  so  much  of 
the  former  that  the  latter  becomes  in  its  turn  predominant, 
and  even  after  a  few  years'  cultivation  only,  the  white  sand 
shows  itself  through  and  upon  the  surface  of  the  black  vege- 
table matter,  and  soon  afterwards  it  appears  in  sufficient 
quantities  to  form  white  ridges  over  the  cultivated  field. 
When  this  takes  place,  the  soil  has  already  begun  to  exhibit 
unmistakable  evidences  of  partial  exhaustion. 

The  soils  in  which  vegetable  matter  predominates,  apper- 
ently  in  great  excess,  not  injuriously  however,  prevail  orer 
large  tracts  or  areas  in  the  eastern  counties,  and  are  beginning 
to  be  esteemed  the  most  valuable  lands  of  any  in  North- 
Carolina.  They  are  not  confined  to  one  or  two  counties,  but 
may  be  found  in  most  of  them  which  lie  east  of  the  Wil- 
mington railroad.  They  also  prevail  in  the  south-eastern 
section,  especially  in  New  Hanover  and  Columbus. 

Some  of  the  tracts  are  classed  as  pocosin  and  swamp  lands, 
but  they  agree  in  having  a  very  large  percentage  of  veget- 
able matter,  and  in  being  also  thoroughly  wet  and  frequently 
covered  with  water,  I  have  found  that  there  is  no  constant 
percentage  of  vegetable  matter  where  different  and  distant 
tracts  are  compound  together.  It  is  as  variable  as  the  clay 
or  sand  in  argillaceous  and  sandy  soils.  There  is  also  a 
variableness  as  to  its  condition;  it  is  often  perfectly  disorgan- 
ized and  presents  a  compact  appearance  when  cut  into  blocks ; 
or  it  may  be  in  the  condition  of  coarse  fibres  with  their  tex- 
ture or  structure  perfectly  preserved.  In  the  first  case,  it  is 
in  the  proper  condition  for  cultivation,  and  the  latter,  it  has? 
not  passed  into  that  state  and  condition  which  is  fitted  for 
the  nutrition  of  the  cereals.  The  coarse  vegetable  fibre  pre- 
dominates in  the  open  prairie  grounds  of  Carteret,  and  the 


NORTH-CAROLINA  GEOLOGICAL   SUKVET.  19 

former' in  those  of  Hyde  and  Ooslow  counties.  So  also  these 
vegetable  soils  vary  endlessly  with  respect  to  the  amount  of 
soil  and  sand.  The  Hyde  county  soils  may  be  regarded  as 
the  standard  eoils  for  excellence  of  this  class,  and  hence  it  is 
important  to  determine  their  composition.  On  their  own  ac- 
count, it  is  important  to  determine  the  composition,  as  well  as 
for  the  purpose  of  comparing  their  composition  with  others 
which  resemble  them  in  their  external  characters.  Many 
mistakes  have  been  made  in  the  swamp  lands;  for  when  wet 
and  examined  in  the  ordinary  way  they  look  rich — with 
the  presence  of  a  superabundance  of  vegetable  matter,  their 
true  characters  may  be  concealed.  In  many  cases  the  con- 
dition of  the  earthy  matter  is  overlooked.  It  may  indeed 
be  too  small;  or  it  may  be  a  coarsish  marine  sand  destitute 
of  fine  earth.  In  all  cases  it  is  possible,  and  indeed  easy  to 
determine  whether  it  will  be  productive  or  comparatively 
valuable.  This  is  an  important  fact  to  make  out,  for  all  these 
lands  require  to  be  drained  thoroughly,  and  it  is  certainly  an 
object  worth  attention  to  be  able  to  determine  before  hand 
whether  the  tract  is  worth  the  expenditure  before  it  is  in- 
curred. 

The  Hyde  county  soils  have  acquired  a  deservedly  high 
reputation  for  fertility.  Some  tracts  have  been  cultivated 
over  a  century,  and  the  crops  appear  to  be  equally  as  good 
as  they  were  at  an  early  period  of  their  culture ;  and  yet  no 
manure  has  been  employed,  and  they  have  been  under  cul- 
ture in  indian  corn  every  year;  or  what  would  be  equivalent 
thereto.  If  this  crop  has  been  omitted,  wheat  has  been  sub- 
stituted for  it;  not  because  they  are  properly  wheat  soils, but 
if  they  are  uncultivated,  the  weeds  acquire  a  size  that  it  is 
impossible  to  cover  them  the  next  year.  The  same  difficulty 
occurs  in  part  in  the  culture  of  corn  ;  the  stalks  are  so  numer- 
ous and  large  that  it  is  difficult  to  bury  them  so  completely 
that  they  shall  be  concealed,  and  preserve  at  the  same  time 
an  even  handsome  surface.  For  this  reason  critics  of  a  mor- 
bid ivtamp  have  said,  that  the  Hyde  county  planters  are 
slovenly,  overlooking  the  facts  refered  to,  which  are  really 
the  sole  causes  of  the  defects  complained  of.  Though  the 


20  NORTH-CAROLINA   GEOLOGICAL   SURVEY. 

defects  are  not  very  palpable  under  any  circumstances,  still 
it  is  sometimes  useful  to  a  community  to  have  faultfinders, 
and  to  have  their  doings  overhauled  by  a  would  be  wise  critic. 
§  13.  Hyde  county  appears  to  be  nearly  a  dead  level.  It 
rises  of  course  a  few  feet  above  the  sound,  but  it  is  impercep- 
tible to  the  eye.  Buildings  may  be  seen  for  great  distances, 
and  were  the  whole  surface  laid  out  in  proper  order,  it  might 
be  made  to  appear  like  an  immense  park.  The  depressions 
of  the  surface  are  due  to  fires  which  have  consumed  the 
vegetable  matters  to  the  depth  of  from  four  to  ten  and  per- 
haps fifteen  feet.  In  these  depressions  the  surface  water  has 
accumulated,  and  in  a  few  instances  large  lakes  are  the  re- 
sult. Mattamuskeet  lake  is  the  largest  of  the  surface  drain- 
age. Its  former  extent  was  not  less  than  twenty  miles.  Its 
circumference  now  exceeds  sixty  miles  by  the  road, — and  as 
the  traveller  proceeds  on  his  route,  there  is  nothing  more  sur- 
prising than  the  succession  of  corn  fields  which  are  always  in 
view. 

The  most  common  natural  growth  of  the  best  swamp  land 
of  Hyde  county  is  cypress  and  black  gum. 

In  one  respect  this  region  differs  from  others  farther  from 
the  sea.  There  is  no  difficulty  in  the  cultivation  of  the  gras- 
ses. It  is  evident  the  climate  is  more  humid,  and  the  sea 
breezes  moderate  the  heat  sufficiently  in  summer  to  favor  the 
developement  of  this  family  of  plants.  There  is  no  doubt, 
also,  that  if  the  attention  of  the  planters  was  turned  to  the 
cultivation  of  grasses  adapted  to  the  climate,  greater  profits 
might  be  realized  than  from  the  cultivation  of  maize.  It  is 
less  expensive,  and  as  hay  bears  a  high  price,  and  is  obtained 
from  a  distance,  in  all  the  villages  of  this  part  of  the  State, 
and  as  there  is  always  -a  communication  with  them  by  water, 
there  can  be  no  doubt  that  the  profits  which  would  arise  from 
hay  making,  would  considerably  exceed  those  of  corn.  The 
green  surface  of  the  lake  shore,  the  yards  of  the  houses,  and 
the  appearance  of  the  small  pasturages  sustain  this  view. 

§  14.  The  peculiarities  of  the  soil  of  Hyde  county,  that 
particularly  of  the  lake  region,  are  comprised  in  two  particu- 
lars: 1st,  the  large  quantity  of  fine  vegetable  matter  they 


NORTH-CAROLINA   GEOLOGICAL   SURVEY. 


21 


contain;  2d,  the  extreme  fineness  of  the  intermixed  earthy 
matter.  The  earthy  matter  is  invisible  in  consequence  of  its 
fineness,  and  is  evenly  distributed  through  the  mass.  An 
inspection  of  it  even  under  a  common  lens  will  deceive  most 
persons,  and  they  would  be  led  to  infer  that  it  was  entirely 
absent.  Unlike  other  soils  it  contains  no  coarse  visible  par- 
ticles of  sand ;  and  hence  it  appears  that  during  the  growth 
of  the  vegitables  which  form  at  least  one-half  of  the  soil,  it 
was  subjected  to  frequent  overflows  of  muddy  water;  or  else 
the  area  over  which  these  peculiar  soils  prevail  was  usually  a 
miry  swamp  which  communicated  with  streams  which  brought 
over  it  the  finest  sediment  of  some  distant  region.  This  sedi- 
ment is  frequently  a  fine  grit,  and  fine  enough  for  hones,  and 
w-hen  the  vegetable  matter  is  burnt  off,  it  assumes  a  light 
drab  color.  The  character  of  the  Hyde  county  soils  has 
never  been  understood.  The  cause  of  their  fertility  has  never 
been  explained,  and  many  persons  who  are  good  judges  of 
land  have  overated  the  value  of  swamp  lands  in  consequence 
of  the  close  external  resemblance  they  have  borne  to  those 
of  Hyde.  Analysis,  however,  will  in  every  case  detect  the 
difference  between  the  common  swamp  soils,  and  those  of 
Matamuskeet  lake. 

It  is  unnecessary  to  dwell  farther  upon  the  points  I  have 
stated  respecting  the  characteristics  of  these  remarkable  soils. 
It  will  appear  in  the  sequel  that  there  is  a  great  uniformity 
in  the  composition  of  these  soils,  both  as  it  regards  the  amount 
and  condition  of  the  vegetable  matter,  and  the  quantity  and 
condition  of  the  fine  grit  intermixed  with  it. 

Eegarding  as  I  do  these  soils  as  the  proper  standard  for 
the  valuable  swamp  soils  of  the  eastern  section  of  the  State, 
I  have  subjected  many  samples  to  a  rigid  chemical  analysis. 

The  result  of  these  analyses  have  thrown  much  light  over 
them,  and  explains  satisfactorily  their  steady  productiveness 
for  long  periods.  It  will  appear  that  their  fertility  is  due  not 
only  to  their  vegetable  matter,  but  also  to  the  composition 
and  condition  of  the  earth  in  combination  with  it. 

Hereafter,  it  appears  to  me,  it  will  be  unnecessary -to  sub- 


22 


NORTH-CAKOLINA   GEOLOGICAL   SURVEY. 


ject  soils  of  this  character  to  a  strict  analysis,  for  reasons 
which  will  be  stated  in  the  sequel. 

In  my  journey  to  Hyde  my  principal  objects  were  to  select 
the  standard  soils  for  analysis,  and  to  investigate  upon  the 
ground,  the  peculiar  conditions  which  seemed  to  favor  the 
production  of  indian  corn ;  for  of  all  crops  this  seems  to  be 
the  one  to  which  the  soils  are  specifically  fitted. 

In  accomplishing  the  objects  of  my  visit  I  was  ably  secon- 
ded by  Dr.  Long,  of  Lake  Landing,  who  has  become  the 
owner  of  a  tract  which  has  borne  this  crop  for  one  hundred 
years  without  manures.  It  does  not  seem  to  have  deteriorated 
by  this  long  cultivation ;  or  the  crops  do  not  show  a  percepti- 
ble falling  off;  still  there  has  been  a  large  consumption  of 
materials  during  the  one  hundred  years  of  cultivation  which 
may  be  made  to  appear  by  analysis.  The  great  supply  of 
nutriment,  however,  still  holds  out,  and  the  one  hundred  years 
to  come,  if  subjected  to  no  greater  drains  upon  its  magazine 
of  food,  will,  at  such  a  distant  period,  continue  to  produce  its 
ten  or  twelve  barrels  of  corn  to  the  acre. 


CHAPTER  II. 

,v  v     j 

The  best  soil  of  Dr.  Long,  of  Hyde  county — its  composition — its  common 
yield  per  acre  of  corn.  Mr.  Burrough's  soil  of  the  north  side  of  Matta- 
muskeet  Lake.  Amount  of  inorganic  matter  which  a  crop  of  corn  re- 
moves from  the  soil.  Each  organ  to  be  furnished  with  appropriate  nu- 
triment. Maize  an  exhausting  crop.  Soils  from  the  plantation  of  Gen. 
Blount,  Beaufort  county.  Gen.  Blount's  letter,  etc. 

§  15.  The  soil  which  Dr.  Long  regarded  as  his  best,  and 
which  had  been  under  cultivation  only  three  years,  I  shall 
now  speak  of,  and  state  its  composition,  and  present  it  as  rep- 
resenting very  nearly  the  original  condition  of  the  best  soil 
of  the  county.  It  is  rather  light  and  loose,  of  a  black  color 


NORTH-CAROLINA   GEOLOGICAL   SURVEY.  23 

like  all  vegetable  soils.  It  is  not  however  spongy.  Bains  do 
not  expose  grains  of  quartz  as  in  many  instances  of  the  gall- 
berry  lands.  It  becomes  rather  lumpy  on  drying.  Its  com- 
position is  as  follows : 

Organic  matter, 48.10 

Silex: 43.00 

Oxide  of  iron  and  alumina,    6.40 

Lime,    0.21 

Magnesia,    012 

Potash 0.16 

Soda, 0.18 

Chlorine,  trace, 

Soluble  Silex, 0.03 

Sulphuric  acid, 0.04 

Phosphoric  acid,  0.30 

98.55 

The  silex,  after  the  removal  of  the  organic  matter,  is  of  a 
light  drab  color,  exceedingly  fine,  or  nearly  fine  enough  for 
sharpening  fine  edge  tools.  If  all  the  vegetable  matter  was 
removed,  this  fine  earth  would  probably  be  too  compact  and 
close  for  cultivation ;  but,  intermixed  as  it  is  with  the  debris 
of  vegetables,  it  is  sufficiently  porous  to  admit  all  the  light 
and  air  required  for  the  luxuriant  growth  of  any  crop  which 
may  be  put  upon  it. 

The  composition  of  this  soil,  it  is  evident,  shows  a  large 
proportion  of  vegetable  matter.  This  is  intimately  blended 
with  fine  earthy  matter,  the  basis  of  which  is  silex.  In  com- 
bination with  it  we  find  a  full  pioportion  of  iron  and  alumina, 
or  clay,  which  gives  coherency  to  the  grains,  and  besides  the 
nutritive  elements,  lime,  magnesia,  potash,  phosphoric  acid, 
exist  in  as  large  proportions  as  in  other  rich  and  productive 
soils.  The  regular  yield  of  this  soil  to  the  acre  is  from  ten 
to  twelve  barrels  of  Indian  corn.  In  favorable  seasons  it 
amounts  to  twelve,  in  less  favorable  it  may  reach  only  ten 
barrels.  It  is -also  easy  to  cultivate. 

The  composition  of  a  soil  of  a  similar  character,  and  which 
has  been  under  culture  by  Mr.  Burroughs,  of  the  north  side 
of  the  lake,  is  as  follows : 


NOETH- CAROLINA  GEOLOGICAL  SURVEY. 

Silex,   34.60 

Water, 12.30 

Organic  matter,   41  90 

Peroxide  of  iron,    3.70 

Alumina, 5.10 

Soluble  silica,    0.40 

Lime, 0.48 

Magnesia, 0.27 

Potash,    0.13 

Soda, 0.10    , 

Phosphoric  acid,   0.12 

This  soil,  though,  exposed  in  paper  in  a  dry  room  for  two 
months  to  the  air,  contained  more  water  than  the  preceding. 
Its  composition  should  be  calculated  without  the  water.  So 
it  is  probable  that  the  phosphoric  acid,  if  obtained  and  calcu- 
lated from  the  full  proportion  of  earthy  matter,  would  show  a 
more  striking  result.  But  it  is  evident  that  there  can  be  no 
deficiency  of  this  important  element,  inasmuch  as  the  crop  is 
one  which  is  necessarily  rich  in  phosphates.  The  depth  of 
this  rich  vegetable  soil  varies  from  5  to  10  feet,  rarely  less 
than  five  feet.  This  may  be  taken  too  as  the  usual  depth  of 
the  soils  of  this  description,  not  only  in  Hyde,  but  in  all  the 
eastern  counties  where  swamp  and  pocosin  lands  prevail. 

§  16.  There  are  but  few  instances  on  record,  where  a  soil 
has  been  under  cultivation  a  century,  and  still  retains  its  ap- 
parent original  fertility.  It  must  of  course  have  lost  a  large 
amount  of  phosphoric  acid,  potash  and  lime ;  still  the  crops 
are  equal  in  measure  to  what  they  were  when  first  cultivated. 
In  order  to  test  the  value  of  a  soil  which  had  borne  a  crop  for 
one  hundred  years,  and  during  the  whole  period  had  not  re- 
ceived a  bushel  of  manure,  I  selected  a  parcel  of  it  at  a  dis- 
tance from  buildings,  or  from  a  spot  which  could  not  Jiave 
received  any  artificial  aid. 

This  parcel  gave  the  following  result,  on  submitting  it  to 
analysis : 

Silex, 59.00 

Organic  matter, 22.20 

Peroxide  of  iron  and  alumina,  8.00 

Lime,   0.10 

Magnesia,    0.09 

Potash,   .                                           0.02 


NORTH-CAROLINA   GEOLOGICAL   SURVEY.  25 

Soda,    0.03 

Sol.  silica,   0.20 

Water,   8.90 

Phosphoric  acid,    trace. 

99.44 

These  remarks  are  justified  on  comparing  the  results  of  this 
analysis  with  Dr.  Long's  soil,  which  has  been  under  cultiva- 
tion only  three  years  ;  thus,  the  silica  is  in  greater  proportion, 
and  the  organic  matter,  less ;  and  it  is  due  no  doubt  to  the 
fact  that  it  has  been  under  cultivation  for  the  time  specified. 
It  still  retains,  however,  a  magazine  of  food  for  future  crops ; 
and  if  not  exhausted  at  a  greater  rate  than  during  the  last 
century,  it  will  be  a  rich  soil  at  the  close  of  the  next  century. 
It  will  be  perceived  that  all  the  elements  of  fertility  which 
belong  to  new  and  unexhausted  soils  still  belong  to  this.  The 
inorganic  matter  is  extremely  fine,  like  the  finest  grit,  and  in 
the  proportion  required  for  the  production  of  the  most  valua- 
ble crops.  Growing,  as  we  perceive,  in  a  magazine  of  food, 
it  seems  to  show  that  it  is  a  crop  upon  which  it  is  scarcely 
possible  to  overmanure,  and  that  it  is  unlike  other  corn  crops, 
which,  when  over  supplied  with  food,  run  to  stalks  and  leaves 
to  the  detriment  of  the  grain. 

§  17.  If  we  calculate  the  amount  of  inorganic  matter  which 
a  hundred  crops  of  maize  remove  from  the  soil,  we  should 
find  it  to  amount  to  many  thousand  pounds. 

From  data  in  my  possession,  I  am  led  to  believe  that  five 
hundred  pounds  per  acre  of  inorganic  matter  is  removed  in 
every  crop.  This  inorganic  matter  is  contained  in  the  ker- 
nels, cobs,  husks,  silks,  leaves,  sheaths,  stalks  and  tassels ; 
each  organ  containing  its  own  appropriate  amount: 

The  number  of  plants  which  are  allowed  to  grow  upon  an 
acre,  amount  to  fourteen  thousand  and  seven  hundred.  Each 
plant  removes  from  the  soil  a  specific  amount  of  the  earthy 
compounds,  and  nearly  in  the  following  proportions,  viz : 

In  Silica,    195  Ibs. 

Earthy  phosphates, 108  " 

Lirne, 25  " 

Magnesia,  18  " 


*0  NORTH-CAROLINA   GEOLOGICAL   SURVEY. 

Potash,   78  " 

Soda, 30  " 

Chlorine, 29  " 

Sulphuric  acid,   34  " 

507 

If  five*  hundred  pounds  of  the  earthy  constituents  of  this 
soil  are  removed  from  one  acre  in  one  year  or  in  a  single 
crop,  it  will  amount  in  one  hundred  years  to  fifty  thousand 
pounds — a  quantity  which  would  exhaust  most  perfectly  any 
of  the  ordinary  soils  of  the  country. 

In  an  analysis  which  I  have  made  of  the  kernels  and  cobs 
of  the  yellow  corn,  I  found : 


COBS. 

Silica, 4.67  5.93 

Earthy  phosphates,   8.22  22.18 

Lime,   0.10  0.10 

Magneisa, 0  30  1 .50 

Potash,   12.31  14.95 

Soda, 2.03  14.11 

Chlorine,    0.04  0.39 

Sulphuric  acid,  0.11  2.74 

61.81 


That  the  composition  of  the  leaves  may  be  compared  with 
the  foregoing,  I  subjoin  an  analysis  of  the  leaves  made  at  the 
same  time  and  growing  upon  the  same  plant: 

LEAVES. 

Silica, 82.88 

Earthy  phosphates, 29.27 

Lime,  9.40 

Magneisa,    1.91 

Potash,    19.70 

Soda,    13.14 

Chlorine,   15.07 

Sulphuric  acid,  6.46 

It  might  be  supposed  that  as  the  sheaths  of  the  leaves  be- 
long in  one  sense  to  the  leaves  themselves,  that  their  composi- 
tion would  be  the  same ;  but  this  is  not  the  case  as  may  be 
seen  by  the  following  analysis : 


NORTH-CAROLINA   GEOLOGICAL   SURVEY.  27 


Silica, 39.66 

Earthy  phosphates,  7.54 

Lime,  1.58 

Magnesia,   58 

Potash, 5.57 

Soda, 9.26 

Chlorine,   2.20 

Sulphuric  acid, 8.92 

In  the  sheaths  the  earthy  phosphates  and  alkalies  are  much 
less  than  in  the  leaves.  In  the  cobs  too  the  earthy  phosphates 
are  less  than  in  the  kernels ;  it  seems,  therefore,  that  each 
part  or  organ  has  its  own  peculiar  composition.  To  complete 
this  view  of  the  composition  of  the  plant  of  the  maize,  I  sub- 
join an  analysis  of  the  stalks;  thus,  they  contain: 

Silica, 8.78 

Earthy  phosphate,   10.30 

Lime,    1.92 

Magnesia,   0.64 

Potash, 11.08 

Soda,   17.09 

Chlorine,  7.42 

Sulphuric  acid, 7.38 

It  should  be  observed  that  these  several  analyses  were 
made  of  a  single  plant,  and  the  proportions  are  those  belong- 
ing to  the  plant,  or  its  parts,  and  not  properly  percentages. 
The  ash  was  obtained  from  all  the  leaves,  or  stalks,  and  kern- 
els, and  the  whole  ash  obtained  analyzed.  Hence  the  differ- 
ence of  composition  of  those  parts  are  presented  in  a  strong 
light,  as  well  as  in  a  true  proportion. 

From  the  foregoing  it  will  be  perceived  that  where  a  crop 
is  to  be  manured  or  a  fertilizer  applied,  it  is  not  sufficient  to 
apply  the  earthy  phosphates,  for  we  perceive  that  every  organ 
or  part  requires  all  the  elements  whick  we  find  in  them. 
The  notion,  therefore,  should  be  dispelled,  that  bone  earth  is 
the  main  fertilizer  for  the  maize  crop,  or  that  it  is  enough  to 
furnish  substances  which  consist  of  elements  found  in  the  grain 
or  fruit.  For  the  perfection  of  the  crop  it  is  necessary  that 
the  leaves  and  stalks,  tassel  and  cobs  should  be  furnished  with 
appropriate  elements  of  food  as  well  as  the  grain ;  for  that  the 


NOKTH-CAROLINA    GEOLOGICAL   SURVEY. 

grain  may  ripen  and  acquire  perfection,  the  leaves  and  stalks 
also  should  be  equally  perfected.  It  can  scarcely  be  doubted 
that  the  grain  itself  depends  for  its  full  development  upon  the 
perfection  of  the  parts  which  precede  it.  They  are  the  organs 
which  bring  up  the  nutriment  from  the  soil.  Remove  the 
leaves  at  an  early  day,  and  the  grain  is  destroyed,  or  never 
comes  to  maturity ;  but  supply  matter  suitable  for  their  in- 
crease and  perfection,  and  the  grain  is  supplied  also.  It  will 
be  observed  that  the  different  subordinate  parts  frequently 
contain  elements  which  are  not  found,  except  in  very  small 
proportions,  in  the  seed  or  grain  ;  yet,  there  is  no  doubt  these 
elements  are  quite  essential  to  the  perfection  of  the  plant. 

§  18.  Maize  must  be  ranked  among  the  most  exhausting 
crops ;  and  it  is  evident  that  poor  soils  will  scarcely  repay 
the  farmer  for  its  cultivation.  It  is  evident  that,  unlike  other 
cereals,  there  is  little  danger  of  using  too  much  manure  in  its 
cultivation,  as  it  will  bear  almost  any  amount  without  injury, 
provided  all  the  elements  of  fertility  exist  in  the  magazine  of 
food  provided  for  it.  It  is  not  liable  to  run  to  foilage,  and 
thereby  fail  to  produce  grain ;  neither  will  it  lodge  or  fall 
down  by  its  own  excessive  disproportion  of  organic  to  its  in- 
organic nutriment. 

While  it  must  be  admitted  that  maize  is  an  exhausting  crop, 
it  is  equally  clear  and  conclusive  that  it  is  one  of  the  most 
important  and  valuable,  and  hence  it  may  be  regarded  as  one 
which  pays  the  best. 

§  19.  The  foregoing  remarks  respecting  the  maize  crop 
have  been  made  in  consequence  of  the  peculiar  adaptation 
of  the  soil  of  Hyde  county  to  this  cereal.  It  is  the  granary 
of  the  South.  It  is  true  that  the  number  of  bushels  per  acre 
which  constitute  the  average  crop  is  less  than  the  number 
frequently  made  on  other  kinds  of  soil.  Thus  a  hundred 
bushels  of  corn  may  be  grown  upon  an  acre,  but  the  Hyde 
county  soils  rarely  exceed  sixty  bushels  per  acre, — but  from 
fifty  to  sixty  bushels  are  grown  annually  per  acre  for  an  in- 
definite term  of  years,  without  the  expense  -of  fertilizers, 
while  the  heavy  premium  crops  require  a  great  expenditure 
on  them ;  and  these  have  to  be  repeated  in  order  to  keep  the 


NOETH-CAKOLINA   GEOLOGICAL   SUKVEY.  29 

ground  in  a  good  condition ;  and  hence,  in  the  long  term  of 
years,  the  profits  of  these  rich  lands  greatly  exceed  those 
which  are  only  moderately  so,  naturally,  and  require  every 
few  years  an  instalment  of  manure. 

§  20.  The  similarity  in  the  composition  of  the  soils  and 
lands  surrounding  Matarnuskeet  lake  in  Hyde  county  is  re- 
markable. They  are  all  eminently  rich  in  vegetable  matter, 
and  all  are  supplied  with  a  sufficiency  of  fine  earthy  matter; 
in  which  respect  they  differ  greatly,  as  will  be  perceived 
from  the  open  ground  prairie  in  Carteret  county.  The  simi- 
larity appeared  so  great  that  I  have  not  multiplied  analyses 
of  them.  I  have,  however,  specimens  received  from  Gen. 
Blonnt,  from  Beaufort  county,  which  I  have  analyzed ;  all  of 
which  will  go  to  show  that  there  is  an  extension  of  similar 
swamp  lands  of  that  direction  in  the  county  of  Beaufort, 
which  I  have  submitted  to  analysis ;  all  of  which  go  to  prove 
the  extension  of  the  Matarnuskeet  lands  westward,  or  of 
swamp  lands  quite  similar  in  composition  to  these  justly  cele- 
brated soils. 

The  soils  which  were  collected  by  Gen.  Blount  were  four 
in  number,  and  were  taken  from  tracts,  some  of  which  had 
been  under  cultivation  several  years,  while  others  were  com- 
paratively new. 

After  having  submitted  these  soils  to  analysis,  I  stated  to 
Gen.  Blount  my  opinion  of  the  samples  I  had  operated  upon, 
and  requested  a  statement  from  him  also  of  all  the  facts  con- 
nected with  them  which  he  regarded  as  of  sufficient  import- 
ance to  be  made  public. 

In  reply  to  this  request  I  received  the  following  interesting 
communication  which  I  propose  to  incorporate  with  this  re- 
port. 

It  should  be  stated,  however,  for  the  benefit  of  those  who 
are  not  acquainted  with  Gen.  Blount's  husbandary,  that  he 
has  been  engaged  in  the  successful  culture  of  swamp  lands 
between  forty  and  fifty  years,  and  hence  is  amply  qualified 
to  express  an  opinion  respecting  their  productiveness  and 
value. 

The  following  is  the  communication  referred  to : 


<5U  NOETH-CABOLINA   GEOLOGICAL   SURVEY. 

MADISON  VILLE,  (NEAR  WASHINGTON,)  BEAUFORT  COUNTY, 

January  30£/i,  1858. 

PKOF.  EMMONS — My  Dear  Sir : — Your  letter  was  duly  received.  I  will 
now  give  you  a  description  of  the  land  of  which  the  four  parcels  sent  you 
are  specimens : 

No.  1. — A  dark  soil,  from  fifteen  to  twenty  inches  deep,  incumbent  on 
porous  clay,  with  some  fine  sand  intermixed;  through  this  substratum  the 
water  percolates  freely.  The  natural  growth  on  this  land,  (before  being 
cultivated,)  was  a  heavy  growth  of  black  gum,  a  scattering  growth  of  large 
poplars,  some  maples,  a  few  laurels ;  here  and  there  a  large  short  stra\ved 
pine.  This  land  has  been  cultivated  in  corn  for  three  years,  and  has  pro- 
duced from  40  to  50  bushels  per  acre, 

No.  3. — When  cleared,  some  ten  years  since,  was  considered  by  me  second 
quality  swamp  land.  The  growth  is  formed  of  gums,  but  more  laurels,  pines, 
and  poplars  than  No.  1.  For  ten  consecutive  years  it  has  been  cultivated 
in  indian  corn ;  when  in  its  prime  it  produced  40  bushels  per  acre — the  last 
crop  30 — the  past  season  it  was  sown  in  oats,  produced  20  bushels  per 
acre.  The  specimen  sent  you  was  taken  from  the  poorest  spot  I  could  find 
in  the  field,  (judging  from  the  growth  of  oats  then  on  it ;)  the  soil  where 
the  specimen  was  taken  from  was  about  12  inches  deep,  the  balance  of  the 
field  18. 

No.  2. — Unreclaimed  swamp— soil  from  18  to  24  inches  deep ;  subsoil  u 
different  clay  from  that  which  underlays  the  previously  described  land,  it 
is  lumpy  and  resists  the  spade.  My  opinion  is  that  the  water  does  not  pass 
freely  through  this  subsoil,  and  consequently  the  surface  soil  is  wetter  than 
on  the  lands  above  mentioned.  The  natural  growth  of  this  land  is :  reeds 
standing  ver.y  thick,  of  moderate  size,  small  sickly  pine  saplings,  red  and 
white,  bay  bushes  and  gallberry,  I  have  no  doubt  that  this  land  has  been 
often  burnt.  I  find  strata  of  ashes  at  different  depths  below  the  surface, 
and  the  stumps  of  large  pine  trees  charred.  I  own  about  3000  acres  of 
this  description  of  land — it  lays  between  the  long  leaf  pine  land  and  the 
gum  lands,  and  is  the  greater  part  of  the  year  filled  with  water  to  the  sur- 
face. For  some  time  after  every  heavy  rain  the  surface  is  partially  covered, 
and  the  water  slowly  disappears ;  every  foot  of  it  can  be  drained ;  it  ad- 
joins my  farm.  Why  should  not  such  land,  when  thoroughly  drained,  be 
fertile  V  If  it  would  not  be,  what  should  be  the  proper  treatment  to  make 
it  productive? 

No.  3  lies  between  Nos.  1  and  2. 

No.  4. — Soil  of  the  complexion  of  the  specimen  sent  you.  It  is  from  2 
to  3  feet  deep ;  incumbent  on  soapy  clay,  which  is  porous,  and  allows  an 
easy  descent  of  the  water.  The  growth  of  timber  on  this  land  is  magnifi- 
cent :  black  gums,  from  one  to  two  feet  diameter  at  the  stump,  fifty  to 
sixty  feet  to  the  limbs,  straight  bodies,  the  limbs  not  drooping,  but  forming 
with  the  body  an  angle  of  about  30  degrees,  limbs  and  twigs  showing  that 


NOKTH-CAKOLINA   GEOLOGICAL   SURVEY.  3l 

the  growth  is  healthy  and  vigorous;  a  few  very  large,  long  bodied  poplars; 
some  maples,  corresponding  in  appearance,  as  regards  size,  &c.,  with  the 
gums  above  described ;  cypress  trees,  averaging  from  8  to  10  in  number 
per  acre,  from  two  and  a  half  to  four  and  a  half  feet  diameter  at  the  stump ; 
one  hundred  feet  to  the  limbs,  straight  bodies,  small  bulky  tops,  limbs  not 
drooping  but  erect.  I  have  none  of  this  land  in  cultivation,  but  have  just 
commenced  to  reclaim  it.  My  opinion  is  it  will  be  found  equal  in  produc- 
tion to  the  lands  on  the  south-side  of  Matamuskeet  lake. 

On  a  farm  laying  on  said  lake  that  I  once  owned  I  have  made  one  hun- 
dred and  fifteen  bushels  of  indian  corn  per  acre,  and  thirty  bushels  of  wheat 
per  acre.  I  think  this  last  described  land,  No.  4,  with  perfect  drainage 
and  judicious  cultivation,  will  produce  as  much  as  the  Matamuskeet  lake 
land  spoken  of ;  appearances,  however,  may  be  deceptive. 

I  have  been,  for  a  period  of  forty  years,  engaged  in  reclaiming  and  cul- 
tivating swamp  lands,  such  as  I  have  described,  and  have  found  it  a  profit- 
able business.  I  am  located  near  the  margin  of  the  swamp,  (of  which  my 
plantation  is  a  part;)  it  contains  about  30,000  acres,  and  is  south  of  my  re- 
sidence. The  health  of  my  family,  white  and  black,  will  compare  favorably 
with  the  healthiest  locations  in  eastern  North-Carolina. 

We  have,  as  you  are  aware,  large  bodies  of  rich  swamp  lands  in  this 
portion  of  the  State.  Within  a  few  years  wealth  and  population  has 
flowed,  and  is  still  flowing  in  upon  them,  which  promises  the  happiest  re- 
sults to  the  good  Old  North  State.  Rich  swamp  land,  like  almost  every 
thing  else,  will  show  after  a  while  the  effects  of  bad  treatment,  but  fortun- 
ately for  us,  if  we  impoverish  our  land  by  severe  and  injudicious  cultiva- 
tion, we  have  in  close  contiguity  inexhaustible  supplies  of  shell  marie, 
which  has  proved  itself  a  panacea  to  worn  down  swamp  land.  Guano  and 
the  other  manures  in  common  use  produce  as  fine,  perhaps  a  better  effect, 
on  swamp  land  than  any  other  description  of  land  of  which  I  have  any 
knowledge.  I  fear,  sir,  I  have  taxed  you  too  severly ;  the  interest  I  feel  as 
a  citizen  of  the  eastern  part  of  the  State  I  mention  as  my  justification. 
Should  you  wish  more  specific  information  than  I  have  given,  it  will  afford 
me  pleasure  to  furnish  it. 

Such  is  my  great  aversion  to  writing,  I  have  been  compelled  to  enlist 
the  aid  of  my  daughter,  Mrs.  B.,  who  is  now  with  me.  You  will  perceive 
that  a  lady  has  been  my  amanuensis. 

Most  respectfully, 

WILL.  A.  BLOUNT,  SEN'K. 

From  the  foregoing  communication  the  reader  will  be  pre- 
pared to  form  a  correct  opinion  of  the  character  of  the  swamp 
lands  referred  to,  especially  when  taken  in  connexion  with 
their  composition  as  determined  by  analysis. 

No.  1. — On  being  exposed  for  a  few  weeks  to  the  air  be- 


OZ,  NORTH-CAROLINA   GEOLOGICAL   SURVEY. 

comes  dry.  Its  color,is  blackish  brown,  it  contains  undecom- 
posed  bark,  wood  and  some  roots,  but  is  mostly  made  up  of 
decomposed  vegetable  matter.  The  earthy  part  is  not  visible 
as  in  many  vegetable  soils  of  the  poorer  class. 

On  submitting  it  to  analysis  I  found  it  composed  of  the 
following  elements: 

Silex,    65.540 

Organic  matter,   26.100 

Water,   6.050 

Peroxide  of  iron  and  alumina, 4.920 

Garb,  lime,    0.490 

Magneisa,   0.050 

Potash, 0.003 

Soda,  0.020 

Phosphoric  acid,   0.003 

The  silex,  as  in  most  of  the  good  swamp  soils,  is  extremely 
fine.  Its  color  is  drab,  and  hence  probably  contains  a  small 
quantity  of  alumina  which  cannot  be  detached  without  being 
attached  by  potash. 

This  soil,  it  is  evident,  still  contains  the  elements  of  fertility. 
and  it  is  also  evident  that  it  will  bear  cultivation  for  years  to 
come  without  exhaustion.  It  will  be  observed  that  the 
natural  growth  upon  this  soil  is  one  which  indicates  fertility. 
as  the  poplar  and  black  gum,  and  a  large  growth  of  short 
leaved  pine,  the  growth  being  very  heavy. 

No.  2. — This  specimen  or  mass  of  soil  consists  apparently  of 
vegetable  matter  without  any  earth.  It  is  black,  and  pre- 
serves a  moist  state,  though  it  has  been  exposed  to  the  air  in 
a  box  for  several  months ;  and  on  being  exposed  in  a  drying 
oven  lost  its  moisture  very  slowly.  It  contains  fresh  vegeta- 
ble fibres,  portions  of  partially  decomposed  wood  and  bark, 
etc.  Still  it  is  rather  homogeneous,  and  is  unlike  the  coarse 
fibrous  soil  of  the  open  prairie  of  Carteret. 

On  submitting  this  soil  to  analysis,  I  found  it  composed  of 
the  following  elements : 

Silex, 74.600  74.SOO 

Organic  Matter,  18.000  18.100 

Peroxide  of  iron  and  alumina,    3.100  3.100 

Phosphoric  acid,  0.021  trace, 


NOBTH-CAKOLINA  GEOLOGICAL   SURVEY.  33 


Lime, 0.049  0.040 

Magnesia,   0.005  0.005 

Potash,  0.040  trace,  % 

Soda, , 0.030  trace, 

Water,    4.000  4.000 


98.845  99.845 

This  soil  was  dried  before  the  quantity  was  weighed  for 
analysis.  When  exposed  to  about  300  degrees  of  Fah.,  it  lost 
fifteen  per  cent,  of  water. 

This  soil  has  not  been  cultivated,  and  though  it  looks  rich, 
still  I  am  inclined  to  regard  it  as  a  poorer  soil  than  No.  1. 
It  contains  more  sand,  is  rather  coarser,  and  less  alumina, 
iron  and  vegetable  matter.  The  alkaline  earths,  as  lime  and 
magnesia,  are  much  less.  The  same  may  be  said  of  the  alka- 
lies, potash  and  soda.  The  depth  of  this  soil  is  from  eighteen 
to  twenty-four  inches,  resting  on  a  hard  and  rather  impervious 
bottom.  Its  natural  growth  is  also  different ;  as  it  consists  of 
reeds  standing  very  thick,  and  small  sickly  pine  saplings,  red 
and  white  bay  bushes,  gallberry,  etc. 

This  growth,  it  is  evident,  might  be  due  to  the  impervious 
bottom,  or  its  low  temperature  ;  but  it  is  also  in  part  due  to 
the  absence  of  the  most  important  elements  of  fertility. 
There  is  no  doubt,  however,  but  a  low  temperature,  which  is 
due  to  the  presence  of  water,  is  competent  to  produce  an  ap- 
parent sterility,  low  bushes  of  peculiar  kinds,  as  bay,  gall- 
berry,  alder  and  willow. 

No.  3. — The  color  of  this  soil  is  a  dark  ash  or  gray.  It  has 
become  dry  in  the  box  in  which  it  was  sent,  while  No.  2  has 
remained  wet.  It  is  pulverulent  and  light,  though  somewhat 
lumpy.  The  vegetable  matter  exists  evidently  in  a  large  pro- 
portion, yet  a  close  observer  would  perceive  that  it  is  less 
than  in  No.  2. 

On  submitting  it  to  analysis,  I  found  its  composition  as  fol- 
lows: 

Silei 81.600 

Vegetable  matter, 12.800 

Peroxide  of  iron  and  alumina, 4.100 

Carb.  of  lime,  0.020 

4 


GEOtOGICAL  StFKVET. 

Magnesia,    ..*...; •...     0.010' 

Phosphoric  acid, trace, 

Potash, „ trace. 

This  soil  was  regarded  by  Gen.  Blount  as  second  quality. 
Its  growth  consisted  of  low  pines,  gums  and  poplars.  It  how- 
ever produced  forty  bushels  of  corn  to  the  acre,  but  the  last 
crop  was  only  thirty  bushels.  Afterwards,  it  gave  twenty 
bushels  of  oats  to  the  acre. 

The  proportion  of  silex,  it  will  be  perceived,  is  much  greater 
than  in  No.  1.  The  specimen  was  taken  from  a  poor  spot  in 
the  field.  It  had  been  under  culture  for  ten  years.  Depth 
of  soil  twelve  inches. 

In  attempting  the  solution  of  the  question,  why  a  poor 
crop  was  at  last  produced,  we  should  not  forget  that  certain 
soils  in  this  climate  become  dry  at  an  early  day ;  and  if  so, 
we  invariably  find  the  cereals  growing  very  slim  and  slender, 
and  perhaps  soon  cease  to  grow,  turn  yellow,  and  produce,  if 
any,  a  very  small  ear  of  grain.  In  a  shallow  soil  such  a  re- 
sult may  be  expected,  notwithstanding  the  soil,  on  analysis, 
may  be  found  to  contain  the  elements  of  fertility.  In  the 
same  field,  plants  growing  in  the  same  soil,  a  part  may  yield 
seed  and  fruit,  and  another  will  fail;  the  results  being  de- 
pendent on  the  existence  of  moisture  surrounding  the  roots 
of  the  plant. 

No.  4. — The  color  is  grayish  black,  and  contain  half  decom- 
posed roots,  bark,  etc.  It  has  also  partially  dried  in  the  box, 
and  in  drying,  becomes  lighter  colored.  This  soil  is  deeper 
than  either  of  the  preceding,  being  between  three  and  four 
feet  deep,  and  incumbent  on  a  porous  bottom. 

The  growth  is  very  large,  consisting  of  black  gum  from  one 
to  two  feet  in  diameter,  and  from  fifty  to  sixty  feet  high.  The 
limbs  are  straight  as  well  as  the  bodies.  Very  large  poplars 
also  are  found  scattered  over  the  field,  also  cypress  in  clusters 
from  eight  to  ten  in  each. 
This  sample  I  found  composed  as  follows  : 

Silex,    77.500 

Organic  matter,  15.400 

Peroxide  of  iron  and  alumina, 4     6.900 


GEOLOGICAL  SURVEY.  35 

Lime,  0.500 

Magnesia, 0100 

Potash,  0.019 

Soda,    0.028 

Phosphoric  acid,  * * 0.400 

Sulphuric  acid,  . , 0.180 

Portions  of  this  soil,  on  being  dried  in  an  oven  at  300  deg. 
lost  thirty-four  per  cent,  of  water.  The  silex  is  extrmely  fine, 
and  similar  in  appearance  to  the  Hyde  county  soils.  It  is, 
however,  in  a  greater  proportion,  and  there  is  less  organic 
matter.  But  there  is  no  doubt  this  soil  will  be  productive 
when  drained  and  put  under  cultivation.  It  appears  estab- 
lished from  observation  and  experiment  upon  the  swamp 
lands  of  the  eastern  counties,  that  much  depends  on  the  fine- 
ness of  the  earthy  matter ;  for  when  there  is  a  perceptible 
coarseness,  the  land  will  not  bear  cultivation  many  years. 
There  is  in  those  cases,  however,  less  alumina  and  iron,  and 
hence  this  kind  of  soil  dries  readily;  and  in  certain  seasons 
crops  will  be  very  short,  and  in  reality  fail.  Where  the 
earthy  matter  is  fine  it  retains  moisture,  and  furnishes  a  sup- 
ply for  those  seasons  when  the  rains  are  unseasonable.  In 
certain  cases  the  extreme  fineness  of  the  earth  would  present 
other  defects.  It  would  become  too  compact  and  close,  and 
exclude  the  air.  But  the  vegetable  matter  counteracts  this 
defect  in  the  swamp  lands. 

The  gallberry  lands  often  appear  rich,  if  their  vegetation 
did  not  remind  one  of  their  poverty.  It  will  be  found,  in 
most  cases  of  the  poorest  kinds  of  this  class  of  lands,  that  the 
sand  may  be  seen  in  the  mass,  or  shows  through  its  black 
covering  of  vegetable  mould.  On  examination,  the  sand  will 
be  found  to  be  coarse.  Under  cultivation  the  vegetable  mat- 
ter disappears  rapidly ;  it  is  readily  burnt — and  the  surface 
soon  becomes  white  with  the  marine  sand,  and  in  extreme 
cases  blows  into  ridges.  Lands  of  this  description  do  not  pay 
the  expense  incurred  in  draining.  It  is  sometimes  necessary 
to  drain  them,  in  order  to  effect  the  drainage  of  other  con- 
tiguous tracts. 

Neither  of  the  four  foregoing  soils  of  Gen.  Blount's  planta- 


36  NORTH-CAROLINA  GEOLOGICAL  SUB  VET.  % 

tions  belong  to  the  poor  gallberry  lands,  though  No.  2  might 
be  ranked  in  the  better  class  of  this  description  of  soils. 

The  texture  of  the  gallberry  lands  has  much  to  do  with 
their  poverty ;  for  generally  they  are  made  up  of  stiff  whitish 
clays  and  coarse  sand.  From  analysis  we  might  prove  that 
their  constituents  were  the  same  as  in  productive  kinds  of 
soils.  Such  facts  prove  that  productiveness  is  not  entirely 
dependent  on  composition. 


CHAPTER  III. 

e 

Topography  of  the  Eastern  Counties,  from  Wake  eastward  to  Onslovr 
County.  Character  of  the  soil  of  the  White  Oak  Desert.  Mr.  Francke's 
Pocosin  and  Swamp  Lands.  Better  kind  of  Gallberry  Swamp  Land. 
Swamp  Lands  of  the  Brown  Marsh.  Green  Swamp  Lands.  Mr.  Mc- 
Neil. Will  pay  for  drainage.  Barren  soil  of  Bogue  Sound,  furnished 
by  D.  A.  Humphrey,  Esq.,  with  his  letter.  Cause  of  barrenness  in  these 
soils. 

§  21.  From  "Wake  county  eastward  to  the  shore  of  the  At- 
lantic the  country  slopes  gently,  the  greatest  inclination  being 
of  course  on  the  western  side  of  the  plane.  Between  "Wake 
and  Johnston  the  country  is  rolling.  From  Smithfield,  in 
Johnston,  to  Clinton,  in  Sampson  county,  the  country  is  still 
somewhat  rolling ;  but  much  less  so  than  between  Johnston 
and  Wake.  A  large  proportion  of  the  country,  however,  be- 
tween Smithfield  and  Clinton  is  a  flat  piney  woods.  The 
land  seven  or  eight  miles  west  of  Clinton  is  level  and  rather 
sandy. 

In  Duplin  county  the  level  swamp  lands  begin.  Between 
Magnolia  on  the  railroad  and  Onslow  county,  the  country  is 
low  and  swampy,  and  in  Onslow  there  are  large  tracts  of  un- 


NORTH-CAROLINA  GEOLOGICAL  SURVEY.  37 

settled  or  unreclaimed  swamp  and  pocosin  lands  of  an  excel- 
lent quality.  One  tract  in  particular  contains  a  hundred 
square  miles,  and  a  large  proportion  of  it  is  excellent  swamp 
land — and  some  tracts  are  equal  to  the  corn  lands  of  Hyde 
county. 

Johnston  county  contains  large  tracts  of  flat  piney  woods, 
the  soil  of  which  produces  only  the  shrubs  which  indicate 
unproductiveness,  as  the  gallberry,  ilex,  and  magnolia  or  bay, 
with  a  small  growth  of  the  long  leaved  pine.  The  surface,  if 
not  covered  with  water,  is  liable  to  be  overflowed — and  as  it 
consists  of  sand  and  clay,  with  a  mixture  of  vegetable  mould, 
may  be  said  to  be  quite  impervious  to  water ;  and  hence,  the 
surface  water  stands  over  it  for  a  long  time,  and  its  tempera- 
ture remains  too  low  for  the  growth  of  the  more  valuable 
trees  and  plants.  Towards  Sampson  county  the  country  im- 
proves, and  upon  the  branches  of  the  Six  Run  there  are  rich 
plantations.  The  best  swamp  lands  are  still  farther  east ;  and 
these,  while  they  are  usually  high  enough  to  admit  of  drain- 
age, are  rarely  more  than  fifty  feet  above  tide  level.  The 
Hyde  county  corn  lands  are  about  five  feet  above  tide  level, 
or  may  be  less  than  four  feet.  Sometimes,  in  close  proximity 
to  the  sounds,  as  in  Carteret,  the  swamps  are  heaped  up  as 
it  were,  and  hence  may  be  from  twelve  to  sixteen  feet  above 
the  level  of  the  sea. 

In  Onslow  county,  the  soil  between  Thompson's  and  Jack- 
sonville is  very  good.  Some  of  it  is  suitable  for  the  ground 
pea,  being  a  light  soil  with  considerable  vegetable  matter. 

§  22.  In  Onslow,  the  White  Oak  desert  is  the  most  inter- 
esting tract  of  swamp  land  in  the  county,  it  is  at  the  head  of 
White  Oak  creek.  This  tract  may  be  drained  into  Trent 
river.  The  timber  is  very  large,  and  consists  of  white  oak, 
poplar  and  pines. 

The  most  important  work  which  has  been  undertaken,  is 
the  drainage  of  a  part  of  this  tract  by  Mr.  Francke.  He  has 
been  able  to  secure  two  objects,  the  drainage  of  the  land  and 
a  good  water  power,  with  a  fall  of  about  twelve  feet.  The 
cost  of  cutting  the  main  drain  or  canal  is  fifteen  cents  per 
square  yard.  The  thickness  of  the  soil  in  Mr.  Franke's  po- 


38 


NORTH-CAROLINA   GEOLOGICAL  SURVEY. 


eosin*  is  five  feet  towards  the  outer  rirn,  and  still  thicker  to- 
wards the  middle,  attaining  at  least  ten  feet  of  rich  soil. 
This  pocosin  is  said  to  vary  much  in  its  depth  and  quality ; 
some  parts  are  sandy,  and  the  trees  are  still  large  and  nu- 
merous. These  sandy  knowles  are  called  islands.  But  the 
excellent  quality  of  parts  of  it  which  are  covered  with  heavy 
timber,  prove  by  cultivation  that  it  is  equal  to  the  Matamus- 
keet  lands  of  Hyde — their  average  yield  being  twelve  barrels 
of  corn  to  the  acre. 

I  have  not  seen  the  land  referred  to  in  Jones  county,  but  I 
am  confirmed  in  the  statement  from  its  composition,  which  I 
have  determined  by  a  careful  analysis.  Thus  the  drained 
portion  of  Mr.  Francke's  pocosin  gave  me  a  result  on  analysis 
equal  in  value  to  the  best  of  the  Hyde  county  soils.  It  is  as 
follows : 

Silex,    , , 60.000 

Organic  matter, 25.000 

Peroxide  of  iron  and  alumina, 11.030 

Phosphoric  acid,  , , 0.312 

Lime,  1  500 

Magnesia,    , , 0.300 

Potash,  0.010 

Soda ...,'•  0.020 

Soluble  silica,  0.100 

Water,  2.713 

From  the  foregoing  results,  when  compared  with  those 
obtained  by  anaylsis  of  the  Hyde  county  soil,  it  will  be 
acknowledged  that  if  composition  is  a  test  which  can  be  relied 
upon,  the  Onslow  swamp  lands  must  be  very  valuable ;  and 
furthermore,  that  this  value  justifies  the  expense  required  in 
draining.  This  is  the  first  question  to  be  settled  in  all  swamp 
lands:  are  their  qualities  good  enough  to  justify  this  neces- 
sary expense  ?  because  they  must  be  drained  before  the  cereals 
can  be  cultivated.  The  encouragement  to  incur  this  first  ex* 
pense  arises  from  the  fact  that  when  drained  they  do  not 


*  This  pocosin  is  partly  in  Onslow  and  partly  in  Jones  county.    The  portion 
which  has  been  drained  and  cleared  is  in  Jones  Bounty.    The  only  meaning  which 
J  can^attach  to  the  word  pocosin  is,  that  it  is  a  large  swamp. 


.  NORTH-CAROLINA   GEOLOGICAL   SURVEY.  39 

wear  out  in  the  life  time  of  man  ;  they  require  no  manures, 
they  are  easily  tilled,  and  they  produce  large  crops  annually, 
and  besides  are  less  affected  by  droughts ;  or,  in  other  words, 
the  corn  crop  is  more  sure  and  certain  than  upon  up  lands. 

Where  there  are  large  continuous  tracts  as  in  Onslow,  Jones, 
Hyde  and  Beaufort,  a  systematic  plan  of  drainage  should  be 
undertaken.  This  should  be  based  upon  a  topographical 
survey  of  the  whole  tract,  ascertaining  first  the  area  and  its 
irregularities,  if  any,  then  the  regular  slope  and  the  most 
feasible  points  to  which  the  drains  and  canals  should  run.  If 
a  main  canal  can  be  cut  which  will  take  water  sufficient  for 
boat  navigation,  it  should  be  regarded  as  an  important  means 
for  transportation.  It  is  surprising  that  swamp  lands  hold  so 
much  water — so  that  most  of  the  largest  tracts  of  pocosin 
lands  furnish  a  sufficiency  for  this  purpose. 

The  earthy  matter  in  the  pocosin  of  Onslow  is  very  fine, 
and  of  drab  color,  in  which  respects  it  is  similar  to  the  best 
lands  of  Hyde. 

It  is  evident  also  from  an  inspection  of  the  results  of  this 
analysis,  that  there  is  a  full  supply  of  lime,  and  of  the  more 
.expensive  elements,  and  hence  it  may  be  expected  that  when 
these  lands  have  been  brought  under  full  cultivation  by 
thorough  drainage  and  other  means  necessary  to  favor  the 
growth  of  the  cereals,  that  farms  or  plantations  as  valuable  as 
any  in  North-Carolina,  will  be  formed  out  of  this  desert 
swamp.  The  determination  of  the  high  value  of  this  part  of 
Onslow  I  consider  of  great  importance ;  for  there  seems  to 
have  been  hitherto  great  backwardness  in  attempting  to  re- 
claim the  lands  of  "White  Oak  desert.  It  is  true  the  under- 
taking is  a  formidable  one,  but  the  rich  results  which  will 
certainly  be  secured  thereby  fully  warrant  the  undertaking. 

§  23.  The  character  of  the  gallberry  lands  require  also 
new  investigation.  These  have  usually  been  regarded  as 
worthless.  They  are  usually  flat  and  wet,  and  hence  the 
temperature  of  the  surface  is  always  too  low  for  the  vigorous 
growth  of  the  most  valuable  trees :  aside  from  this  fact  it  is 
probable  that  the  soil  is  really  poor  and  unfertile,  and  no 
measures  within  a  reasonable  expense  could  be  employed  to 


40  NORTH-CAROLINA  GEOLOGICAL  SURVEY. 

change  this  semi-barren  condition  to  one  of  fertility.  But  it 
is  equally  probable  that  many  large  tracts  of  land  which  are 
classed  among  the  gallberry  lands  may  be  reclaimed  and  will 
become  fertile  by  thorough  drainage. 

In  forming  a  judgment  upon  the  expediency  of  draining 
these  flat  and  wet  lands  with  a  view  to  their  cultivation,  it  is 
necessary  to  examine  the  texture  of  the  materials  which  com- 
pose them  as  well  as  their  composition.  As  there  is  a  large 
proportion  of  black  vegetable  matter  upon  the  surface,  it  is  im- 
portant to  ascertain  if  it  is  intermixed  with  earth,  and  if  so 
whether  it  is  coarse  or  fine,  and  whether  it  is  mostly  sand,  whose 
particles  are  large  or  visible  at  once  on  inspection.  If  the 
earth,  after  the  vegetable  matter  has  been  consumed,  is  fine 
and  impalpable,  it  is  a  fact  which  speaks  well  of  its  character ; 
if  on  the  contrary  it  is  a  white  and  coarsish  sand,  it  is  unfavor- 
able, for  it  cannot  be  expected  that  it  holds,  in  mechanical 
combination  the  more  essential  earths,  alumina,  lime  and 
magnesia,  or  the  alkalies,  potash  and  soda.  If  it  is  sand  these 
important  elements  will  be  in  combination  with  the  vegetable 
matter,  and  when  this  has  become  an  ash,  or  is  partly  con- 
sumed, the  soil  will  be  destitute  of  the  elements  of  fertility. 
Observation  and  experience  prove  the  correctness  of  the 
foregoing  observations.  If,  for  instance,  the  soils  of  Hyde 
county  are  examined,  the  fine  impalpable  material  is  always 
found  intermixed  with  the  vegetable  matter ;  and  so,  in  cases 
where  the  sand  is  found,  and  soon  appears  after  cultivation, 
the  lands  do  not  wear  well  but  soon  give  out. 

But  the  gallberry  lands  are  frequently  stiff,  whitish  clays 
intermixed  with  sand.  These  have  undergone  very  little 
change  from  the  influence  of  atmospheric  agencies.  "When 
ploughed  and  exposed  for  a  few  years  to  the  atmosphere  the 
color  slowly  changes  to  a  light  brown,  and  finally  to  a  deeper. 
These  changes  are  also  favorable,  and  it  will  be  found  that 
these  lands  improve  by  cultivation. 

As  an  example  of  the  better  kind  of  gallberry  land,  I  pro- 
pose to  give  the  composition  of  one  which  occupies  a  large 
area  in  Onslow  county,  which,  on  being  submitted  to  analy- 
sis, gave  the  following  results : 


NOKTH-CAEOLINA   GEOLOGICAL   SURVEY.  - 

Silex, 82.300 

Peroxide  of  iron  and  alumina, 8.700 

Lime,   0.020 

Magnesia,    0.010 

Phosphoric  acid, 0.150 

Organic  matter,  . . ., 3.350 

Potash  and  soda, traces, 

Soluble  Silica, 0.100 

Water, 6.000 

The  color  of  this  soil  is  a  light  yellow,  and  its  texture  rather 
fine,  and  is  disposed  to  be  lumpy.  Its*texture  and  composi- 
tion favor  the  growth  of  wheat  rather  than  corn,  and  I  have 
no  doubt  when  reclaimed  by  drainage  will  prove  an  excellent 
soil  for  the  cultivation  of  this  grain. 

§  24.  The  swamp  lands  of  Brunswick  and  New  Hanover, 
and  the  adjoining  counties,  resemble  in  many  respects  those  of 
Hyde  and  Onslow.  In  order  to  determine  as  far  as  possible 
from  analysis  the  expediency  of  draining  a  certain  tract  or  a 
portion  of  it  lying  in  Brunswick  county,  which  is  known  as 
the  Green  swamp,  Mr.  McNeil  *  furnished  me  with  a  few 
samples  of  muck  which  were  obtained  as  it  appeared  from 
beneath  the  water.  It  was  similar  to  black  mud,  but  on  dry- 
ing I  found  it  contained  partially  decayed  pieces  of  bark, 
wood  and  roots,  though  its  structure  did  not  appear  to  be 
fibrous. 

On  drying  in  the  paper  in  which  it  was  orginally  wrapped, 
it  became  rather  hard  and  firm,  showing  that  it  contained 
earth,  for  if  made  up  of  peaty  matter  destitute  of  earth,  it 
would  have  been  much  less  firm  and  compact. 

On  submitting  this  material  to  analysis,  I  found  it  was 
composed  of  the  following  elements : 


*JACK  FOREST,  24th  November,  1857. 

DEAR  SIR  : — I  send  you  four  packages  of  soil  from  our  swamp  lands :  one  from  the 
heavy  timbered  land  on  the  Brunswick  marsh;  one  from  the  low  lands  of  the  Brown 
marsh,  and  lands  requiring  ditching;  one  from  the  original  Green  swamps,  but  now 
timbered  with  young  growth,  and  one  from  a  ditch  draining  the  land  near  the  dwainp, 
which  I  suppose  contains  lime. 

Yours  truly, 

H.  J.  McNEIL. 


42  NORTH-CAROLINA  GEOLOGICAL   SURVEY. 

Silex,    , 35.350 

Peroxide  of  iron  and  alumina, 10.85 

Organic  matter, „ 37.50 

Water,     15.8 

Lime,  1.40 

Magneisa, 0.15 

Potash,  , 0.10 

Soda, 0.15 

This  soil  was  found  to  be  much  richer  than  I  anticipated, 
and  on  drying  in  paper,  it  retained  a  larger  quantity  of  water 
than  I  expected.  If  the  composition  had  been  obtained  after 
most  of  the  water  was  expelled  by  heat,  the  proportion  of 
the  elements  of  fertility  would  have  been  proportionally 
greater.  As  the  soil  is  composed,  there  can  scarcely  remain 
a  doubt  of  the  value  of  these  lands.  The  earthy  matter  is  as 
fine  as  that  of  the  Onslow  or  Hyde  county  lauds,  and  its 
quantity  and  condition  proves,  as  it  appears  to  me,  the  same 
capability  with  them  for  a  productive  cultivation  for  a  series 
of  years.  Hence  the  cost  of  drainage  should  be  incurred, 
and  these  valuable  lands  reclaimed,  inasmuch  as  they  pay 
better  than  the  uplands.  The  extent  of  unreclaimed  lands 
of  this  description  makes  it  still  more  expedient,  inasmuch  as 
the  general  results  are  proportionately  greater  than  when  the 
surface  embraces  only  a  few  acres. 

The  depth  of  this  material  is  from  eighteen  to  twenty-five 
or  thirty  inches,  but  like  the  Onslow  pocosin  it  is  variable, 
and  like  the  latter  also,  the  swamp  abounds  in  islands,  which 
are  frequently  occupied  by  inhabitants  who  contrive  to  live 
by  basket  making.  The  timber  consists  of  cypress  and  black 
gum,  and  various  pines  and  oaks,  which  frequently  attain  a 
large  size,  proving  by  the  natural  method  a  productive  soil. 
In  passing  through  these  low  lands,  the  water  is  frequently 
deep  in  the  common  highway ;  sometimes  it  is  due  to  the 
prevalence  of  rains,  in  others  it  is  produced  by  dams  to  ob-^ 
tain  a  water  power  for  mills.  As  it  respects  the  practice 
of  maintaining  mills  in  this  low  and  half  inundated  country, 
it  seems  to  me  to  be  inexpedient.  It  certainly  prevents  in 
part  the  reclamation  of  these  lands  by  drainage,  and  when  it 
is  taken  into  consideration  that  steam  power  cannot  be  very 


NORTH-CAROLINA   GEOLOGICAL   SURVEY. 


43 


expensive  in  a  country  abounding  in  wood,  it  becomes  quit® 
plain  that  all  such  mills  should  be  suffered  to  go  down  and 
their  places  supplied  by  the  much  more  efficient  steam  mills. 
The  soil  taken  from  the  bank  of  a  ditch  is  of  a  dark  drab 
or  purplish  gray.  It  coheres  strongly  on  drying  and  loses 
most  of  its  water.  It  is  gritty  to  the  feel  and  is  composed  of 
moderately  fine  quartz  and  clay.  On  submitting  it  to  analy- 
sis I  found  it  composed  of 

Silex,   83.00 

Organic  matter, 21.20 

Peroxide  iron  and  alumina, 7.40 

Lime, trace, 

Magnesia, trace, 

Potash  and  soda,  undetermined,  

Water, , 3.20 

The  lime  and  magnesia  were  scarcely  perceptible.  It  re- 
sembles in  appearance  and  composition  the  poorer  gallberry 
lands,  though  it  is  probably  better  than  many.  If  a  soil  of 
this  description  was  to  be  put  under  cultivation  it  would  re- 
quire steady  and  constant  marling.  It  forms  a  good  subsoil 
in  one  respect,  that  of  being  impervious  and  capable  of  hold^- 
ing  manures.  It  unlies  the  cultivable  soil  iof  the  swamp 
lands  in  this  neighborhood.  The  soil  taken  from  the  Bruns- 
wick swamp  is  brown  or  brownish  ;  contains  undecomposed 
twigs,  bark,  &c.,  but  on  drying  forms  a  firm  mass  and  con- 
tains a  sufficiency  of  earthy  matter.  It  is  not  unlike  much 
of  the  soil  of  Hyde  county,  and  it  appears  that  it  has  been 
heavily  timbered.  I  found  it  composed  of 

Silex,    . . , , 45.470 

Water, 8.000 

Organic  matter,  34.000 

Peroxide  of  iron  and  alumina, 10.490 

Lime,   0.490 

Magnesia,    , 0.060 

Potash,  0.581 

Soda, 0.326 

Soluble  silica, 0.580 

This  soil  possesses  a]l  the  good  qualities  of  the  Hyde  county 
soils.  It  absorbs  and  retains  water  strongly.  The  mass  of 


44  NORTH-CAROLINA  GEOLOGICAL  SURVEY. 

soil  on  drying  becomes  hard  and  tough,  requiring  force  to 
break  it,  and  yet  when  apparently  perfectly  dry  holds  eight 
per  cent  of  water.  It  is  also  sufficiently  rich  in  lime,  and 
particularly  in  organic  matter.  The  question  to  be  solved  by 
analysis  was  whether  these  lands  would  become  valuable  by 
drainage.  We  may  be  assured  this  is  proved  by  the  results 
obtained  by  analysis.  The  expediency  of  drainage  depends, 
however,  very  much  upon  the  cost  of  the  undertaking,'  but  if 
the  lands  admit  of  drainage  at  the  ordinary  cost  of  such  un- 
dertakings there  is  no  doubt  but  that  the  soil  would  rank 
among  the  most  valuable  in  the  State. 

§  25.  The  foregoing  analysis  furnish  examples  of  soils,  most 
of  which  may  be  regarded  as  highly  productive.  In  the 
midst  however  of  productive  lands,  there  are  very  frequently 
limited  tracts  wrhich  are  really  barren,  so  far  as  the  cereals 
are  concerned.  To  the  eye,  or  upon  a  mere  cursory  exam- 
ination, these  tracts  would  be  regarded  as  valuable  as  any 
which  lie  adjacent  to  them ;  yet  experience  would  prove,  in 
an  attempt  to  cultivate  them,  that  they  are  worthless.  Corn 
takes  root  and  grows  a  few  weeks,  when  it  begins  to  turn 
yellow,  and  finally  dries  up,  or  lives  on  in  a  stinted  condition. 

The  cause  of  this  unexpected  termination  is  not  well  un- 
derstood. Some  planters  believe  that  the  soil  is  lacking  in 
one  or  more  of  the  elements  of  growth ;  others,  that  there 
is  some  substance  of  a  poisonous  quality  in  the  soil.  If  either 
of  these  suppositions  or  guesses  were  true,  the  fact  might  be 
determined  by  submitting  the  soil  to  a  careful  analysis. 

But  there  are  other  causes  which  affect  unfavorably  the 
growth  of  vegetables.  It  may  be  too  tenacious,  it  may  be 
compact  and  prevent  the  access  of  air,  (an  element  always 
required,)  or  it  may  be  so  porous  and  open  that  the  necessary 
amount  of  moisture  cannot  be  retained.  In  addition,  there- 
fore, to  the  chemical  composition  of  a  soil  which  a  plant  may 
require  to  insure  its  perfection,  there  may  be  an  incompati- 
ble physical  one,  whose  operation  is  equally  effective  in  stint- 
ing its  growth.  We  must  not,  therefore,  regard  barrenness 
as  always  the  result  of  the  absence  of  fertilizing  elements. 
In  investigating  any  particular  case  of  infertility,  it  is  neces- 


NOKTH-CAROLINA   GEOLOGICAL   SURVEY.  45 

sary  in  the  first  place  to  inquire  into  its  physical  condition — 
to  ascertain  its  texture,  the  size  of  its  particles,  and  at  the 
same  time  ascertain  whether  they  are  silicious  and  coarse, 
and  insusceptible  of  retaining  water  or  fertilizing  matter. 

Many  examples  of  these  unproductive  tracts  belong,  geo- 
logically, to  the  most  recent  formation,  as  the  Postpliocene  of 
authors.  They  are  properly  marine  formations,  in  which 
sand,  as  will  be  seen  in  the  sequel,  forms  the  largest  propor- 
tion of  the  elements  of  the  compound. 

A  specimen  of  the  unproductive  soil  was  received  from  D. 
A.  Humphrey,  Esq.,  of  Swansboro',  Onslow  county,  accom- 
panied with  a  letter  containing  a  brief  account  of  the  mate- 
rial under  consideration,  the  copy  of  which  is  in  the  follow- 
ing words : 

SWANSBORO',  N.  C.,  Jan.,  1858. 

DEAR  SIR  : — You  will  remember,  that  at  Beaufort,  last  May,  when  I  had 
the  pleasure  of  an  introduction  to  you,  you  told  me  if  I  would  send  you  a 
specimen  of  some  of  that  peculiar  land  of  which  we  talked,  you  would  an- 
alyze and  inform  me  of  its  constituents,  and  advise  me  of  the  necessary 
change  to  be  made  in  it,  so  as  to  make  it  produce  the  ordinary  crops. 

The  land  from  which  this  specimen  was  taken  produces  weeds  and  vege- 
tables common  to  all  the  sound  land,  very  scantily,  except  the  sweet  fennel 
(Foenicuhim)  which  grows  very  luxuriantly,  so  large  even,  that  I  have 
them  taken  up  with  a  grub-hoe.  It  will  produce,  with  the  best  cultivation, 
(without  manure,)  say  100  Ibs.  seed  cotton  to  the  acre,  and  one  bushel  corn. 
When  the  corn  first  springs  up,  it  grows  rapidly  for  a  short  time ;  then 
turns  yellow  and  falls.  The  land  is  quite  elevated.  I  have  shipped  to 
Wilmington  a  small  bag  containing  the  specimen,  from  which  place  you 
will  soon  receive  it,  and  when  it  suits  your  convenience  to  examine,  please 
do  so,  and  let  me  hear  from  you. 

And  oblige,  very  much, 

Your  friend  and  humble  serv't, 

D.  A.  HUMPHREY. 

PROP.  E.  EMMONS,  Raleigh,  N.  C. 

On  submitting  the  soil  described  in  the  foregoing  letter,  I 
found  it  composed  of  the  following  elements : 

Silex,    85.200 

Peroxide  of  iron  and  alumina,  2.862 


4:6  NORTH-CAROLINA   GEOLOGICAL   SURVEY. 

Carbonate  of  lime,  1.85 

Magneisa,    trace, 

Organic  matter,  f  .0$ 

Water,    2.50 

The  phosphates  and  potash  scarcely  distinguishable  in  200 
grains.  The  sand  representing  the  silex  is  rather  coarse, 
grains  distinctly  visible  and  rather  angular.  The  color  of  the 
mass  is  black,  and  it  seems  to  be  made  up  of  fine  vegetable 
matter.  It  contains,  as  will  be  seen,  a  sufficient  quantity  of 
lime  and  inorganic  matter — the  former  is  derived  from  parti- 
cles of  marine  shells,  sometimes  of  a  large  size,  and  it  is 
probable  all  the  lime  is  coarse ;  it  effervesces  with  acids." 
The  silex,  though  large,  is  not  in  greater  proportion  than  in 
many  productive  soils.  It  would  be  regarded  as  a  light  soil^ 
though  the  vegetable  matter  might  deceive  one  who  has  had 
no  experience  in  cultivating  soils  of  this  description.  A  soil 
of  this  character  presents  two  questions  for  solution :  1st, 
whether  its  present  or  natural  state  will  justify  an  expendi- 
ture sufficient  to  make  it  fertile  ?  and  2d,  if  so,  what  course 
should  be  adopted  to  secure  the  object  sought  for?  My  first 
impression  is  that  it  cannot  be  made  productive  at  all,  in  con- 
sequence of  its  composition.  It  has  really  only  a  base  of 
coarsish  sand  of  considerable  depth.  Hence  it  Is  loose  and 
'porous,  and  transmits  all  the  water  through  it.  Besides,  it  is 
evident  that  there  is  a  deficiency  of  alumina  and  all  the  most 
expensive  elements  except  lime,  and  the  lime,  instead  of  be- 
ing fine  and  in  a  condition  to  furnish  to  vegetables  this  neces- 
sary element,  aids  rather  in  giving  it  porosity,  as  it  is  in 
coarse  particles.  But  still,  so  far  as  this  element  is  concerned, 
the  soil  is  well  enough ;  but  in  a  combination  or  mixture 
which  is  loose  and  porous,  it-  is  doubtful  whether  the  neces- 
sary chemical  changes  do  take  place  at  all.  considering  the 
nature  of  the  tract  of  land,  I  believe  the  first  step  to  be  taken 
towards  its  improvement  would  be  to  give  it  a  heavy  dressing 
of  clay,  to  change,  if  possible,  its  physical  condition.  Less 
clay  would  be  required,  if  one  which  is  calcareous  could  be 
employed ;  for  less  would  answer  the  purpose  than  if  it  were 
pure.  In  order  that  chemical  changes  should  take  place,  it 


&ORTH-CAROLINA   GEOLOGICAL   SURVEY,  4:7 

is  necessary  that  water  should  be  retained,  or  that  it  should 
pass  through  slowly. 

The  fertilizers  which  are  best  adapted  to  a  case  like  the 
Swansboro'  soil  are  green  crops,  peas  or  clover,  which  may 
be  ploughed  in.  By  either  crop  we  secure  in  part  the  end 
we  aim  at,  condensation  of  the  soil  or  compactness,  by  which 
water  is  retained,  and  by  which  also  time  is  given  for  the 
consummation  of  the  chemical  changes  required.  The  water 
being  retained,  the  crop,  whatever  it  may  be,  the  plant  is 
supplied  both  with  water  and  nutriment. 

But  the  necessary  dressing  of  clay  is  always  expensive, 
even  when  it  is  near  or  at  hand,  unless  indeed  it  can  be 
reached  by  the  plough.  There  are  very  few  cases  where  the 
expense  of  hauling  clay  is  ever  returned  in  an  increased 
amount  of  crops.  We  may  be  able,  as  I  believe,  to  point 
out  in  what  way  given  defects  in  a  soil  may  be  remedied. 

When  that  is  done,  it  still  remains  a  question  for  solution^ 
whether  the  mode  proposed  will  pay.  It  is  evident  that  a 
calculation  of  the  cost  of  the  mode  prescribed  is  very  impor- 
tant, if  it  is  to  be  put  in  execution.  A  garden  may  be  put 
into  a  high  state  of  fertility,  when  a  large  cornfield  cannot  be 
treated  in  the  same  mode. 

It  is  not  easy,  in  the  case  before  us,  to  account  for  the  bar- 
renness of  the  soil  of  the  coast,  unless  we  adopt  the  theory 
that  it  is  mainly  owing  to  its  mechanical  condition.  A  soil 
having  a  very  close  resemblance  to  this,  at  Cape  Cod,  in 
Massachusetts,  is  quite  fertile.  President  Hitchcock,  of  Am- 
herst  College,  who  conducted  the  geological  survey  of  the 
State,  found  on  examination  and  analysis,  that  the  blowing 
sands  of  the  cape  owed  their  productiveness  probably  to  the 
comminuted  shells,  intermixed  with  the  sand.  Or,  at  least, 
the  sands,  under  a  microscope,  exhibited  particles  of  shells ; 
and  hence,  as  the  soil  consisted  of  sand  and  finely  commi- 
nuted shells,  its  productiveness  was  attributed  to  the  presence 
of  this  fine  lime  dust  commingled  with  the  sand.  But  the 
climate  of  Massachusetts  bay  is  much  more  moist  and  cool 
during  the  summer  than  the  coast  of  Bogue  sound.  The  sun 
in  the  latter  case  acts  with  more  force  upon  vegetables  than 


48  NOKTH-CAKOLINA  GEOLOGICAL   SURVEY. 

at  the  north.  A  soil  which  might  bear  corn  in  Massachusetts 
would  not  sustain  it  on  the  coast  of  North- Carolina,  on  ac- 
count of  the  more  rapid  evaporation  of  water ;  in  conse- 
quence of  which,  a  plant  would  be  early  deprived  both  of 
water  and  nutriment,  though  it  might  be  found  in  the  medi- 
um in  which  it  had  been  growing. 


CHAPTER  IY. 

Soils  of  Jones  county,  taken  from  the  plantation  of  J.  H.  Haughton,  Esq. 
Composition  of  a  brown  earth  overlying  and  resting  upon  the  marl  beds. 
Recapitulation. 

§  27.  Several  specimens  of  soil  have  been  furnished  me  for 
analysis  from  Jones  county,  which,  as  they  may  be  employed 
to  illustrate  the  composition  of  the  cultivated  lands  in  that 
section  of  the  State,  I  Shall  give  the  results  in  this  place. 
They  were  furnished  by  John  H.  Haughton,  Esq.,  from  a 
plantation  which  he  recently  purchased.  Four  kinds  were 
forwarded,  marked  1,  2,  3,  4  respectively.  No.  1.  Color, 
brown  or  blackish  brown,  and  to  the  eye  appears  rich  in  veg- 
etable matter.  When  ignited  it  loses  readily  this  part  of  the 
soil  and  becomes  a  light  drab,  leaving  a  fine  residue  resem- 
bling that  of  the  Hyde  county  soils.  Its  appearance  shows 
that  it  is  a  silicious  soil.  One  hundred  parts  gave  me  th« 
following  proportions : 

Silex 82.300 

Peroxide  of  iron  and  alumina,  4.300 

Organic  matter,  4.500 

Lime,  0.102 

Magnesia,    : 0.02C 

Potash,   0.003 

Soda 0.001 

Water, 8.800 

Sulphuric  acid,  trace, 

Chlorine,., trace, 

100.025 


NORTH-CAROLINA   GEOLOGICAL   SURVEY.  49 

This  soil  has  evidently  been  worn  by  long  cultivation,  still 
it  has  sufficient  matter  to  sustain  moderate  crops ;  but  it  has 
reached  that  stage  which  requires  additional  applications  of 
manure.  , 

All  the  most  important  elements,  as  phosphoric  acid,  sul- 
phuric acid,  lime,  magnesia  and  potash,  are  considerably  less 
than  the  standard  soils  contain  ;  and  as  they  maintain  about 
the  usual  proportions  to  each  other,  it  is  probable  that  they 
have  been  reduced  simultaneously  by  cultivation. 

No.  2.  Color,  a  light  drab,  resembles  clay,  but  contains 
coarse  particles  of  sand,  and  hence  is  very  gritty.  This 
variety  of  soil  contains  greater  excess  of  sand,  and  is  defici- 
ent in  organic  matter,  etc.  One  hundred  grains  gave  me 

Silex,    93.000 

Peroxide  of  iron  and  alumina, 2.000 

Organic  matter,  1.300 

Lime, 0.001 

Magnesia, "0.010 

Water,   3.000 

Potash, trace, 

Soda,  trace, 

Sulphuric  acid, trace, 

99.311 

This  evidently  ranks  among  the  poorest  of  soils.  It  ap- 
pears quite  similar  to  much  of  the  poor  gallberry  lands  of  the 
eastern  part  of  the  State. 

A  larger  proportion  of  alumina  and  iron  could  have  been 
obtained  by  fusion  with  baryta  or  soda,  but  the  exhaustion  by 
boiling  with  hydrochloric  acid,  I  deemed  sufficient  for  my 
purpose,  or  the  objects  to  be  obtained  by  analysis.  This  kind 
of  soil  no  doubt  might  be  put  into  a  condition  for  raising 
wheat  by  thorough  drainage,  and  a  large  application  of 
manures. 

The  best  application  to  a  soil,  the  composition  of  which  re- 
sembles the  foregoing,  is  a  compost  of  marl  with  organic 
matters  derived  from  the  stable ;  or,  the  leaves  of  a  forest. 
In  materials  of  this  description  a  supply  of  organic  matters 
is  obtained  in  combination  with  the  phosphates  of  lime  and 
5 


50  NOKTH-CAKOLINA   GEOLOGICAL   SUUVEY. 

potash,  all  of  which  are  required  to  impart  fertility  to  a  soil 
defective  as  this  is  in  each  of  those  elements. 

No.  3.  Color,  brown,  fine  grained,  and  has  apparently  con- 
siderable vegetable  matter  in  its  composition.  It  has  no 
lumps  of  earth,  but  is  reduced  to  a  granular  state;  or  in  other 
words  it  is  pulverulent  and  light. 

One  hundred  grainsj  on  being  submitted  to  analysis,  gave 
me — 


Silex,   80.300 

Alumina  and  peroxide  of  iron, 2.550 

Lime,   0.151 

Magnesia,    0.020 

Phosphoric  acid,  trace, 

Sulphuric  acid, 0.020 

Potash, 0.001 

Soda, O.C02 

Organic  matter, 3.100 

Water, 3.000 


98.144 


The  quantity  of  organic  matter  is  less  than  its  appearance 
before  analysis  indicated,  and  this  is  often  the  case  in  the  soils 
in  the  eastern  part  of  the  State. 

Many  chemists  regard  the  organic  matter  as  of  little  im- 
portance. Experience  and  the  best  conducted  experiments, 
however,  prove  that  it  is  a  necessary  constituent  of  a  good 

soil. 

Here^  also,  the  lime  or  alkaline  earths  and  alkalies  are  defi- 
cient, at  least  to  raise  good  crops  of  maize,  or  any  of  the 
cereals.  Besides  there  is  a  great  excess  of  silex,  but  it  is  in 
a  fine  condition,  indeed  in  none  of  the  samples  is  it  ever 
coarse  ;  it,  therefore,  makes  a  better  basis  upon  which  to  work 
than  if  this  were  a  coarse  sand,  inasmuch  as  it  is  better  con- 
ditioned to  hold  or  retain  water. 

No.  4.  Color,  nearly  black,  with  organic  matter,  and  fine 
grained.  Ignition  leaves  it  of  a  drab  color. 

I  found  its  composition,  on  submitting  it  to  analysis,  to  be 
as  follows: 


IsORTH-CAKOLINA  GEOLOGICAL   SURVEY.  51 


Silex,  ..........................................  88.700 

Peroxide  of  iron  and  alumina,    ...................  3.350 

Lime,    ......................................  0.100 

Magnesia,   .......................................  0.022 

Sulphuric  acid,  .................................  0.010 

Chlorine,  .......................................  trace, 

Potash,    ....................  .'  ...................  0.048 

Soda,    ..........................................  0.010 

Organic  matter,  .................................  1.800 

Water,  .....  .....................................  5.000 

99*040 


This  specimen  of  soil  lias  a  better  composition  than  either 
of  the  four  of  this  lot.  There  is  less  silica,  more  lime  and 
potash ;  though  the  amount  of  organic  matter  and  peroxide 
of  iron  and  alumina  is  still  comparatively  small,  and  we  infer 
from  that  fact,  that  the  amount  of  phosphates  is  also  small. 

This  soil  has  no  doubt  been  under  cultivation  for  years.  It 
has  a  good  basis  to  build  upon,  as  the  silex  is  fine  and  not 
very  excessive  in  quantity.  It  is  evidently  a  better  soil  than 
No.  1,  and  does  not  rank  in  the  class  with  ~No.  2,  which  is  a 
coarse  clayey  silicious  soil,  the  particles  of  which  are  very 
coarse.  In  all  these  samples  the  cultivation  should  not  be 
carried  to  that  extent  which  would  effect  an  entire  exhaus- 
tion. 

The  remarks  upon  the  four  foregoing  soils  have  been  sug- 
gested by  the  analyses  and  their  physical  properties.  ~No 
information  has  been  obtained  respecting  the  treatment  to 
which  they  have  been  subjected. 

§  28.  A  soil  of  a  somewhat  remarkable  appearance,  and 
having  a  good  composition,  is  spread  over  large  portions  of 
the  eastern  counties.  It  is  not  always  a  surface  soil ;  indeed 
it  is  rather  rare  to  meet  with  it  under  cultivation.  It  occupies 
a  distinct  position  in  the  series  of  soils,  and  is  really  one  of 
the  deposits  which  is  alwa}rs  associated  with  the  marl  beds. 
It  cannot,  with  propriety,  be  regarded  as  a  marl,  though 
under  favorable  circumstances  it  may  be  used  as  a  fertilizer. 

It  has  a  brown  color,  and  when  wet  is  as  tenacious  as  the 
ordinary  clays,  though  it  has  less  alumina  in  its  composition ; 
it  is  very  adhesive  to  the  shoe  or  boot,  and  if  it  is  ever  profit- 


52  NOKTH-CAROLINA   GEOLOGICAL   SURVEY. 

able  to  haul  clay  for  fertilizing  the  sandy  soils,  this  is  especial- 
ly adapted  to  the  fulfilment  of  all  the  ends  which  may  be  ob- 
tained by  the  use  of  clay. 

It  rests  upon  the  shell  marl  in  some  places,  and  in  others 
upon  the  eocene  marl.  The  circumstances  attending  its  de- 
position were  peculiar.  It  appears  to  have  been  deposited 
immediately  after  a  period  of  denudation,  as  it  rests  not  only 
upon  the  marl,  but  extends  into,  and  fills  deep  channels  which 
had  been  cut  out  of  the  marl  during  the  period  alluded  to. 
Hence  it  appears  to  send  down  long  tapering  columns  which 
extend  sometimes  to  a  point  near  the  bottom  of  the  bed. 
This  formation,  however,  was  formed  from  quiet  waters,  as 
there  is  no  evidence  of  a  rush  or  violent  flow  of  waters,  by 
the  presence  of  large  rocks,  or  even  coarse  pebbles.  It  has 
some  coarse  sand  intermixed  with  pebbles.  It  has  the  appear- 
ance of  a  sediment,  which  was  probably  derived  from  the 
decomposing  slates  and  granite,  which  lie  beneath  the  terti- 
ary, but  which  is  now  concealed,  except  in  a  few  isolated 
places. 

On  submitting  this  soil  to  analysis  I  found  it  composed  of 

Silex,    ... 77.850 

Alumina  and  peroxide  of  iron,    10.107 

Lime,   2.000 

Megneisa,    1.810 

Organic  matter,  3.950 

Water,  5.750 

Sulphuric  acid,  0.010 

Chlorine,  0.010 

Potash, 0.185 

Soda,  0.345 

Soluble  silica,    0.100 

'.,,/"'  99,815 

This  soil  is  rich  in  lime,  which  is  in  part  derived  from  a  few 
small  fragments  of  shell  which  it  contains,  but  it  efferveses 
but  slightly,  and  hence  it  is  probable  the  lime  is  diffused 
rather  uniformly  through  the  mass.  When  this  mass  lies 
immediately  beneath  the  sandy  soil,  and  within  reach  of 
the  plough,  it  would  improve  it  very  much  to  commingle  it 
with  the  surface  material,  and  it  need  not  be  rejected  in  load- 


NORTH-CAROLINA   GEOLOGICAL   SURVEY.  53 

ing  marl  at  the  pit,  inasmuch  as  its  composition  shows  that  it 
is  an  important  improver  of  the  common  sandy  soil  so  preva- 
lent in  the  eastern  counties. 

The  phosphoric  acid  remains  to  be  determined.  In  itself 
this  soil  has  a  composition  admirably  adapted  to  the  growth 
of  wheat,  or  indeed  cotton.  It  contains  also  a  large  amount 
of  potash. 

It  was  taken  from  a  mass  which  overlies  the  eocene  marl 
of  the  plantation  of  Sam'l  Biddle,  Esq.,  of  Craven  county. 
It  is,  however,  found  on  the  Cape  Fear,  resting  upon  the  shell 
marl,  a  more  recent  deposit,  and  may  be  found  on  the  plant- 
ation of  Dr.  .Robinson,  of  Elizabethtown. 


RECAPITULATION  OF  THE  LEADING  FACTS   RESPECTING  THE  SOILS  OF 
THE  EASTERN  COUNTIES  OF  NORTH-CAROLINA. 

§  29.  (1.)  The  soils  of  the  eastern  counties,  without  excep- 
tion, are  marine  formations,  being  deposited  from  water,  and 
are  truly  sediments.  They  are  therefore  in  their  origin  un- 
like those  of  the  middle  and  western  counties,  inasmuch  as 
the  latter  are  the  products  of  slow  decomposition,  and  are  in 
situ,  or.  occupy  the  place  upon  the  rocks  from  which  they  are 
derived. 

The  eastern  soils  have,  on  the  contrary,  been  transported, 
or  wTere  first  the  products  of  a  disintegration  and,  afterwards, 
transported  from  the  places  from  whence  they  were  derived. 
As  they  are  frequently  composed  of  one  or,  at  most,  two  ma- 
terials which  can  be  distinguished  by  the  naked  eye,  it  is  im- 
possible to  determine  the  source  from  whence  they  came. 
They  were  probably  derived,  however,  from  the  granite 
which  borders  the  tertiary  formation  upon  the  west.  Their 
distinguishing  features  are  siliceous ;  and  it  seems  that  most 
of  the  aluminous  compounds,  as  felspar  and  certain  slates, 
were  finely  comminuted,  and  were  transported  to  distant 
points,  leaving  the  heavy  and  coarser  materials  in  the  bays 
which  jut  up  from  the  ocean  in  the  depressions  of  the  land. 


54:  NOKTH-CAKOLINA  GEOLOGICAL   SURVEY. 

These  sandy  deposits  were  not  laid  down  at  one  period, 
though  they  are  comparatively  modern.  They  alternate  with 
a  few  beds  of  clay,  but  there  is  but  one  near  the  surface 
which  is  extensively  distributed.  The  last  of  the  marine  de- 
posits was  mostly  a  pure  white  sand  ;  and  it  not  unfrequently 
washes  white  when  it  is  deprived  of  its  vegetable  coating. 
The  last  or  most  recent  bed  of  sand,  is  formed  by  waves  of 
the  ocean  into  swells  or  undulations.  A  belt  thus  thrown  up 
and  moulded  by  this  agency,  extends  obliquely  across  the 
country.  One  of  the  most  distinguished  features  of  this  belt 
is  intersected  by  the  Wilmington  railway,  at  Everettsville, 
ten  miles  S.  "W.  from  Goldsborough.  These  swells  of  sand 
are  sufficiently  large  and  extensive  to  give  origin  to  perma- 
nent mill-streams.  They  seem  to  have  been  derived  from 
the  Atlantic  side,  and  to  have  been  cast  up  by  waves  which 
in  their  operation  have  denuded  all  the  eastern  portions  lying 
between  this  belt  and  the  Atlantic  ocean,  and  hence  it  not 
unfrequently  happens  that  the  upper  stratum  of  sediment  is  a 
stiff  clay. 

(2.)  The  denuded  clay  is  often  a  stiff  brick  clay,  and  is 
about  four  feet  thick.  Shallow  depressions  are  hollowed  out 
of  it,  which  are  always  the  receptacles  of  water,  and  have 
also  favored  the  growth  of  moss  and  small  vegetables.  To 
the  growth  of  these  humble  plants  we  attribute  the  origin  of 
the  vegetable  matter  which  is  so  extensively  prevalent  in 
many  of  the  eastern  counties,  and  which  are  known  by  the 
names  ofpocosin  and  swamp  lands. 

(3.)  A  slight  elevatory  movement  of  the  whole  coast  of 
North-Carolina,  has  reclaimed  those  tracts  from  water ;  and, 
though  not  dry  yet,  they  are  not  submerged,  and  are  no  long- 
er the  recipients  of  sediment. 

While  these  lands  were  but  half  reclaimed  from  the  do- 
minion of  water,  they  were  subjected  to  inundations  which 
transported  fine  silt,  and  which  required  much  time  to  settle. 
This  fine  silt,  or  mud,  is  now  the  soil  which  is  so  productive 
in  corn  in  Hyde  county  and  other  parts  of  the  Atlantic 
border. 

This  singular  soil  is  characterized   by  its  vegetable  matter, 


NORTH-CAROLINA   GEOLOGICAL   SURVEY.  55 

and  by  the  extreme  fineness  of  its  inorganic  matter ;  and  the 
two  compound  elements  are  well'  fitted  to  each  other,  and 
admirably  adapted  to  the  growth  of  maize  in  this  climate, 
whereas  in  a  northern  climate  it  is  very  doubtful  whether  the 
same  results  could  be  obtained.  In  Canada  East  there  are 
somewhat  similar  soils,  but  they  are  treated  quite  differently 
in  order  to  bring  the  soil  under  cultivation.  There,  the  sur- 
face is  first  burned,  and  the  ash  and  debris  remaining  sup- 
plies the  nutriment  for  a  succession  of  heavy  crops.  When 
this  first  fertilizing  matter,  obtained  by  burning,  is  exhausted, 
it  is  subjected  to  the  same  treatment  again,  and  again  put 
under  cultivation.  The  lands  of  the  eastern  counties  would 
not  bear  this  mode  of  cultivation ;  neither  do  they  require  it. 
The}''  become  productive  by  draining. 

§  30.  The  composition  of  the  soil  of  Canada  East,  taken 
from  a  tract  which  is  there  known  by  the  name  of  Savanna 
of  St.  Dominique,  is  composed,  according  to  Mr.  Hunt,  of 

Fixed  carbon, 29.57 

Ashes,    6.75 

Volatile  matter, 63.68 

The  ash  or  inorganic  matter  in  100  parts  contained : 

Carb.  Lime,   52.410 

Lime""}  as  silicates,    1(U3° 

Magnesia,  '  3.150 

Peroxide  of  iron,  4.680 

Alumina,  2.440 

Oxide  of  magnesia,  0.040 

Phosphate  of  lime,  2.019 

Sulphate  of  lime,   15.085 

Sulphate  of  potash,   0.605 

Sulphate  of  soda,   0.076 

Chloride  of  Sodium,  0.412 

Silica, 4.920 

Sand,    4.040 

100.308 

In  the  foregoing  analysis  we  can  readily  perceive  that  the 
material  subjected  to  this  process  is  an  ash,  with  only  faint 
traces  of  soil,  but  in  appearance  the  North-Carolina  pocosin 


56  NORTH-CAROLINA   GEOLOGICAL   SURVEY. 

lands  resemble  the  turf  or  peat  soils  of  Canada  and  New 
York,  but  the  better  kinds  or  those  of  Hyde,  contain,  inter- 
mixed with  the  vegetable  matter,  fine  earth,  which  gives 
them  a  substantial  body.  In  this  respect  they  differ  from  the 
peaty  or  turf  soils  of  other  places.  They  differ  also  in  en- 
durance. They  continue  productive  through  several  genera- 
tions. Those  of  Hyde  have  been  tilled  through  three  genera- 
tions, and  the  fourth  has  them  under  culture.  I  attribute 
this  extended  period  of  endurance  to  the  temperature  which 
the  soil  enjoys.  Below,  in  immediate  proximity  to  the  roots 
of  corn,  the  water  remains  through  the  season.  Hence  there 
is  a  temperature  preserved  which  is  only  moderately  high  in 
the  midst  of  summer,  in  consequence  of  evaporation.  Even 
the  water  often  surrounds  the  hill  of  corn,  and  remains  on 
the  surface  for  a  long  time,  without  injuring  the  growth  of 
the  plant.  The  external  heat  is  sufficient  for  the  crop.  If  it 
were  higher  it  would  slowly  consume  the  vegetable  matter. 
Besides,  the  low  temperature  of  these  peculiar  soils,  the 
proximity  to  the  ocean,  favors  a  constantly  moist  climate,  or 
atmosphere;  and  hence,  through  the  influence  of  water  be- 
neath, and  a  moist  atmosphere  above,  the  growth  of  veget- 
ables is  promoted. 

In  the  midland  counties  the  vegetable  matter  is  consumed, 
or  so  nearly  consumed  that  the  blackened  belt  at  the  surface 
is  never  formed.  Upon  the  mountains,  the  whole  of  the  blue 
ridge,  vegetable  matter  accumulates  in  the  soil.  The  heat  is 
insufficient  to  destroy  it,  while  in  the  midland  counties  it 
never  accumulates  even  in  forests,  and  though  there  is  a  large 
annual  addition  of  vegetable  matter  from  the  leaves  which 
fall  in  autumn  and  winter,  still  no  accumulation  takes  place 
in  the  soil.  It  is  literally  consumed. 

§  31.  The  pocosin  and  swamp  lands  present  a  great  variety 
in  the  proportions  of  vegetable  matter  present  in  the  soil. 
Some  passing  to  the  extreme  limit,  from  10  to  93  per  cent,  of 
organic  substance.  The  latter  percentage  is  near  the  boun- 
dary which  limits  the  capability  of  growing  the  cereals.  A 
greater  excess  of  vegetable  matter  scarcely  admits  of  the 
continued  growth  until  the  crop  ripens,  it  soon  ceases  to  grow, 


NOETH-CAROLHSTA   GEOLOGICAL   SURVEY.  57 

becomes  yellow  after  it  has  appeared  above  the  ground  when 
it  has  reached  the  height  of  10  or  12  inches.  The  most 
valuable  swamp  and  pocosin  lands  lie  in  Hyde,  Beaufort, 
Jones,  Onslow  and  Brunswick  counties ;  those  of  Hyde  have 
been  steadily  cultivated  for  more  than  one  hundred  years 
without  manures,  and  still  the  crops  are  equally  as  good  as 
when  first  planted.  Hundreds  of  square  miles  of  the  most 
valuable  of  these  lands  still  remain  unsubdued.  It  may  be 
inferred  that,  as  these  swamp  lands  are  so  low  and  wet,  that 
they  must  necessarily  be  extremely  unhealthy,  or  become  so 
when  drained  and  the  vegetable  matter  begins  to  decompose. 
Experience,  however,  does  not  support  this  view.  The  testi- 
mony of  those  who  have  cultivated  them  for  forty  years  is, 
that  their  families  have  enjoyed  as  much  health  as  their 
neighbors  who  have  lived  at  a  distance.  Persons  who  are  in 
the  habit  of  plunging  into  the  swamp  lands  knee  deep  for 
draining,  and  when  drained  to  live  in  the  immediate  vicinity 
of  the  extended  surface  of  black  vegetable  mould  for  years, 
are  rarely  sick  with  fevers.  The  points  which  are  unhealthy 
are  those  which  are  exposed  to  winds  which  blow  over  ex- 
tended surfaces  of  the  waters  of  the  IsTeuse  or  Cape  Fear 
rivers.  Miasm,  which  generates  fever,  arises  more  'from  the 
banks  of  rivers  than  from  the  swamp  and  pocosin  soils. 

§  32.  The  soil  which  is  known  as  the  gallberry  soil  is  not  of 
a  uniform  composition  or  appearance ;  one  of  the  most  com- 
mon kinds  is  formed  of  sand,  intermixed  with  black  vegetable 
matter.  On  exposure  to  rains  by  the  road-side,  or  where 
ditches  are  cut  through  it  so  as  to  expose  a  section  one  or  two 
feet  thick,  it  has  a  grayish  look  from  the  presence  of  the  white 
marine  sand  which  is  exposed  by  washing.  A  microscope 
shows  at  once  the  naked  sand.  A  soil  of  this  description,  and 
which  is  widely  spread  over  the  flat  low  grounds  of  the  mid- 
dle section  of  the  eastern  counties,  I  submitted  to  careful 
analysis  for  the  purpose  of  determining  the  amount  of  avail- 
able material  which  it  contains.  It  was  taken  from  the  plant- 
ation of  Mr.  Lane,  of  Craven  county,  but  is  a  fair  representa- 
tion of  the  soil  of  the  Dover  pocosin.  It  contained : 


M  NORTH-CAROLINA   GEOLOGICAL   SURVEY. 

Sand  or  silex,  70.50 

Organic  matter,   25.20 

Peroxide  of  iron  and  alumina,  0.76 

Lime 0.01 

Magnesia, trace, 

Water, 2.70 

Soluble  silica, „ trace. 

The  silex  is  a  perfectly  white  marine  sand. 

Although  this  analysis  is  not  carried  through,  yet  it  is  evi- 
dent that  the  available  matter  for  crops  is  extremely  small. 
The  seventy-six  hunclredths  of  a  grain  of  peroxide  of  iron 
and  alumina  is  too  small  a  quantity  to  have  much  chemical 
or  mechanical  influence  upon  the  organic  matter  with  which 
it  is  mixed;  neither  can  it  furnish  phosphoric  acid  to  supply 
the  wants  of  vegetation  if  put  under  cultivation.  This  variety 
of  gallberry  land  belongs  to  the  poorest  class  of  soils.  It  is  not 
expected  it  would  pay  a  profit  if  cleared,  and  hence  all  such 
lands  should  remain  wild,  or  in  their  natural  state. . 

Another  variety  of  low  ground  soil  is  of  a  better  quality, 
though  still  it  ranks  low  for  the  purposes  of  agriculture.  It 
is  of  a  light  color,  and  hence  contains  much  less  vegetable 
matter.  It  is  a  marine  sand,  intermixed  with  a  small  quanti- 
ty of  clay,  a  portion  of  which  can  be  dissolved  in  hydrochloric 
acid.  This  soil  is  from  Sampson  county.  It  forms  extensive 
areas  in  Johnston,  Sampson  and  Duplin  counties.  There  is, 
however,  an  improvement  in  the  character  of  the  low  grounds 
towards  the  east  from  Johnston  county.  The  color  of  this 
soil  is  a  light  brownish  or  purplish  drab  ;  in  drying  it  becomes 
hard  and  loses  most  of  its  water  of  absorption.  It  resembles 
the  green  swamp  soil  in  Brunswick  county.  It  is  composed 
of 

Silex,    1 88.40 

Peroxide  of  iron  and  alumina,  2.92 

Lime, 0.02 

Magnesia,    0.03 

Water,   3.09 

Organic  matter, 4.20 

Potash  and  soda,  traces, 

Phosphoric  acid,  undetermined,  


NORTH-CAROLINA   GEOLOGICAL   SURVEY.  59 

In  this  variety  of  soil  from  the  swampy  grounds  there  is 
still  a  deficiency  of  the  alkalies  and  alkaline  earths ;  this, 
however,  may  be  cultivated  with  medium  results,  if  marl  is 
at  hand  from  which  to  supply  the  deficient  matter. 


CHAPTER  Y. 

FERTILIZERS. 

What  constitutes  a  Fertilizer. — Sources  of  Fertilizers. — Those  from  the 
Vegetable  kingdom  are  the  Ash. — Ash  of  difFernt  Vegetables  — Ash  of 
Plants  resembles  in  composition  the  Inorganic  Matter  of  Soils. — Quantity 
of  Fertilizing  Matter  removed  from  the  Soil  by  different  Plants. — Me- 
thods to  be  adopted  by  which  a  Waste  of  Fertilizing  Matter  may  be  Pre- 
vented.— Fertilizing  Matter  Restored  by  Plowing  in  Green  Crops, 

§  33.  Any  substance  in  husbandry  is  a  fertilizer  which  im- 
proves the  soil.  They  are  numerous  and  are  derived  from 
numerous  sources.  The  air  is  a  reservoir  of  substances  which 
improve  the  soil,  and  water  is  the  medium  of  communication. 
As  in  the  laboratory  substances  do  not  act  upon  each  other 
unless  one  or  both  are  in  a  fluid  condition  ;  so  fertilizers  must 
be  in  solution  in  a  menstrnm,  of  which  water,  in  the  kingdom 
of  nature,  is  the  universal  solvent  The  air  contains  ammo- 
nia and  carbonic  acid.  These  are  the  most  direct  fertilizers. 
They  are  both  transferable  agents,  passing  from  the  atmos- 
phere to  the  earth  dissolved  in  rain  water,  and  escaping  up- 
ward from  the  earth  in  the  ascending  vapors,  when  they 
have  fulfilled  their  mission  to  the  grown  and  perfect  vege- 
table. They  escape  when  it  decays,  and  wait  for  another 
mission  to  the  earth  or  soil.  The  interchange  is  almost  per- 
petual. There  are  vegetables  at  all  times  undergoing  decay, 
or  [eremacausis,~]  a  slow  combustion,  during  which  the  com- 
pound atoms  are  undergoing  a  change,  and  each  one  of  which 


60 


NORTH-CAROLINA   GEOLOGICAL   SURVEY. 


is  finally  resolved  into  new  forms  and  conditions.  Ammonia 
and  carbonic  acid  are  the  common  products  of  change  in  all 
these  cases.  Both  are,  however,  compound  bodies.  The 
first  is  a  body  recognized  by  its  extremely  pungent  smell, 
and  commonly  known  as  hartshorn,  and  is  formed  by  the 
union  of  two  elements — nitrogen  and  hydrogen.  The  latter 
is  the  lightest  substance  known — it  is  .069,  the  weight  of  air. 
Carbonic  acid  is  an  air,  also,  or  gas,  and  is  heavier  than  at- 
mospheric air,  and  hence  is  sometimes  found  in  depressed 
places,  not,  as  is  usually  maintained,  by  falling  down  from 
the  atmosphere  in  consequence  of  its  greater  weight,  but  by 
its  escape  from  beneath,  or  from  the  soil  or  fissures  of  rocks. 
Rain  water  and  snow  hold  both  ammonia  and  carbonic  acid 
in  solution,  and  hence,  as  has  been  remarked,  they  are  the 
media  from  which  growing  plants  derive  these  important 
fertilizers.  jSnow,  particularly,  is  rich  in  ammonia.  From 
this  material  it  may  be  obtained  by  evaporation.  To  this 
substance,  probably,  the  beautiful  greenness  of  vegetation  is 
due,  which  appears  on  the  melting  of  a  March  snow. 

These  two  substances,  however,  may  be  derived  from  any 
organic  matter  in  the  earth,  when  it  is  undergoing  decay ; 
hence,  most  if  not  all  bodies  which  have  lived  may  furnish 
them  if  buried  in  the  soil  and  within  reach  of  the  roots  of  a 
growing  plant.  There  are,  therefore,  two  modes  by  which 
these  fertilizers  become  subservient  to  nutrition — 1,  by  wrater 
falling  from  the  atmosphere  and,  2,  by  water  in  the  soil  which 
dissolves  them  out  from  particles  of  earth  and  organic  matter. 

In  the  application  of  the  first  mode,  husbandry  has  nothing 
to  do.  It  is  a  part  of  the  machinery  of  nature,  by  which  she 
maintains  the  balance  between  the  vegetable,  animal  and 
mineral  kingdoms.  This  machinery  in  its  workings  is  per- 
fectly competent  to  preserve  this  balance,  to  furnish  food  and 
sustain  in  perpetual  existence  all  the  species  which  belong  to 
the  present  system.  In  a  temperate  climate,  however,  with- 
out artificial  aid,  the  cereals  would  cease  to  grow,  or  yield  the 
harvests  they  nowT  do,  because  of  the  exhaustion  they  bring 
about  in  the  progress  of  time  and  of  cultivation. 

§  34.  Fertilizers  may  be  divided  into  kinds  according  to 


NOKTH-CAKOLINA   GEOLOGICAL   STJKVEY.  61 

the  source  from  whence  they  are  derived,  as  those  which  be- 
long to  the  three  kingdoms  of  nature,  the  mineral,  vegetable 
and  animal,  but  such  a  division  is  really  of  small  importance, 
inasmuch  as  it  will  be  perceived  from  the  foregoing  remarks 
that  all  fertilizers  may  be  traced  back  to  the  mineral  kingdom, 
even  ammonia  is  strictly  a  mineral,  although  it  abounds  in 
both  the  vegetable  and  animal  kingdoms  in  certain  combina- 
tions. Proximately,  they  are  either  animal  or  vegetable ;  but 
in  either  case  they  are  of  a  mineral  origin.  The  fertilizers 
which  will  come  up  for  examination  are  ashes,  marls,  excre- 
ments of  animals  and  green  crops. 

§  35.  It  needs  no  argument  to  prove  the  value  of  ashes  as 
fertilizers,  we 'have  only  to  inspect  the  foregoing  tables  of  the 
composition  of  the  ashes  of  wheat,  maize,  oats  and  potatoes. 
The  composition  of  the  ashes  of  forest  trees  brings  us  to  the 
same  results,  and  as  much  dependence  is  placed  upon  the 
decomposition  of  the  standing  trees  in  the  cultivated  fields  it 
is  important  that  the  fertilizers  thus  obtained  may  be  shown. 
We  are  obliged,  in  this  case,  to  resort  to  the  analyses  of  the 
ash  obtained  directly  by  combustion.  The  results,  however, 
are  the  same  in  the  natural  process  of  decay  as  by  combustion, 
and  the  decayed  bark,  limbs  and 'twigs  furnish  ultimately 
what  they  would  have  furnished  were  they  consumed  by 
fire. 

The  white  oak,  for  example,  quercus  alba,  furnishes  by 
cumbustion  an  ash  composed  of  the  following  elements.  First 
the  bark  of  the  trunk,  which  contains : 

Potash,  0.25 

Soda,    2.57 

Sodium,  0.08 

Chlorine, 0.12 

Sulphuric  acid,  0.03 

Phosphates  of  lime  and  magnesia,  10.10 

Carbonic  acid,  29.00 

Lime,  54.89 

Magnesia,  0.20 

Silica,    0.25 

Soluble  silica,  0.25 

Organic  matter,  1.16 


02  NOETH-OAEOLINA  GEOLOGICAL  SURVEY. 

The  bark  of  the  twigs  gave  me,  on  submitting  the  asli  to 
analysis : 

Potash,  1.27 

Soda,  4.13 

Chlorine,  0.13 

Sulphuric  acid,  trace, 

Phosphates  of  lime,  magnesia  and  peroxide  of  iron,  14.15 

Carbonic  acid,  - 30.33 

Lime,    47.72 

Magnesia,    '0.20 

Silica,  0.65 

Soluble   silica, 0.55 

Organic  matter, 1.52 

100.09 

The  wood  of  the  twigs  decays  with  the  bark,  but  the  wood, 
as  will  be  seen,  is  richer  in  fertilizing  matter  than  the  bark. 
It  has  the  following  elements : 

Potash,    9.74 

Soda,    6.89 

Sodium,  0.16 

Chlorine, 0.25 

Phosphates  of  lime,  magnesia  and  peroxide  of  iron,  23.60 

Carbonic   acid,  17.45 

Lime,    34.10 

Magnesia,  0.50 

Silica,  0.55 

Soluble   silica,  0.60 

Organic  matter, 5.90 


The  outside  wood  slowly  decays  beneath  the  bark,  or  afber 
it  has  fallen  and  furnishes  an  ash  rich  in  potash  and  the  phos- 
phates of  lime,  magnesia,  etc.  While  standing  the  process  ie 
certainly  very  slow,  but  it  will  ultimately  be  reduced  to  a 
substance  equivalent  to  an  ash  having  the  following  composi- 
tion, viz: 

Potash,   13.41 

Soda, 0.62 

Sodium,  2,78 


NOETH-CAKOLINA  GEOLOGICAL  SURVEY.  63 

Chlorine,  4.24 

Sulphuric  acid,    0.12 

Phosphates  of  lime,  magnesia  and  iron, 32.25 

Carbonic  acid, 8.95 

Lime,    30.85 

Magnesia,  0.36 

Silica,   0.21 

Soluble  silica, 0.80 

Organic  matter,  5.70 

100.18 

The  pine  tree  gives  an  ash  on  combustion  differing  slightlj 
from  the  foregoing,  viz : 

BARK. 

Potash,  2.86 

Soda,  3.17 

Chloride  sodium,  0.03 

Sulphuric  acid, 3.48 

Carbonic  acid, 24.33 

Lime,    31.48 

Magnesia,  0.01 

Phosphate  of  lime,  magnesia  and  peroxide  of  iron,  22.12 

Organic  matter,  3.58 

Silica,  13.40 

The  most  important  addition  which  the  bark  of  this  species 
of  pine  will  add  to  the  soil  is  soluble  silica  and  lime,  the  alka- 
lies are  comparatively  unimportant. 

§  36.  The  benefit  which  has  been  attributed  to  the  stand- 
ing dead  trees  is  not  probably  due  entirely  to  the  ash  which 
the  bark  and  limbs  furnish.  A  more  important  effect  may 
be  obtained  by  the  moisture  which  is  retained  by  the  spread- 
ing roots  in  the  soil,  each  of  which  must  absorb  considerable 
water  and  retain  it  for  a  long  time.  The  practice  adopted  in 
this  particular  is  better  adapted  to  a  warm  than  a  colder 
climate.  The  shade  even  of  the  trunks  of  forest  trees  would 
be  detrimental  to  the  maize  crop  in  JSTew  England  or  New 
York,  more,  as  I  believe,  than  all  the  benefits  to  be  expected 
either  from  its  decaying  wood  or  the  increased  water  in  the 
soil. 


t>*  NORTH-CAROLINA   GEOLOGICAL   SURVEY. 

The  leaves  of  forest  trees  are  richer  in  the  phosphates  than 
the  bark  or  wood. 

In  the  fruit  these  elements  exist  in  still  greater  proportion. 
In  the  leaves  of  the  Catawba  grape  I  found  them  to  exist  in 
the  following  proportions : 

Potash,  13.394 

Soda,  9.698 

Phosphates  of  lime  and  magnesia,  32.950 

Lime,  4.391 

Magnesia,    1.740 

Chlorine 0.740 

Sulphuric  acid,  2.620 

Silica,    29.650 

Carbonic  acid,  :. 3.050 

99.026 

The  fruit  of  the  common  black  butternut  is  composed  of 

BIND.  SHELL. 

Potash 41.43  47.00 

Soda,  7.12  10.21 

Earthy  phosphates, 15.60  18.50 

Lime, 23.75  5.60 

Magnesia,  1.55  0.10 

Chlorine,  1.50  2.15 

Silica,  1.36  0.40 

Sulphuric  acid,  2.65  9,84 

Org'ic  matter  and  alkaline  phosphates,     2.30  5.40 

§  37.  The  oat  plant  furnishes  similar  facts.  The  dry  crop 
in  the  grain  weighs  975  Ibs.  per  acre,  and  furnishes  39  Ibs.  of 
ash,  with  a  percentage  of  4.00.  The  elements,  per  acre, 
are: 

Phosphoric  acid,    6.00 

Sulphuric  acid,  0.40 

Chlorine,   0.20 

Lime,  .. 12.00 

Magnesia,  7 3.00 

Potash  and  soda,  5.00 

Silica,  21.00 

Oxide  of  iron,  0  60 


HDRTH-CAEOLINA   GEOLOGICAL   SURVEY.  6'5» 

In-  the  straw,  per  acre,  the  proportion  of  elements  is : 

Phosphoric  acid, 1-50 

Sulphuric  acid,   2.50 

Chlorine, 3.00 

Lime, 5.00 

Magnesia, 15.00 

Potash  and  soda, 17.00 

Silica, 24.00 

Oxide  of  iron, 1.00 

* 

§  38.  The  clover  plant  weighs,  when  dry,  3693  Ibs.  per  acre. 
The  percentage  of  ash  is  7.70,  which  is  quite  large,  and  the 
weight  of  the  ash,  per  acre,  284  Ibs.  It  contains,  of 

Phosphoric  acid,  18.00 

Sulphuric  acid,  7.00 

Chlorine, 7.00 

Lime 70.00 

Magneisa, 18.00 

Potash  and  soda, 77.00 

Silica, 15.00 

Oxide  of  iron,  0.90 

The  clover  plant,  it  will  be  perceived,  contains  about  equal 
proportions  of  lime,  potash  and  soda ;  the  limey  however,  is 
in  excess,  but  its  composition  shows  why  it  is  so  well  adapted 
as  a  fertilizer  to  the  wheat  crop.  The  vigorous  growth  of 
clover  upon  a  soil  which  has  been  marled  \vith  green  sand, 
which  contains  both  lime  and  potash,  illustrates  and  places  in 
a  strong  light  the  advantages  of  special  fertilizers. 

If  the  ash  of  the  foregoing,  or  any  other  plant  is  compared 
with  the  composition  of  the  best  soils,  or  marls,  it  will  not 
fail  to  strike  almost  any  one  that  there  is  a  close  resemblance 
between  them.  The  soil  furnishes  phosphoric  acid,  iron,  sul- 
phuric acid,  chlorine,  magnesia,  silica,  potash  and  soda.  All 
the  remarkable-  fertilizers  contain  the  same  elements.  Those 
which  are  the  most  striking  in  their  effects  contain  lime,  phos- 
phoric acid,  potash  and  soda  in  large  proportions,  furnishing 
thereby  the  expensive  elements,  the  most  essential  ones,  or 
those  which  exist  in  the  soil  in  the  smallest  proportions,  in. 
6 


*>  NORTH-CAKOLINA   GEOLOGICAL   STJKVEY. 

great  abundance.  The  effects  of  a  fertilizer  are  the  most  per- 
ceptible where  these  are  the  most  abundant.  Hence  guano 
which  contains  a  large  amount  of  phosphoric  acid,  ammonia 
and  lime,  rarely  fails  to  satisfy  the  wants  of  the  plant  and  to 
become  the  Efficient  means  of  producing  a  greatly  increased 
crop.  Of  certain  elements  it  may  be  said  there  is  never  a 
deficiency.  Silica  is  one,  as  it  is  always  present  in  the  largest 
proportion.  The  same  may  be  said  of  iron ;  but  lime,  mag- 
nesia, and  especially  the  alkalies,  are  frequently  wanting,  if 
not  altogether,  yet  not  in  a  sufficient  quantity  to  supply  the 
wants  of  vegetation.  Hence,  in  fertilizers,  the  test  of  their 
value  consists  in  determining  the  quantity  of  lime,  potash 
and  phosphoric  acid,  whiclrthey  contain;  or,  the  amount  of 
those  special  elements  which  are  always  in  the  smallest  pro- 
portion m  the  soil ;  and  hence  too  it  is  easy  to  perceive  why 
soils  become  barren  by  cultivation,  as  those  elements  are  early 
removed  in  the  crops  which  the  soil  has  borne. 

§  39.  To  illustrate  this  point  and  make  it  sufficiently  clear 
to  be  comprehended  by  every  reader,  I  propose  to  state  the 
quantity  of  nutriment  which  several  of  our  most  important 
plants  consume;  and  which  is  derived  directly  from  the 
soil. 

In  order  to  do  this  it  is  necessarry  to  ascertain  what  ele- 
ments exist  in  the  plant,  and  which  must  of  necessity  be  taken 
from  the  soil  in  which  it  grows.  These  elements  are  obtained 
when  a  plant  is  burned.  The  residue  of  the  combustion  are 
earths,  intermixed  with  alkalies,  the  mass  of  which  is  known 
as  ashes ;  wheat,  oats,  potatoe  and  clover,  will  furnish  striking 
examples  suitable  for  the  illustration  of  the  point  in  question. 

An  ordinary  wheat  crop,  according  to  Bousingault,  when 
dried,  weighs,  upon  an  average,  in  grain,  1052  Ibs. ;  in  straw 
¥558  Ibs.,  and  the  grain  furnishes  2.40  per  cent  of  ash,  and 
the  straw  7.00.  The  quantity  of  ash  per  acre,  in  the,  grain 
amounts  to  25  Ibs.,  in  the  straw  per  acre  is  179  Ibs. 

The  proportion  ot  the  elements  contained  in  the  25  Ibs.  of 
ash  are : 


NORTH-CAKOLINA   GEOLOGICAL   SUEVEY.  67 


LB8. 

Phosphoric  acid,  12.00 

Sulphuric  acid, 0.30 

Chlorine,  trace, 

Lime,   0.80 

Magnesia,  ..  4.00 

Potash  and  soda,  , 7.00 

Silica, 0.04 

Oxideof  iron, 0.00 


In  the  straw  the  proportions  are : 


Phosphoric  acid, 5.00 

Sulphuric  acid, 1.50 

Chlorine, 1.00 

Lime, 15.00 

Magnesia, 9.00 

Potash  and  soda, 17.00 

Silica,  121.00 

Oxide  of  iron, 1.75 


One  remark  may  be  made  in  this  place,  that  the  phosphoric 
acid  of  the  grain  greatly  exceeds  that  of  the  straw ;  while 
the  lime  of  the  straw  is  in  much  greater  proportion  than  it  is 
in  the  grain,  and  the  silica  is  reduced  in  the  grain  to  the 
smallest  percentage,  but  greatly  abounds  in  the  straw.  We 
have  in  this,  as  in  many  other  instances,  the  exercise  of  a 
species  of  elective  affinity,  by  which  the  elements  select  their 
appropriate  organic  materials. 

A  potatoe  crop,  when  dried,  weighs,  in  tubers,  2828  Ibs., 
and  gives,  in  ashes,  4  per  cent.,  and  weighs  113  Ibs.  per  acre. 
The  percentage  of  composition  is : 


Phosphoric  acid,   13.00 

Sulphuric  acid, 8.00 

Chlorine, „ 3.00 

Lime,   2.00 

Magnesia,    6.00 

Potash  and  soda,  58.00 

Silica, 6.00 

Oxide  ofiron, 17.00 

The  percentage  in  tops,  5042  Ibs.,  with  6  per  cent,  of  ash. 


68  NORTH-CAROLINA   GEOLOGICAL   SURVEY; 

and  weighing  303  Ibs.  per  acre.     The  percentage  of  composi- 
tion is : 

LB8. 

Phosphoric  acid, ... 33.00 

Sulphuric  acid,    7.00 

Chlorine, 4.00 

Lime,   , 7.00 

Magnesia, 5.0C 

Potash  and  soda, 135.00 

Silica, 39.00 

Oxideof  iron,  36.01 

The  potatoe  plant  abounds  in  the  oxide  of  iron  and  pqtash, 
and  there  is  no  doubt  the  character  of  the  soil  influences  to  a 
considerable  extent  the  quality  of  the  tuber. 

§  40.  Among  the  substances  which  of  all  others  would  be 
expected  to  be  destitute  of  inorganic  matter  are  cotton  wool, 
and  the  fine  fibre  of  flax.  Indeed  it  was  at  one  time  main- 
tained that,  these  substances  w^ere  composed  of  carbon,  oxy- 
gen and  hydrogen,  and  hence  would  be  entirely  volatilized 
by  heat;  and  hence,  too,  as  they  were  composed  of  those 
bodies,  their  cultivation  would  not  impoverish  the  soil,. pro- 
vided the  other  parts  were  duly  returned  to  it.  But  these 
views  proved  fallacious.  Prof.  Shepard,  on  submitting  the 
cotton  wool  to  analysis  several  years  ago,  found  the  percent- 
age of  the  ash  to  be  0.9247,  nearly  one  per  cent.  The  ash,. 
as  obtained,  gave  the  following  results  in  his  analysis,, viz: 

Carbonate  of  potash,  (traces  of  soda,)  4-1.19 

Phos.  of  lime,  (traces  magnesia,)    25.44 

Carbonate  of  lime,   8.87 

Carbonate  of  magnesia, 6.85 

Sulp.  potash,  2.70  ^ 

Alumina,  (accidental,) 1.40 

Chjorides,  potassium  and  magnesium,  ) 

Sujp.  qf  lime,  Phos,  potash,  oxide  C  6.43 

of  iron  and;  loss,  ) 

This  analysis  is  quoted  for  the  purpose  of  showing  that  the 
finest  fibre  contain  matter  derived  from  the  soil.  So  of  the 
finest  flax  fibre  whose  ash  is  found  to  contain : 


NORTH-CAROLINA   GEOLOGICAL   SURVEY.  60 

f Carbonate  of  lime, 62.00 

Sulphate  of  lime, 7.15 

Phosphate  of  lime,  13.66 

Oxideofiron 3.99 

Garb,  of  magnesia,  with  traces  of  chloride  of  sodium,  2.00 

-Silica, - 11.20 

100.00 

'The  steep  water  in  which  flax  is  rotted  contains  a  small 
'amount  of  matters  dissolved  out  of  the  flax,  but  neither  the 
addition  to  the  soil  of  this  water,  nor  the  refuse  of  its  dress- 
ing is  sufficient  to  restore  the  soil  to  the  state  it  was  in  prior 
to  the  growth  of  the  crop. 

§  41.  Various  methods  are  adopted  to  supply  the  waste  in 
fertilizing  matter,  or  to  diminish  it  during  cultivation.  One 
of  the  cheapest  methods  is  to  allow  as  much  of  the  crop  to 
decay  upon  the  field  as  possible. 

This  course  is  adopted  when  a  planter  ploughs  in  the  stalks 
of  indian  corn,  cotton,  or  the  stubble  of  rye  and  wheat.  There 
is  an  advantage  in  ploughing  in  the  stubble  of  all  cereals. 
-Another  method  has  been  adopted.  The  stubble  is  first  burn- 
ed and  the  ashes  have  been  strewed  over  the  field  under  the 
impression  that  they  contain  all  the  fertilizing  matter.  This 
method,  however,  has  never  proved  successful.  This  is  due 
in  part  to  the  nature  of  the  ash.  All  silicious  stems,  when 
heated  to  redness  and  burned,  undergo,  so  far  as  their  silica 
is  concerned,  an  important  change,  which  consists  in  convert- 
ing the  soluble  into  an  insoluble  silica,  and  is  therefore  not 
immediately  available  to  the  plant ;  when  ploughed  in  entire 
and  allowed  to  waste  in  the  soil,  all  the  soluble  silica  is  pre- 
served in  a  condition  to  meet  the  wants  of  the  growing  vege- 
table. 

The  plants  which  belong  to  the  corn  family,  however,  are 
not  so  profitably  employed  as  fertilizers  as  clover,  buckwheat 
and  the  pea.  This  fact  becomes  obvious  from  an  inspection  of 
the  composition  of  the  corn  stalk,  or  the  stubble,  or  straw  of 
wheat,  and  comparing  it  with  the  composition  of  the  latter. 
'Still,  the  use  of  the  corn  stalk  is  highly  important.  I  have 
'found  it  composed  of  the  following  elements : 


70  .          NORTH-CAROLINA   GEOLOGICAL   SURVEY. 

Potash, 16.210 

Soda,  24.699 

Phosphates  of  lime  and  magnesia,  15.150 

Lime,    2.820 

Magnesia,    0.936 

Silica,  12.850 

Sulphuric  acid,  10.793 

Chlorine,  10.453 

Carbonic  acid, 1.850 

Organic  matter,  3.200         -,* 

99.461 

§  42.  The  inspection  of  the  composition  of  the  ash  of  the 
corn  stalk  shows  that  it  should  not  be  wasted,  inasmuch  as  a 
quantity  of  the  most  valuable  elements  would  be  lost ;  it  would 
be  equivalent  to  the  wasting  of  so  much  bread  or  corn,  inas- 
much as  the  whole  of  the  matter  may  be  converted  into 
bread  or  corn  in  the  process  of  cultivation. 

The  straw  of  wheat  is  less  rich  in  phosphates  and  the  alka- 
lies than  corn ;  and  yet  it  is  entitled  to  preservation  and  use 
as  a  fertilizer. 

The  ash  of  the  straw  amounts  to  2.660  per  cent.,  and  con- 
sists of 

Silica, 1.235 

Phosphates, 0.422 

Thus  the  phosphates  bear  a  very  small  proportion  to  the 
silica. 

The  complete  analysis  of  the  straw  of  wheat  gave  me : 

Potash, 22.245 

Soda,  5.195 

Earthy  phosphates, 19.600 

Silica,   49.100 

Lime, 3.460 

Magnesia,    0.324 

Sulphuric  acid, 0.876 

Chlorine, 0.121 

In  a  ton  of  straw  the  loss  which  would  be  sustained  by 
wasting  it,  amounts,'  in  pounds,  to 


NORTH-CAROLINA   GEOLOGICAL   SURVEY.  71 

Silica,  29.255 

Potash,    13.253 

Soda,    ... 3.095 

Earthy  phosphates, *   11.678 

Lime, 2.061 

Magnesia,    0.193 

Sulp.  acid, 0.521 

Chlorine,  0.072 

60,128 

The  organic  matter,  which  is  not  taken  into  the  account,  is 
equally  valuable  and  important,  both  as  furnishing  materials 
of  growth  and  the  preservation  of  an  open  condition  of  the 
soils. 

§  43.  Certain  crops  are  raised  expressly  for  the  purpose 
of  improving  the  soil.  These,  when  in  blossom,  are  ploughed 
in,  and  their  subsequent  decay  furnishes  the  manure  for  the 
succeeding  crop.  The  kinds  usually  selected  are  those  which 
grow  vigorously  and  send  their  roots  deep.  Such  plants 
bring  from  a  great  depth  the  fertilizing  matter  to  the  surface 
where  it  becomes  accessible  to  the  succeeding  crop. 

The  red  clover  is  the  favorite  plant  in  the  Northern  States. 
Buckwheat  is  also  employed,  but  it  is  objectionable :  it  con- 
tinues to  spring  up  from  the  seed  as  some  will  ripen  and  mix 
with  the  wheat  crop  or  appear  as  a  weed  in  the  corn  and  re- 
quire eradication  by  the  hoe. 

For  the  South  the  pea  has  become  a  favorite  with  intelligent 
planters,  and  is,  from  its  composition  and  adaptation  to  climate 
the  best  crop  to  precede  wheat  and  to  act  as  its  fertilizer. 

The  composition  of  red  clover  is  well  adapted  to  the  end 
which  it  is  designed  to  fulfil ;  besides,  its  root  is  large,  spreads 
widely  and  sinks  deeply,  and  hence  it  brings  to  the  surface  a 
large  amount  of  fertilizing  matter. 

The  ash  of  the  green  plant  amounts  to  1.06  per  cent.,  when 
dry  to  5.87. 

On  submitting  the  dry  clover  in  the  condition  of  hay  to 
analysis  I  found : 

'      « 

Potash, 25.930 

Soda,  14.915 


NORTH-CAROLINA   GEOLOGICAL   SUKTEY. 

Earthy  phosphates,    20.60& 

Garb,  of  Lime, 30.950 

Chlorine,  1.845 

Sulphuric  acid,  0.495 

If  a  ton  of  this  hay  or  a  plant  in  its  green  state  was  ploughed 
in,  it  would  add  the  following  amount  of  elements  reckoned 
in  pounds  as  follows : 

Potash, , 32.153 

Soda, ,... . . ., . 18.394 

Earthy  phosphates, 25.544 

Carbonate  of  lime, 38.378 

Magnesia,   4.873 

Chlorine,  , 2.288 

Sulphuric  acid, 0.624 

/         Silica, 1.054 

Amounting  to 123.508  Ibs. 

§  44.  It  is  not  perhaps  possible  to  estimate  the  real  value 
of  a  clover  crop  as  a  fertilizer.  Two  hundred  pounds  of  guano 
cost  $5.  May  we  not  infer  that  its  value  exceeds  that  of  this 
popular  fertilizer,  especially  when  it  is  considered  that  the 
organic  part  must  exercise  considerable  influence  and  always 
furnishes  a  large  amount  of  food  ?  It  is  true  that  new  ele- 
ments are  added  by  the  clover,  but  then  the  cost  of  the  crop 
is  trifling,  and  the  effects  are  more  lasting  than  guano  in  this 
climate. 

The  clover  crop  is  from  two  and  a  half  to  three  tons  per 
acre  of  dry  hay.  It  is  more  profitable  to  feed  cattle  upon  it 
before  it  is  ploughed.  By  this  course  or  plan  of  treatment 
the  manure  which  is  added  by  feeding  cattle  nearly  suffices 
for  the  diminished  amount  of  clover  consumed.  It  is  not  re- 
garded as  expedient  to  plough  in  a  very  heavy  green  crop  of 
any  kind.  It  is  better  to  feed  it  in  part,  if  there  were  no 
valuable  returns  in  meat  or  flesh. 

On  account  of  the  grain  in  food  for  cattle  the  clover  crop 
is  preferable  to  buckwheat,  and  yet  this  plant  is  rich  in  fer- 
tilizing products.  » 

§  45.  In  the  South  the  heavy  or  large  stalks  of  corn  are 


NORTH-CAROLINA   GEOLOGICAL   SURVEY.  73 

'broken  down  and  laid  flat  and  longitudinally  with  the  furrow 
:and  covered  in  that  position. 

The  cotton  stalk  is  also  laid  flat  and  ploughed  under.  The 
real  importance  of  this  operation  becomes  evident  on  an  in- 
spection of  the  composition  even  of  the  dried  stalks,  bolls  or 
capsules. 

I  found  from  the  composition  of  the  capsules  that  they  are 
richer  than  the  stalks. 

The  percentage  of  ash  of  the  dry  capsules  is  5.402,  nearly 
six  per  cent.  It  was  obtained  from  capsules  left  in  the  field 
growing  in  the  county  of  ISTash. 

§  46.  The  ploughing  in  of  the  dry  plant  returns  a  certain 
amount  to  the  soil.  From  the  capsules  there  will  be  returned 
'in  every  hundred  parts  of  ash  of  percentage  of  ash  5.60 : 

EarUhy  and  alkaline  phosphates  and  potash, 21.480 

Soda, 5.230 

Earthj  phosphates, 22.923 

Lime,    81 .940 

Magnesia, 11.627 

Sulphuric  acid,  0.400 

'Chlorine,  0.231 

Soluble  silica, 1.302 

Adherent  sand, .-.'..... 2.601 

97.734 

In  the  stalks  of  cotton  in  the  condition  in  which  they  are 
broken  down  preparatory  to  ploughing  the  field  I  found  the 
following  elements: 

Alkaline  and  earthy  phosphates, .. . . 14.400 

Potash,  ...... 17.400 

Soda, 20.860 

Lime, 31.200 

Mrgneisa, 13.160 

Sulphuric  acid, 3.04fi 

Chlorine, 0.400 

Soluble  silica, 0.100 

100.5615 

§  47.  From  the  foregoing  analysis  it  is  evident  that  the 


74  NORTH-CAROLINA   GEOLOGICAL    SURVEY. 

custom  of  ploughing  in  the  old  stalk  after  the  cotton  is  saved 
is  an  important  measure. 

I  have  no  means  of  determining  the  number  of  tons  of  the 
stalks  per  acre,  but  the  amount  thus  saved  to  the  soil  or  suc- 
ceeding crops  is  very  great  and  prolongs  the  fertility  of  a 
cotton  plantation  for  years. 

In  this  connection  it  is  proper  to  state  the  composition  of 
the  cotton  seed,  which  is  now  always  employed  as  a  fertilizer. 
Its  real  value  will  be  duly  appreciated,  though  it  is  scarcely 
necessary  to  confirm  by  analysis  what  experience  had  long 
determined  by  its  use.  But  the  planter  will  understand  bet- 
ter what  he  is  adding  to  his  soil,  and  also  how  much  from  the 
following  results  of  analysis  ; 

Earthy  phosphates, 32.000 

Potash,   15.560 

Soda,    10.960 

Lime,    4.000 

Magnesia,    0.200 

Sulphuric  acid, 2.720 

Chlorine,    0.120 

Carbonic  acid,  8.540 

Soluble  silica,   2.000 

Adherent  sand,  23.600 

99.700 

The  large  quantity  of  sand  is  due  to  cotton  adhering  to  the 
seed  which  had  been  exposed  in  a  pile  to  the  weather.  It 
was  not  suspected  until  the  ash  was  subjected  to  the  action 
of  hydrochloric  acid.  It  is  of  course  foreign  matter. 

After  making  all  the  allowance  necessary  for  this  foreign 
matter  it  will  not  fail  to  strike  every  cotton  grower  of  the 
value  of  the  cotton  seed  as  a  fertilizer. 

§  48.  Analysis  of  the  seed  of  buckwheat : 

Potash,  21.27 

Soda,    2.32 

Phosphoric  acid, 49.85 

Lime,  3.01 

Magnesia,      15.84 

Sulphuric  acid,  1.55 


NOKTH-CAROLINA   GEOLOGICAL    SURVEY.  75 

Silica,  1.95 

Chlorine, 0.30 

Carbonic  acid,  1.95 

Organic  matter,  2.75 

In  the  cultivation  of  this  plant  it  will  be  seen  that  a  large 
amount  of  fertilizing  matter  is  removed  in  the  gathering  of 
.  seed,  or,  if  it  remains,  a  large  amount  is  preserved  for  subse- 
quent crops. 

Every  ten  bushels  of  seed  contains  6.281  Ibs.  of  phosphoric 
acid,  two  pounds  of  magnesia,  and  over  two  pounds  and  a 
half  of  potash.  The  whole  amount  of  valuable  fertilizers  re- 
moved in  every  ten  bushels  of  buckwheat  is  12.450  Ibs. 
The  buckwheat  in  drying  loses  about  the  same  quantity  of 
water  as  wheat  and  rye.  Thus,  on  being  dried  in  a  water 
bath  at  212,  it  lost  12.875  parts ;  and  hence  there  remains  of 
dry  matter,  87.125  of  which  gives  4.132  per  cent,  of  ash. 

The  organic  constitution  of  buckwheat  is  similar  to  the 
cereals,  consisting  of 

Starch,    42.47 

Sugar   and  extractive  matter 6.16 

Dextrine,   1.60 

Epidermis  or  insoluble  matter,  16.42 

A  peculiar  gray  matter,  soluble  in  potash, )  in  10 
but  insoluble  in  water  or  alcohol,             } 

Albumen,  6.70 

Casein,    0.78 

Oil,  0.47 

Water,  12.88 

§  49.  The  foregoing  does  not  relate  so  much  to  matters 
which  can  be  employed  as  fertilizers,  but  is  introduced  here 
for  the  purpose  of  showing  its  nutrient  properties. 

The  pea  will  no  doubt  take  the  place  of  the  red  clover  in 
this  State.  Experience  has  already  proved  its  superiority. 
It  is  easily  cultivated  and  is  not  liable  to  so  many  accidents. 
It  takes  deep  root  and  spreads  widely,  and  is  rich  in  valuable 
fertilizers.  By  careful  extraction  from  the  hill  I  have  found 
its  roots  spreading  through  six  feet  of  ground. 

That  the  value  of  the  pea  may  be  appreciated,  and  its  fer- 


76  'NORTH-CAROLINA   GEOLOGICAL   SURVEY. 

tilizing  matter  applied  to  the  best  advantage,  I  have  carefully 
determined  the  composition  of  its  ash  from  specimens  which 
I  obtained  in  Wake  county. 

The  percentage  of  ash  of  the  pea  vine,  destitute  of  leaves 
and  in  the  condition  in  which  it  is  fed  to  cattle,  and  as  derived 
from  268  grains  of  the  stems  and  branches  in  a  perfectly  dry 
state,  I  found  to  be  4.570. 

On  submitting  this  ash  to  analysis  I  found  it  composed  of 

'Potash,  .„ 7.800 

Soda,  5.650 

Earthy  phosphates, 19.800 

Lime,    , 16.400 

Magnesia,    30.040 

Sulphuric  acid,  11.710 

Chlorine,  1.710 

Silica,  , , 10.900 

Soluble   silica,  6.000 

When  we  find  so  large  a  percentage  of  ash,  and  a  compo- 
sition clearly  rich  in  inorganic  constituents,  we  may  not  doubt 
the  utility  of  employing  this  plant  as  a  fertilizer  instead  of 
the  clover  plant,  as  it  is  considerably  richer  in  the  expensive 
^elements  of  nutrition. 

§  50.  The  pea  with  its  pod  is  richer  in  phosphates  than  the 
vine,  and  as  these  are  ripe  when  turned  under  the  value  of 
the  crop  for  this  purpose  is  increased. 

The  percentage  of  ash,  as  determined  from  365  grains  of 
the  dried  pod  with  the  pea,  is  3.13.  The  percentage  of  ash 
is  greater  from  the  presence  of  the  pod.  But  this  being 
ploughed  in  the  result  is  more  accurate  from  their  combina- 
tion. If  the  nutrient  matters  of  the  pea  were  to  be  deter- 
mined it  should  be  analyzed  by  itself. 

The  composition  of  the  pea,  with  its  pod,  I  found  as  follows : 

Potash,    ..., 24.200 

Soda,  -., , 10.7^9 

Earthy  and  aikaline  phosphates,  32.200 

Carbonate  lime,   11  000 

Magnesia,  .„.,.. 3.000 


NORTH-CAROLINA   GEOLOGICAL  SURVEY.  77 

Sulphuric  acid,   1.461) 

Chlorine,  0.561 

Soluble  silica, , 10.020 

Silica, 3.800 

Percentage  of  ash,  3.137 

The  pea  in  composition  is  closely  related  to  the  cereals,and 
in  nutritive  powers  ranks  high.  Indeed  the  leguminous  plants 
as  a  class  stand  at  the  head  of  a  certain  class  of  nutrients. 
The  bean  employed  for  food  gives  more  muscle  or  strength 
of  muscle  and  endurance  than  the  cereals.  This  is  due  in 
part  no  doubt  to  its  phosphoric  acid  and  nitrogenous  matters. 

It  appears  from  the  foregoing  that  the  greater  the  amount 
of  nutrient  power  the  more  valuable  they  are  as  fertilizers. 
Weeds  which  bear  only  small  seeds,  or  which  are  composed 
of  lime,  are  less  useful  than  leguminous  plants,  and  others 
which  are  closely  related  to  the  cereals. 

§  51.  The  composition  of  another  plant  which"  may  be  in- 
teresting in  another  point  of  view  is  tobacco.  I  design  to 
show  by  the  analysis  how  much  the  tobacco  exhausts  the  soil, 
and  of  what  elements. 

Thus,  one  hundred  parts  of  the  ash  consist  of 

Potash,  4.260 

Soda,  6.140 

Lime,    48.000 

Magnesia,  9.180 

Phosphates  of  lime  and  magnesia,  etc.,  14300 

Sulphuric  acid, 8.420 

Chlorine 1.100 

Silica  and   sand, 4.800 

Soluble   silica,  3.800 

100.000 

This  tobacco  grew  in  Rockingham  county,,  and  was  regard- 
ed by  the  manufacturer  as  fine  as  any  which  is  grown  in  the 
northern  counties,  The  result,  however,  of  this  analysis  sur- 
prised me,  as  it  contained  so  much  less  of  potash  than  can  be 
expected  in  the  best  of  tobacco.  It  is  found'  by  many  analy- 


78  NORTH-CAROLINA   GEOLOGICAL   SURVEY. 

ses,  however,  that  the  ash  is  variable  in  the  proportions  of 
its  elements. 

The  tobacco  which  obtains  the  highest  price  in  the  Paris 
market  contains  a  much  larger  proportion  of  potash  and  less 
lime.  This  specimen  had  the  fine  yellow  brown  color  which 
is  regarded  as  indicative  of  the  best  quality.  As  it  is,  how- 
ever, it  is  a  lime  plant,  nearly  one-half  being  composed  of 
carbonate  of  lime. 


CHAPTER  VI. 

4* 

FERTILIZERS CONTINUED . 

Marl  beds,  or  Marl  formations. — The  different  periods  to  which  they  belong, 
or  their  relation  to  each  other. 

§  52.  There  are  three  distinct  formations  from  which  marl 
is  obtained.  Enumerating  them  in  the  ascending  order,  or 
according  to  age,  they  lie  relatively  to  each  other  as  follows : 
1.  Green  Sand  /  2.  Eocene  Marl ;  3.  Miocene  Marl. 

The  first,  or  green  sand,  is  the  formation  which  is  so  favor- 
ably known  in  ISTew  Jersey  as  a  fertilizer,  having  been  em- 
ployed for  that  purpose  for  more  than  half  a  century.  It 
derived  its  name  partly  from  its  green  color,  and  partly  from 
its  granular  consistence.  The  beds  thus  named  are  known 
not  only  in  this  country  but  also  in  many  parts  of  Europe 
by  the  same  name,  and  where,  to  a  certain  extent,  they  are 
also  used  as  a  fertilizer. 

In  the  geological  systems  its  beds  are  subordinate  to  the 
cretaceous  system,  and  in  Europe  form  subordinate  beds  be- 
neath the  chalk — the  white  chalk  in  common  use  for  marking. 


NOKTH-CAROLINA   GEOLOGICAL   SURVEY.  79 

In  this  country  this  part  of  the  cretaceous  system  is  wanting, 
or  has  not  yet  been  recognized.  From  its  wide  extent,  both 
in  this  country  and  Europe,  it  is,  geologically  speaking,  an 
important  formation  ;  so  also  in  an  economical  point  of  view 
it  is  equally  important,  for  it  has  been  a  source  of  revenue  to 
the  agricultural  community,  not  second  even  to  guano.  For 
permanent  improvements  in  the  soil  it  is  superior  to  this  far 
famed  substance,  its  effects  lasting  from  ten  to  fifteen  years. 
In  New  Jerse}r  it  first  attracted  attention  from  an  accident : 
some  green  sand  being  thrown  out  of  a  ditch  upon  a  bank, 
an  exceeding  fine  growth  of  clover  was  the  consequence.  It 
was  immediately  inferred  that  the  substance  upon  the  ditch 
bank  was  the  cause  of  this  fine  growth ;  and  hence  a  trial 
was  made  of  it. 

From  many  subsequent  experiments  and  observations  its 
claim  as  a  good  fertilizer  became  established.  This  happened 
more  than  fifty  years  ago,  and  ample  experience  in  the  mean 
time  has  fully  satisfied  the  agricultural  community  at  large 
that  it  is  worthy  the  confidence  which  has  been  reposed  in  it, 

§  53.  In  the  subsequent  pages  I  propose  to  give  a  full 
statement  of  the  grounds  upon  which  its  reputation  rests, 
and  also  to  furnish  numerous  analyses  of  the  best  and  poorest 
varieties  of  this  substance.  In  the  first  place  I  deem  it  proper 
to  show  its  geological  relations,  and  its  relative  position  to 
other  beds  of  marl,  inasmuch  as  it  will  aid  in  determinim'ng 
in  any  given  case  whether  the  substance  or  beds  in  question 
really  belong  to  those  which  have  received  the  common  name 
referred  to.  In  all  cases  this  is  an  economical  question,  or 
may  be  thus  used,  inasmuch  as  the  beds  formed  during  this 
geological  era  have  a  composition  which  fits  them  for  the 
purpose  for  which  they  have  been  so  largely  employed.  Beds, 
therefore,  occupying  their  position  may  be  supposed  without 
trial  and  without  analysis  to  contain  the  active  fertilizing 
matter.  It,  however,  cannot  be  determined  by  these  external 
observations,  how  much  they  contain,  for  it  is  found  that  they 
are  variable  in  composition,  so  far  as  quantity  is  concerned. 
For  the  purpose  of  determining  their  commercial  value,  or 


SU»  NORTH-CAROLINA   GEOLOGICAL   SURVEY. 

to  ascertain  the  amount  which  may  be  profitably  employed 
and  how  far  they  may  be  transported  has  to  be  ascertained 
by  analysis. 

There  are  several  localities  at  which  the  green  sand  occurs. 
The  strongest  marl  beds  occur  at  Black  Rock  on  the  Cape 
Fear  river,  about  twenty-five  miles  above  Wilmington.  It 
forms  low  bluffs  at  several  other  points,  but  it  appears  to 
terminate  from  two  to  five  miles  below  Brown's  landing. 

Striking  across  the  county  to  the  eastward  it  again  appears 
prominently  .at  Rocky  Point,  twenty  miles  above  Wilmington. 
The  green  sand,  unlike  the  shell  marl,  forms  continuous  beds, 
but  as  its  beds  are  undulating,  they  rise  at  certain  points  to 
the  surface,  and  then  sink  beneath  it. 

In  this  State  I  have  been  unable  to  determine  its  thickness, 
or  the  number  of  beds  which  properly  belong  to  it.  For  this 
reason  1  propose  to  describe  them  now,  as  they  are  known  to 
exist  in  New  Jersey,  inasmuch  as  such  a  description  may  aid 
others  where  it  exists,  to  determine  with  accuracy  both  their 
thickness  and  the  number  of  beds  which  compose  the  green 
sand  formation  in  North-Carolina.  The  difficulty  in  the  way 
of  solving  this  question  is  the  slight  elevation  of  the  banks 
of  rivers  and'  ravines  above  the  adjacent  country.  We  find 
at  Black  Rock,  for  example,  a  strong  bluff  of  this  deposit, 
but  the  water  is  never  low  enough  to  disclose  the  bottom 
beds,  or  the  masses  upon  which  it  rests. 

In  order  to  state  all  that  is  known  of  the  green  sand  and 
marl,  and  their  relations  to  each  other,  I  have  prepared  sev- 
eral sections  which  show  how  they  are  situated  with  respect 
to  each  other.  From  these  sections  it  will  be  seen  that  the 
marl  beds  vary  much  in  thickness,  and  in  their  relations  at 
different  places  where  they  are  exposed  to  the  best  advant- 
age. Thus,  section  I,  fig,  1,  exhibits  all  the  beds  as  they  exist 
at  Black  Rock: 


NORTH-CAROLINA   GEOLOGICAL   SURVEY.  SI 

FIG.  1. 


3 


>••  J 


W>  <?  %,-    «  ^  §  s«  @      *  d)     CP  °  sa  8    Q   ^ 


i.  The  upper  bed  is  the  common  marine  sand  spread  wide- 
ly over  the  county.  2.  Beneath  it  there  is  a  mass  of  brown 
soil,  or  earth,  which  is  probably  more  widely  spread  than  any 
other  in  the  eastern  part  of  the  State.  It  is  sometimes  pebbly 
towards  the  upper  part,  and  at  many  places  the  pebbles  are 
cemented  by  oxide  of  iron.  A  pudding  stone  is  thereby 
formed,  which  is  very  firm,  and  has  been  employed  as  a  rough 
building  material.  In  the  vicinity  of  Fayetteville  it  is  not 
unfrequently  used  for  the  more  ordinary  kinds  of  construction. 
From  the  vicinity  of  Raleigh  eastward  it  may  be  seen  by  the 
road-side  where  a  cut  has  been  extended  through  the  super- 
incumbent sand.  This  bed,  which  is  at  least  twelve  feet  thick 
at  Fayetteville,  originated  in  the  decomposition  of  primary 
rocks,  the  debris  of  which  becomes  red,  or  reddish  brown,  by 
exposure  to  the  atmosphere.  If  any  thing,  it  is  more  persist- 
ent towards  the  belt  where  these  rocks  formed  the  surface 
materials.  How  this  stratum  has  been  spread  out  so  evenly 
and  widely  through  the  whole  width  of  the  State  from  south 
to  north  is  not  satisfactorily  accounted  for.  Along  the  wes- 
tern margin  referred  to  it  rests  on  the  rocks  from  which  it  is 
derived.  Eastward,  however,  where  recent  beds  of  different 
kinds  take  their  proper  places,  this  brown  earth  formation  is 
found  near  the  surface,  but  with  ;several  marine  strata  be- 
neath and  upon  which  it  reposes.  J.t  always  maintains  the 
position  I  have  given  it,  or  its  relations  are  never  altered  ; 
and  hence,  though  it  may  be  regarded  as  a  soil,  still  it  must 
7 


NOBTH-CAKOLIKA   GEOLOGICAL   SURVEY. 

have  been  spread  out  by  some  general  cause,  and  at  one 
specific  period. 

This  bed,  however,  is  not  confined  to  this  State.  It  extends 
over  a  part  of  Maryland,  Virginia,  South  Carolina,  Georgia, 
and  Alabama. 

It  is,  therefore,  a  wide  spread  stratum,  having  its  origin 
through  the  influence  of  general  causes.  That  this  cause  or 
force  operated  with  considerable  violence  is  indicated  by  the 
losses  which  one  at  least  of  the  inferior  formations  has  sus- 
tained. The  shell  marl,  for  example,  is  never  a  continuous 
deposit,  and  some  of  the  beds  are  frequently  furrowed  and 
channelled,  apparently  by  a  rush  of  water  over  them,  remov- 
ing not  only  the  upper  layers,  but  cutting  frequently  deep 
into  the  beds.  An  erosion  of  this  kind  is  illustrated  by  fig.  5. 
The  brown  earth  fills  these  eroded  channels  without  mixing 
at  all  with  the  marl. 

The  next  stratum  beneath  is  a  brick  clay,  which  is  also 
general,  but  it  is  absent  occasionally,  in  which  case  the  brown 
bed  occupies  its  place.  This  clay  varies  considerably  in  com- 
position ;  it  is  sometimes  charged  with  sand,  in  others  it  is 
very  fine  and  compact,  and  makes  the  best  of  brick.  It 
passes  also  into  potter's  clay.  It  is  bluish  white,  gray  and 
reddish  at  different  places.  It  never  exceeds  five  feet  in 
thickness. 

4.  The  fourth  stratum  is  sand,  usually  gray,  and  loose  in 
texture,  not  unlike  quick  sand. 

5.  The  shell  marl  occupies  the  fifth  place  in  the  descending 
order.     It  wjll  be  fully  described  hereafter. 

6.  The  beds  of  green  sand  occupy  the  sixth  place,  and  at 
Blackrock  it  may  be  divided  into  two  beds ;  the  upper  con* 
tains  a  large  amount  of  clay,  and  the  lower  is  sandy  with 
more  lime  ;  it  is  also  indurated,  or  partially  consolidated. 

The  lower  mass  forms  a  shelving  projection  from  the  upper, 
some  eight  or  ten  feet  wide,  when  it  falls  off  perpendicularly 
to  a  depth  of  fifteen  fe^j.*  The  lower  part  is  always  under 
water,  and  I  know  of  9b  locality  at  which  this  part  of  the 
formation  is  exposed.  I  regard  this  as  an  unfortunate  circum- 
stance, inasmuch  as  I  have  reason  to  believe  that  the  quality 


tfOETH-CAROLINA   GEOLOGICAL   SURVEY.  83 

of  the  marl  is  better  towards  the  bottom,  or  lower  in  the  bank, 
than  where  it  is  exposed.  At  certain  points  in  New  Jersey 
it  has  a  sandy  base,  but  several  feet  above  it  becomes  a  rich 
marl. 

The  color  of  this  kind  of  marl  is  green  or  dark  green.  It 
is  always  rather  sandy,  but  still  it  is  rich  even  then  in  fertiliz- 
ing matter.  The  Blackrock  beds  here  have  a  dark  green,  or 
greenish  gray,  and  may  be  divided  into  two  parts :  the  upper 
which  has  a  darker  color,  and  is  much  like  clay  to  the  feel ; 
and  the  lower,  which  is  consolidated  and  of  a  greenish  gray, 
and  rather  gritty  to  the  touch.  There  is  no  dividing  line 
which  is  so  clearly  marked  that  we  can  fix  upon  the  termina- 
tion of  the  lower,  and  the  beginning  of  the  upper  division, 
but  still  the  difference  observable  is  sufficiently  strong  to 
admit  of  the  division  I  have  proposed ;  though,  geologically, 
it  may  be  regarded  as  one  mass.  The  division  is  more  im- 
portant in  an  economical  point  of  view,  inasmuch  as  the 
composition  of  the  upper  is  quite  dissimilar  to  the  lower 
bed. 

§  54.  In  New  Jersey  the  green  sand  formation  is  composed 
of  six  distinct  beds  ;  three  of  which  are  known  as  green  sand 
proper,  in  consequence  of  the  peculiar  composition  ;  and  three 
which  are  composed  of  a  common  marine  sand,  and  which 
separates  each  of  the  respective  beds  from  the  other.  In 
North- Carolina  it  is  probable  that  equivalent  beds  exist,  but 
it  has  been  impossible  up  to  this  time  to  recognize  but  two. 
At  Blackrock  the  lowest  is  known  by  its  fossils  :  the  Exogyra 
costata,  Ostrea  falcata,  Belemintes  Americana,  and  casts  of 
the  cucullea  vulgaris.  This  mass  terminates  in  one  which  is 
quite  argilaceous,  and  in  this  part  of  it  no  fossils  have  been 
observed. 

The  third  or  upper  bed  may  be  probably  recognized  at 
Tawboro',  on  the  Tar  river,  at  the  marl  beds  of  Col.  Clark. 
It  is  only  about  four  feet  thick,  but  is  underlaid  by  sand,  in 
which  much  sulphuret  of  iron  is  disseminated. 

The  annexed  section,  fig.  2,  shows  the  relations  of  the  beds 
referred  to  upon  the  Tar  river: 


84  NORTH-CAROLINA   GEOLOGICAL   SURVEY. 

FIG.  2.  Soil.     1.  Ten  feet  of 

^^T-T^v?.  -.• .'/  "  yellow  sand.     2.  Four 

^  ~T  feet  of  greenish    clay. 

/_ _  3.  Six  feet  of  shell  marl. 

j  $  M   "'<$  ^3*0     §  3      4.  Four  feet  of  upper 

yrv~  ~  shell   marl,    containing 

^/  A-      lignite  and  pyrites.     5. 

/**"  p~~  Light    gray    sand,   the 

j£ ^  thickness   of  which    is 

undetermined,  as  it  ex- 
tends below  the  water  of  the  Tar  river,  and  does  not  become 
visible  at  any  other  place  in  the  vicinity.  It  is  probably  one 
of  the  sand  beds  which  seperate  two  of  the  adjacent  beds  of 
green  sand.  But  as  it  has  not  furnished  fossils  it  cannot  be 
confidently  maintained.  It  is,  however,  mineralogically,  a 
green  sand. 

As  all  the  beds  of  green  sand  are  never  exhibited  at  one 
place,  and  as  those  which  have  been  spoken  of,  except  the 
upper,  on  the  Tar  river,  the  thickness  of  this  formation  re- 
mains undetermined. 

Wherever  it  occurs  the  country  is  comparatively  low,  and 
at  no  point  yet  discovered  has  the  base  of  the  Blackrock  mass 
or  lowest  been  sufficiently  elevated  to  disclose,  even  approxi- 
mately, its  thickness. 

§  55.  The  bluffs  which  exhibit  the  tertiary  and  secondary 
formations  of  the  eastern  counties  are  mostly  upon  the  south- 
side  of  the  rivers  and  ravines.  Some  of  these  bluffs  are  high 
and  commanding,  but  they  are  never  continuous  for  long  dis- 
tances. The  green  sand  does  not  appear  in  any  bluff  above 
Brown's  landing.  Indeed  it  disappears  about  three  miles  be- 
low, and  though  this  landing  is  high  and  bold,  yet  I  am  unable 
to  recognize  a  bed  which  can  be  referred  to  the  upper  part 
of  the  secondary  formation. 

At  Brown's  landing  there  are  numerous  distinct  beds.     In 
arrangement  they  belong  to  two  distinct  dates :  1st,  the  upper 
which  is  Miocene,  and  the  lower  which  is  probably  Eocene. 
These  beds  are  exhibited  in  the  following  section  : 


KOBTH-CAROLINA   GEOLOGICAL   SURVEY.  85 

• 

FIG.  3. 


-8- 


vS 


1.  Sand.  2.  Brown  earth.  3.  Clay,  four  or  five  feet  thick. 
4.  Sand  and  pebbles.  5.  Shell  marl.  6.  Sand,  with  consoli- 
dated beds  which  becomes  a  gray  sandstone,  with  fossils  and 
lignite.  7.  Blue  clay.  8,  Sand,  blue  clay,  succeeded  again 
by  sand.  The  formation  below  is  here  concealed  under  water. 

The  most  interesting  points  at  Brown's  landing  are  the 
thick  beds  of  sand  and  clay  beneath  the  shell  marl,  the  latter 
of  which  is  identical  with  that  at  Black  Hock,  where,  it  will 
be  recollected,  this  marl  rests  upon  the  upper  bed  of  green 
sand.  At  the  landing  we  find  interposed  at  least  sixty  feet 
of  material  which  does  not  occur  at  Black  Rock  at  all.  These 
intervening  beds  I  regard  as  Eocene.  It  may,  however,  prove 
to  be  Miocene,  and  as  a  part  of  the  lignite  formation  equiva- 
lent to  that  which  is  spread  over  large  tracts  of  country  in 
Nebraska  and  Kansas.  It  has  consolidated  beds,  cemented 
by  carbonate  of  lime,  in  which  lignite  is  very  common. 
Another  fact  of  interest  is  the  presence  of  green  sand  in  the 
shell  marl,  while  it  is  almost  entirely  absent  in  the  inferior 
beds.  The  marl  contains,  also,  Exogyra,  Belemnites  and  cop- 
rolites  which  belong  to  the  green  sand  which  were  washed 
from  these  beds.  The  change  in  passing  from  the  Eocene  to 
the  Miocene  was  attended  with  considerable  violence,  as  the 
latter  have  abundance  of  pebbles,  rolled  coprolites  as  hard  as 
quartz,  teeth,  etc.  The  bottom  is  truly  a  pebbly  bed. 


00  NOKTH-CAKOLINA   GEOLOGICAL   STTKYEY. 

§  56.  The  sand  beds  beneath  the  shell  marl  extend  nearly 
to  Fayetteville.  They  may  be  examined  at  the  bridge  over 
Rockfish  creek,  seven  miles  from  Fayetteville,  and  at  Mrs. 
Purdyrs  marl  bed,  ten  miles  above  Elizabethtown,  and,  also, 
at  Elizabethtown,  in  the  high  banks  below  the  village. 

The  sand  of  this  formation,  when  it  is  unconsolidated,  is 
loose  and  caves  from  its  banks  continually.  In  addition  to 
lignite  and  a  few  shells  it  contains  an  abundance  of  iron 
pyrites.  Its  whole  thickness  on  the  Cape  Fear  is  about 
seventy  feet. 

It  is  possible  the  beds  may  be  recognized  on  the  Neuse  and 
Tar  rivers,  especially  at  the  Sarpony  hills,  fourteen  miles  be- 
low Goldsboro'. 

§  57.  The  bluff  below  Elizabethtown  presents  the  following 
strata,  as  exhibited  in  fig.  4 : 

FIG.  4.  1.    Sand  with   peb- 

x^T^rrgr^Trs^^oroo.-f'   bles.    2.  Brown  earths. 

/  2.  3.  Sand.  4.  Shell  marl 

/ 3_   three    feet   thick.      5. 

^^  $$• &       &   @>    |^    4.    Sand    containing    lig. 
.-. y.v.-.-.v. -.-. ... ........-....:..•.-. -..-... ..-.-.  y     njte  and  consolidated 

k     layers,  with  numerous 

— .  fossils. 

7_     The  b  e  ds  of  san  d  with 

lignite  or  charred  wood 
are  similar  to  those  of  Brown's  landing  and  Walker's  bluff. 
But  there  are  no  particles  of  green  sand  or  fossils  from  this 
formation  in  the  shell  marl  bed.  It  appears  that  the  shell  marl 
beds  in  which  are  intermingled  the  organic  remains  from  the 
secondary,  are  confined  to  a  narrow  belt  which  may  be  traced 
along  the  eastern  border  of  the  formation. 

Section  No.  5  is  designed  to  show  the  relations  of  the  shell 

O 

marl  to  the  white  Eocene  beds  of  the  Neuse,  which  do  not  ex- 
tend south-westward  to  the  Cape  Fear. 


NORTH-CAROLINA   GEOLOGICAL   SURVEY. 
FlG.   5. 


87 


1.  Soil,  consisting  of  red  earth  penetrating  into  an  excava- 
tion in  the  bed  of  Eocene  marl.  2.  Position  of  the  ordinary 
shell  marl.  3.  Upper  part  of  the  bed  in  which  most  of  the 
fossils  occur.  4.  Body  of  white,  or  light  drab  colored  marl. 

The  section  shows  the  marl  beds  of  Mr.  Wadsworth,  of 
Craven  county. 

It  will  be  observed  that  the  shell  marl  is  in  contact  with 
the  drab  colored  marl,  the  entire  mass  of  the  lignite  forma- 
tion of  the  Cape  Fear  being  absent.  At  this  place,  the 
brown  earth  is  present  filling  the  ancient  fissures  of  denuda- 
tion. The  shell  marl  is  not  present  at  this  point,  but  appears 
in  the  same  relative  position  three  or  four  hundred  yards 
west  from  this  bed. 

§  58.  The  foregoing  sections  show  the  diverse  nature  of  the 
beds  composing  many  of  the  bluffs  of  the  Cape  Fear,  Neuse 
and  Tar  rivers.  The  same  facts  would  be  also  shown  by  sec- 
tions at  many  points  upon  the  Roanoke  and  Meherrin  rivers 
farther  north.  The  position  of  the  shell  marl  seems  to  change, 
as  in  one  case  it  rests  upon  the  green  sand,  in  the  second 
upon  a  lignite  formation  some  sixty  or  seventy  feet  thick, 
and  then  again  upon  a  whitish  marl  which  is  well  known  to 
belong  to  the  Eocene  period. 

The  formation  above  the  shell  marl  is  mostly  a  marine  sand. 
Its  thickness  is  variable,  and  it  is  sufficiently  great  to  prove 
that  a  long  interval  had  elapsed  before  the  present  was  fully 
ushered  in. 

§  59.  The  series  of  beds,  from  the  green  sand  upwards, 
which  hold  a  definite  place  in  the  geological  scale,  have  been 
exhibited  in  the  sections  alluded  to,  do  not  take  in  the  most 


NORTH-CAROLINA    GEOLOGICAL    SURVEY. 

recent.  Upon  the  coast  or  near  it  I  have  observed  limited 
patches  of  peaty  deposits  resting  upon  a  marine  sand,  and 
upon  the  former  beds  of  shells  composed  mainly,  if  not  en- 
tirely, of  those  which  now  live  upon  the  coast.  These  beds 
of  shells  are  rarely  more  than  ten  or  fifteen  feet  above  high 
tide.  The  peaty  beds,  however,  lie  at  the  water's  edge,  and 
at  many  points  are  rapidly  disappearing  by  the  action  of  tides 
and  waves. 

The  mode  in  which  the  shells  are  collected  appeal's  to  have 
been  similar  to  that  which  was  instrumental  in  the  accumula- 
tion of  the  common  shell  marl ;  they  appear  to  be  heaps  of 
dead  shells  thrown  up  by  the  waves, — still  they  are  perfect, 
or  are  but  slightly  worn  by  attrition.  Those  which  are  chang- 
ed the  most  have  become  simply  chalky  from  the  action  of 
the  weather  upon  them  since  they  were  deposited.  The  beds 
which  are  now  forming  have  received  the  name  of  Eolian 
by  Lieut.  Nelson.  The  sands  of  the  entire  coast  come  under 
this  denomination,  and  may  be  regarded  as  deposits  overly- 
ing the  accumulation  of  beds  of  shells  already  alluded  to. 

§  60.  The  formations  then  upon  the  coast  and  interior  of  K. 
Carolina  may  be  subdivided  into:  1.  Green  Sand,  an  import- 
ant part  of  the  secondary;  2.  Eocene,  consisting  of  white 
marl  which  is  made  up  of  comminuted  corals  and  shells,  and 
the  lignite  beds  which  consist  of  gray  sand  and  pebbles,  em- 
bracing consolidated  beds  and  a  few  beds  of  clay  ;  3.  Miocene 
or  Shell  Marl,  which  is  composed  of  fragments  and  entire 
shells  accumulated  in  banks ;  4.  Pliocene  and  Postpliocenc, 
which  are  made  up  of  peaty  beds,  banks  of  shells,  and  finally, 
moveable  sands,  (Eolian  sands,)  which  are  constantly  moving 
beyond  the  present  coast  line.  It  should  be  observed,  how- 
ever, that  the  third  or  Miocene  division  is  regarded  by  Prof. 
Holmes  and  the  late  Prof.  Tuomey  as  Pliocene. 

In  this  State  I  have  obtained  the  same  fossils  in  equal 
numbers  as  those  in  Virginia,  where  the  beds  still  retain  the 
designation,  Miocene.  Not  only,  however,  do  they  contain 
the  Virginia  fossils,  but  those  which  in  South-Carolina  have 
served  to  change  the  name  from  Meiocene  to  Pliocene.  It 
appears  that  many  of  the  Virginia  species  belong  to  a  warm 


NORTH-CAROLINA    GEOLOGICAL    SURVEY.  89 

climate,  that  they  became  extinct  at  an  earlier  period  than  at 
points  farther  south,  and  that  the  same  species  which  were 
once  common  on  the  coast  of  Virginia  and  Maryland,  and 
which  are  now  extinct  so  far  as  that  part  of  our  coast  is  con- 
cerned, still  live  farther  south  where  the  climate  is  congenial 
to  the  species. 


CHAPTER  VII. 

FERTILIZERS CONTINUED. 

Stone  Marl,  its  economical  value. — Composition  of  the  Green  Sand  of  the 
Cape  Fear  River. 

§  61.  The  marls  of  ISTorth-Carolina  do  not  rank  so  high  as 
the  strong  marls  of  other  States.  This  is  in  consequence  of 
the  large  proportion  of  sand  with  which  they  are  intermixed. 
It  appears  that  the  coast  has  been  from  time  immemorial  the 
great  depository  of  sand.  The  rivers  from  the  interior  carry 
sand  or  matter  in  which  silex  greatly  predominates.  The 
rocks  in  the  interior  belong  to  the  silicious  class.  Limestones 
are  absent.  But  the  great  amount  of  sand  of  the  coast  has 
been  probably  derived  from  more  distant  sources,  and  hence 
it  is  probable  we  must  look  to  the  regular  currents  of  the 
ocean  which  flow  in,  more  or  less,  upon  it,  for  the  determa- 
tion  of  the  source  from  which  its  sands  have  been  derived. 
When  the  Atlantic  tide  reached  inland  as  far  as  the  last  of 
the  series  of  falls  of  the  rivers  of  the  State,  as  the  Koanoke, 
Cape  Fear  and  Neuse,  it  acted  upon  a  granite  rock  which 
readily  decomposed,  and  which  must  have  furnished  an  im- 
mense quantity  of  silicious  debris.  This  rock  may,  therefore, 
have  been  one  of  the  sources  of  the  sand  alluded  to.  Some 
beds  of  marl  are  consolidated  into  rock,  and  where  this  con- 


90  NOKTH-CAROLINA  GEOLOGICAL   SUEVEY. 

solidation  was  effected  through  the  agency  of  soluble  silica, 
it  has  become  a  durable  mass,  and  fit  for  being  used  in  build- 
ing. It  has  received  the  name  of  stone  marl,  which  I  propose 
to  speak  of  in  the  first  place. 

§  62.  Stone  Marl.  There  are  two  varieties  of  stone  marl, 
both  of  which  deserve  a  special  notice.  The  first  consists  of 
shells  cemented  strongly  together,  and  which  are  usually  from 
one  to  one  and  a  half  inches  across,  and  very  uniform  as  to 
size.  They  are  very  firmly  cemented  by  silica,  which  seems 
to  have  penetrated  the  shells  more  or  less.  This  rock  has 
been  employed  for  a  long  period  for  small  mill  stones.  Its 
valuable  qualities  consist  in  being  easily  wrought  when  first 
removed  from  the  quarry,  but  subsequently  becomes  very  hard 
and  strong.  Being  made  up  of  shells,  it  has  a  rough  appear- 
ance, even  when  cut  evenly ;  but  this  feature  constitutes  its 
recommendation.  For  certain  structures  it  is  admirably  adap- 
ted. The  enclosure  of  the  cemetery  in  Newbern  is  made  of 
this  rock,  and  the  noble  arches  have  an  imposing  effect.  The 
rock  is  very  durable,  as  appears  to  be  well  sustained  by  the 
rock  itself,  where  it  is  exposed,  or  has  been  exposed  for  ages. 
For  rough  work  it  may  be  used  without  dressing,  but  for 
ornamental,  if  dressed  properly,  it  is  far  superior  to  granite 
for  all  structures,  where  the  material  should  be  indestructible. 
It  is  adapted  to  the  construction  of  dwellings,  as  the  walls  will 
continue  dry  in  wet  weather. 

This  rock  underlies  Newbern  and  the  adjacent  county.  It 
extends  fifteen  or  twenty  miles  in  a  northeast  and  southwest 
direction.  In  some  places  it  reaches  the  surface ;  in  others 
it  is  forty  to  fifty  feet  below.  I  regard  it  as  one  of  the  best 
building  materials  in  the  State. 

The  second  variety  is  a  granular  cream  colored  rock,  and 
rather  destitute  of  shells.  It  might  be  mistaken  for  an  oolite. 
The  grain  is  uniform,  and  like  the  preceding  is  soft,  when  first 
taken  from  the  quarry,  but  becomes  hard  as  any  rock  after  an 
exposure  to  the  air  for  a  few  months.  This  rock  is  not  dis- 
posed to  disintegrate,  and  hence  in  this  respect  is  superior  to 


granite. 


This  granular  variety  occurs  in  "Wayne  county.     The  rocks 


NOKTH-CAROLINA   GEOLOGICAL   SURVEY.  91 

or  consolidated  parts  of  it  are  abundant  on  the  plantation  of 
Maj.  Collier. 

At  a  few  places  it  is  sufficiently  pure  to  be  burnt  for  lime  ; 
as  a  general  rule  it  contains  too  much  silex  to  make  a  strong 
lime. 

The  rock  on  Maj.  Collier's  plantation  contains : 

Silica,   59.400 

Peroxide  of  iron  in  combination  with  )  4  120 

alumina  and  phosphoric  acid,            i  ' ' 
Carbonate  of  lime  and  a  trace  of  magnesia,  36.480 

100.000 

The  amount  of  carbonate  of  lime  is  variable,  and  ranges 
in  the  consolidated  varieties  from  30  to  75  per  cent.  The 
silex  in  the  rock  exists  in  grains  as  sand,  which  are  visible, 
but  a  soluble  silica  is  no  doubt  the  cementing  material,  which 
of  course  once  existed  in  solution,  or  in  a  state  of  minute 
subdivision.  This  marl  may  be  used  in  building,  or  if  suffici- 
ently pure  and  free  from  sand  and  silica,  it  may  be  burnt  for 
lime,  which  will  be  adapted  to  agricultural  purposes.  Its 
composition  fits  it  for  this  purpose  as  it  contains  a  small  pro- 
portion of  phosphoric  acid. 

§  63.  The  green  sand  is  frequently  partially  consolidated, 
but  never  forms  a  building  material.  For  agriculture,  when 
the  amount  of  potash  is  considered,  it  is  the  most  important 
of  the  marls.  In  North-Carolina  I  have  found  no  locality 
where  its  potash  equals  that  of  New  Jersey.  This  I  attribute 
in  part  to  our  inability  to  reach  strata  which  are  upon  the 
same  geological  level,  though  it  is  probable  that  the  amount 
of  sand  will  be  greater,  and  hence  diminish  proportionally 
the  amount  of  available  fertilizing  matter. 

The  lowest  mass  accessible  at  Blackrock  I  found  by  analy- 
sis, has  the  following  composition  : 

Silex  and  sand,  37.000 

Peroxide  of  iron  and  alumina,  6.400 

Carbonate  of  lime,  33.400 

Phosphates  of  peroxide  of  iron,   1.600 

Soluble  silica,  1.460 


NORTH-CAROLINA   GEOLOGICAL   SURVEY. 

Magnesia, 13.600 

Potash,   1.431 

Soda,   2.12S 

Organic  matter,  1.600 

Water, 1.800 

100.614 

The  sand  is  frequently  in  quite  large  angular  grains.  That 
part  of  the  bed  which  is  green,  or  properly  green  sand,  is  not 
so  distinct  as  in  New  Jersey,  and  it  would  be  impossible  to 
separate  the  grains  mechanically,  while  in  New  Jersey  they 
may  be  separated  from  the  other  materials.  These  grains 
have  been  analyzed  by  Prof.  Cook,  who  has  found  them  com- 
posed of 

Silica,    45.510 

Protoxide  of  iron,  21.134 

Alumina,    7.960 

Magnesia,  2.400 

Potash,  6.748 

Lime, 3.842 

Phosphoric  acid,  0.990 

Sulphuric  acid, 1.129 

Carbonic  acid,  0.563 

Sand, 0.850 

Water,    9.110 

100.209 

It  has  been  found  that  the  green  grains  in  the  green  sand 
possess  a  very  uniform  composition,  and  that  taking  the  aver- 
age analysis  of  several  specimens  the  grains  contain  silica, 
protoxide  of  iron,  alumina,  magnesia,  potash  and  water  in 
nearly  equal  proportions,  while  the  other  constituents  are 
variable.  The  absence  of  the  green  grains  in  the  marl  of 
black  rock  may  account  for  the  small  percentage  of  potash 
which  is  the  principal  element  relied  upon  in  the  New  Jer- 
sey marl.  The  lime  and  magnesia  of  the  Blackrock  marl  is 
much  greater  than  any  of  the  New  Jersey  beds,  and  the  sand 
and  silica  are  not  in  great  excess.  It  really  has  as  much  fertil- 
izing matte'r  as  the  New  Jersey  marl,  but  it  is  deficient  in  the 
most  valuable  part,  potash.  This  element,  however,  seems 


NORTH-CAROLINA   GEOLOGICAL   SUKVEY.  9& 

to  be  replaced  by  soda,  which  no  doubt  takes  the  place  of 
potash  in  many  vegetables  where  ash  is  rich  in  the  alkalies. 

§  65.  The  sand  of  the  marl  beds  of  New  Jersey  varies  from 
39  to  70  per  cent. ;  the  remainder  of  which  is  more  or  less 
valuable  in  agriculture. 

The  phosphate  of  lime  is  probably  the  most  variable  in  ite 
quantity  of  all  the  valuable  elements,  and  it  is  regarded  as  a 
mixture,  and  not  forming  a  chemical  union  with  either  of  its 
elements.  Indeed  it  may  in  many  specimens  be  seen  and 
distinguished  by  its  greenish  gray  color. 

But  it  is  never  evenly  distributed  through  the  bed,  as  it  has 
been  ascertained  by  analysis,  that  it  has  occasionally  accumu- 
lated in  the  inside  of  shells.  It  is,  however,  always  present 
in  the  marl,  and  it  no  doubt  exerts  a  favorable  influence  upon 
vegetables. 

The  upper  bed  at  Blackrock  differs  in  composition  from 
the  lower.  It  is  less  gritty  to  the  touch,  is  of  a  darker  green, 
more  compact,  and  resembles  a  dark  green  clay.  The  sand 
in  it  is  greater  in  quantity  than  in  the  lower,  bat  is  much  finer. 

On  submitting  it  to  analysis  I  found : 

Sand  or  silex, s. .  93.4S 

Peroxide  of  iron  and  alumina,  9.00 

Carbonate  of  lime, 11.40 

Magnesia,   0.20 

Potash,   0.38 

Soda,  0.42 

Organic  matter,   4.80 

Water, 3.80 

100.43 

The  specimen  submitted  to  analysis  was  taken  near  the 
upper  part  of  the  bed,  about  four  feet  above  the  line,  along 
which  the  exogyra  are  the  most  numerous. 

The  results  which  I  have  finally  obtained  by  the  analysis 
of  the  green  sand  at  Blackrock  have  disappointed  me.  I 
expected  at  least  twice  as  much  potash  as  I  have  been  able 
to  obtain ;  still  when  the  green  sand  is  carefully  examined 
under  the  microscope  it  shows  such  a  large  intermixture  of 
sand,  and  such  imperfect  green  grains  of  the  silicates,  that 


NORTH-CAROLINA   GEOLOGICAL   SURVEY. 

would  lead  any  one  to  expect  on  analysis  unfavorable  re- 
sults. 

The  upper  bed  has,  however,  been  tested  as  a  fertilizer, 
and  very  excellent  results  have  been  obtained  by  its  use. 

The  field  immediately  adjoining  the  bed  of  green  sand 
had  become  so  much  exhausted  that  it  produceed  but  three 
barrels  of  corn  to  the  acre.  Its  employment  the  first  year 
doubled  the  product  of  the  field.  The  quantity  employed 
was  about  two  hundred  bushels  to  the  acre.  The  stalks  of 
corn  previous  to  its  use  were  but  little  larger  than  the  finger, 
and  about  half  as  long  as  the  common  growth  in  this  latitude. 

Previous  to  my  last  analysis  of  the  marl  of  this  locality  I 
had  hopes  that  it  was  sufficiently  rich  and  valuable  for  trans- 
portation to  the  county  of  Chatham.  If,  however,  on  farther  , 
examination,  beds  can  be  found  which  contain  from  four  to 
six  per  cent,  of  potash,  there  is  no  doubt  it  may  be  freighted 
in  return  boats  to  several  points  along  the  Deep  river. 

§  66.  The  value  of  this  species  of  marl  is  estimated  from 
the  amount  of  potash  and  phosphoric  acid  which  it  contains. 

The  price  of  marl  in  New  Jersey  is  about  eight  cents  per 
bushel.  A  bushel  weighs,  when  it  is  wet  from  the  bed,  one 
hundred  pounds.  It  loses,  on  drying  in  the  atmosphere, 
twenty  pounds. 

The  New  Jersey  fertilizer  company  deliver  marl  on  board 
of  vessels  at  their  wharf  for  nine  cents  per  bushel,  and  the 
white  horse  marl  is  delivered  on  the  line  of  railroad,  not  ex- 
ceeding ten  miles  from  the  beds  or  pits,  for  ninety  cents,  per 
ton.  The  potash  in  the  different  beds  of  New  Jersey  varies 
from  two  to  seven  per  cent.,  very  rarely  as  high  as  the  last 
figure.  At  the  pits  individuals  pay  for  marl  from  twenty-five 
to  seventy-five  cents  per  ton  provided  they  perform  the  labor. 
The  value  of  the  potash  in  marl  has  been  estimated  at  four 
cents  per  pound.  Soluble  phosphoric  acid  is  estimated  at 
five  cents  per  pound,  and  the  insoluble  at  two.  But  this  dis- 
tinction is  uncalled  for,  inasmuch  as  all  the  phosphoric  acid 
becomes  available  in  time.  The  soluble,  it  is  true,  is  more 
rapid  in  its  effects,  and  produces  more  immediate  results :  it 
is  no  better  for  permanent  improvements.  Pro£  Way,  chem- 


NOKTH-CAROLINA   GEOLOGICAL   SURVEY.  95 

1st  to  the  royal  agricultural  society  of  England,  has  estimated 
the  soluble  phosphoric  acid  at  eight  and  a  half  cents  per 
pound,  and  the  insoluble  at  three. 

It  must  be  recollected  that  in  order  to  bring  phosphoric 
acid  to  a  soluble  condition  it  requires  considerable  expense. 
It  is  better  to  purchase  what  is  called  the  insoluble  or  tribasic 
phosphates  than  the  soluble  ones  which  are  found  in  our 
markets  and  sold  as  superphosphate  of  lime. 

The  actual  value  of  the  mineral  fertilizers  to  farmers  is  a 
question  quite  different  from  that  which  considers  the  value 
of  bone  dust,  or  potash  by  the  pound.  Immense  benefits 
have  been  secured  by  the  use  of  marl,  which,  considered  in 
a  commercial  point  of  view,  was  worth  nothing.  The  phos- 
phoric acid  in  a  bushel  of  shell  marl  is  not  worth,  in  com- 
merce, a  penny ;  but  for  use  on  worn  out  lands  the  farmer  is 
enriched  more  than  one-fourth  of  a  dollar  after  paying  for  the 
labor  of  raising  and  applying  it. 

We  are  not,  however,  to  confine  our  estimates  of  the  value 
of  a  marl  from  its  phosphoric  acid  and  potash.  Excluding 
the  sand  and  insoluble  silica,  all  the  soluble  matters  are  valu- 
able to  the  farmer  as  fertilizers,  and  hence  the  determination 
of  how  much  is  soluble,  and  how  much  insoluble,  is  a  more 
correct  mode  of  getting  at  the  value  of  marl  than  by  confin- 
ing our  estimates  to  the  two  elements  referred  to. 

These  remarks  apply  only  to  the  value  of  a  marl  for  the 
private  use  of  an  individual  owner,  wTho  employs  his  own 
hands  in  raising  it  when  there  is  the  least  to  do  and  economises 
his  expenses  to  the  best  advantage. 

Marl,  however,  in  its  crude  state,  as  it  exists  in  the  pits, 
has  a  value  which  admits  of  estimation.  The  common  shell 
marl  may  be  hauled  very  frequently  from  two  to  four  miles, 
and  give  profitable  returns.  This  is  often  done.  The  shell 
marl,  however,  will  not  bear  transportation  as  far  as  the  green 
sand  of  Blackrock. 

§  67.  I  have  alluded  already  to  the  difficulty  of  recognising 
certain  marl  beds  in  consequence  in  part  of  the  absence  of 
characters  upon  which  geologists  can  rely.  Among  the  beds  of 
which  there  are  doubts  respecting  their  epoch,  I  find  a  green 


«*0  NORTH-CAROLINA   GEOLOGICAL    SURVEY. 

sandy  deposit,  which,  if  mineralogical  characters  may  be  re- 
lied upon,  would  be  referred  to  the  green  sand  which  is  now 
under  consideration.  They  contain  the  green  sand  grains, 
but  the  characteristic  fossils  are  absent  except  in  one  or  two 
localities.  The  formation  in  question  exjsts  beneath  the  white 
or  brownish  shell  marl  at  Mr.  Flowers,  Bladen  county,  King- 
ston, Lenoir  county,  on  the  Neuse,  and  at  Tawboro',  on  the 
Tar  river,  and  at  many  intermediate  points  on  the  banks  of 
the  creeks  and  ravines.  It  always  occupies  a  position  inferior 
to  the  shell  marl,  but  as  the  latter  are  frequently  absent,  beds 
of  sand  and  clay  immediately  succeed  it.  The  green  sandy 
beds  at  Mr.  Flowers,  beneath  his  shell  marl,  contain  a  few 
specimens  of  the  Ostrea  falcata,  and  at  one  or  two  of  the  bluffs 
above  Mr.  Flowers,  on  the  Cape  Fear,  I  found  the  vertebra 
of  a  large  saurian,  which  I  am  confident  belongs  to  the  green 
sand,  but  in  both  of  these  cases  their  occurrence  in  these  beds 
may  have  been  accidental.  I  am  inclined,  however,  in  view 
of  the  few  facts  which  bear  upon  the  question  of  age,  to  refer 
these  green  sandy  beds  to  the  cretaceous  system,  occupying 
probably  a  position  above  these  beds  which  have  been  de- 
scribed at  Blackrock. 

The  predominent  element  of  these  beds  is  sand  :  if  a  sample 
is  washed,  a  coarse  sand  remains,  which  amounts  to  two- 
thirds  or  three-fourths  of  the  whole  quantity  employed.  The 
quantity,  in  a  few  instances,  may  not  exceed  60  per  cent. 
Notwithstanding  the  large  percentage  of  sand,  it  has  been 
successfully  employed  as  a  fertilizer.  I  have,  therefore,  sub- 
mitted several  specimens  to  analysis,  taken  from  different 
beds  extending  from  the  waters  of  the  Cape  Fear  to  the 
Tar. 

A  representation  of  the  composition  of  this  formation,  as 
it  exists  at  Mr.  Flowers,  in  Bladen,  and  at  Kinston,  on  the 
Neuse,  is  given  in  the  following  analysis. 

§  68.  The  Kinston  green  sand  marl  is  of  a  dark  green  color 
in  the  bed,  but  becomes  lighter  when  dry.  Imperfect  speci- 
mens of  an  Ostrea  occur  in  it,  but  too  much  broken  to  be  de- 
termined. It  contains: 


NORTH-CAROLINA  GEOLOGICAL   SURVEY.  97 

Sand, 91.000 

Peroxide  of  iron  and  alumina,  , 4.700 

Lime, 1.000 

Magnesia, 0.7oO 

Potash,  0.230 

Soda, 0.260 

Water,  . .  1.500 


Soluble  silica,  0.204 

99.634 


The  marl,  or  this  variety  of  green  sand  at  Kingston,  is  one 
of  the  most  sandy  varieties  known.  It  was  regarded  as  too 
sandy  to  require  the  analysis  to  which  it  was  submitted ;  but 
as  the  marl  bed  only  one  mile  above  had  been  successfully 
employed  as  a  fertilizer,  and  appears  to  be  equally  charged 
with  this  useless  element,  I  was  desirous  of  knowing  how  this 
fact  could  be  explained.  It  will  be  seen  that  the  nine  per 
cent,  of  fertilizing  matter  is  really  rich  in  potash,  soda  and 
lime,  and,  therefore,  where  a  heavy  dressing  is  applied,  quite 
a  large  amount  of  this  matter  is  added  to  the  soil,  and  which 
contains  a  small  quantity  of  potash.  The  sulphuric  acid  was 
nob  determined,  but  all  of  these  beds  contain  it,  which  is  no 
doubt  derived  from  the  sulphuret  of  iron  or  pyrites,  which  is 
always  present. 

An  unfinished  analysis  of  a  parcel  taken  from  a  bed  which 
occupies  a  similar  geological  position  on  the  plantation  of  Col. 
Green,  of  Craven  county,  gave : 

Silex  or  sand,  83.20 

Peroxide  of  iron  and  alumina,  9.00 

Lime, 2.31 

Magnesia,  0.50 

Water, 2.60 

It  lies  beneath  a  white  eocene  marl,  has  a  deep  green  color 
in  the  bed,  but  becomes  brown  after  being  exposed  to  the 
atmosphere.  It  has  not  been  used  as  a  fertilizer,  but  is  un- 
doubtedly richer  than  the  Kingston  marl  which  produces  good 
eifects  upon  corn. 

A  similar  composition  obtained  in  the  same  beds  upon  the 
8 


NORTH-CAROLINA   GEOLOGICAL   SURVEY. 

Tar  river.     A  marl,  for  example,  which  has  been  used  as  a 
fertilizer  by  Hon.  R.  R.  Bridges,  contains : 

Sand  or  silica,  89.700 

Peroxide  of  iron  and  alumina,  5.000 

Lime, 1.500 

Magnesia, 0.200 

Potash  and  soda,  0.250 

Water, 3.510      , 

100.151 

It  is  evident  this  variety  of  marl  cannot  be  transported  far 
because  of  its  excess  of  sand,  and  in  the  instances  in  which 
it  has  been  employed  it  has  been  transported  only  a  short 
distance.  These  marls,  however  weak  as  they  may  appear, 
frequently  destroy  the  existing  vegetation.  It  is  due  to  the 
existence  of  decomposing  sulphuret  of  iron,  which  forms  an 
astringent  salt,  copperas,  or  a  mixture  of  sulphate  of  iron  and 
alumina.  This  injurious  salt  is  not  formed  where  there  is  a 
sufficient  quantity  of  lime  to  neutralize  the  salt,  in  which 
case  gypsum  will  be  formed.  It  should  be  remarked  that  the 
astringent  salts  may  exert  a  beneficial  influence  where  they 
are  formed  only  in  small  quantities. 

Another  similar  outcrop  of  this  sand  appears  in  the  bed  of 
a  creek  adjacent  to  the  dwelling  of  Col.  Clark,  in  Tawboro'. 
On  submitting  this  marl  to  analysis  I  found  it  composed  of 

Sand, 91.300 

Peroxide  of  iron  and  alumina,  5.800 

Carbonate  of  lime, 0.190 

Magnesia,  0.130 

Potash,  O.lnO 

Soda,  0.130 

Sulphuric  acid,  0.300 

Water, 1 .200 

99.200 

A  thin  bed  of  the  supposed  upper  part  of  the  green  sand 
formation  appears  in  the  series  of  beds  on  the  banks  of  the 
Tar  river,  three  miles  from  Tawboro'.  At  this  bank  the  shell 
marl  occurs  in  place,  and  has  been  used  as  a  fertilizer  by  Col. 


NOKTH-CAKOLINA   GEOLOGICAL   SURVEY.  99 

Clark  with  good  success  for  many  years  ;  the  relative  position 
of  this  upper  bed  of  green  sand  is  represented  in  a  section 
already  described.  It  lies,  as  will  be  seen,  immediately  be- 
neath the  shell  marl ;  and  beneath  the  green  sand  a  gray 
sand  crops  out,  which  is  quite  consolidated,  and  to  the  eye 
appears  much  like  a  limestone  formation,  but,  as  will  appear 
in  the  sequel,  is  a  bed  of  sand  of  unknown  thickness. 

The  upper  mass  of  green  sand,  which  does  not  exceed  four 
feet,  has  a  similar  composition  to  those  already  noticed.  It 
is  composed  of 

Sand 79.000 

Peroxide  of  iron  and  alumina,  8.800 

Carbonate  of  lime, 2.752 

Magnesia, 1.600 

Potash,  1.739 

Soda,  0.300 

Soluble  silica, 0.600 

Sulphuric  acid 0.200 

Organic  matter,  2.000 

Water,    2.330 

99.321 

§  69.  Although  the  proportion  of  sand  is  large  in  this  marl, 
yet  I  believe  it  is  a  more  valuable  fertilizer  than  the  shell 
marl  above  it. 

It  contains  more  potash  than  the  green  sand  of  Black  rock 
on  the  Cape  Fear.  It  contains,  it  is  true,  less  lime,  but  if  the 
composition  of  the  ash  of  the  cotton  stalk  is  consulted  it  will 
be  perceived  that  magnesia  is  also  required — this  marl  con- 
tains a  large  percentage  of  this  substance. 

It  may  be  regarded  as  containing  seventeen  or  eighteen 
per  cent,  of  fertilizing  matter.  ~No  trial  has  been  made  of 
this  stratum,  and  of  course  nothing  can  be  said  upon  the 
ground  of  trial. 

§  70.  A  very  useless  bed  of  gray  sand  occupies  the  bank 
at  the  water's  edge,  which  has  been  alluded  to.  Neverthe- 
less, I  submitted  a  specimen  of  it  to  analysis.  It  is  one  of 
those  beds  which  is  charged  with  sulphuret  of  iron,  and  forms 
astringent  salts,  on  decomposition,  of  the  sulphuret  of  iron 


100  NOKTH-OAEOLESTA  GEOLOGICAL  STJKVEY. 

which  is  diffused  through  it.  Beds  of  this  description  may 
be  known  by  pouring  muriatic  acid  over  the  material  when 
a  large  quantity  of  sulphuretted  hydrogen  is  liberated,  which 
has  the  odor  of  rotten  eggs — the  smell  of  which  is  not  usual- 
ly forgotten. 
This  bed  is  composed  of 

Fine  Sand,  93.500 

Peroxide  of  iron  and  alumina,  2.000 

Lime,    trace, 

Magnesia,  trace, 

Sulphuric  acid,  1.000 

Water,    3.200 

Potash  and  soda,  (undetermined,)  

99.700 

The  bed  is  partially  consolidated.  It  is,  without  doubt,  en- 
tirely worthless  as  a  fertilizer.  As  a  geological  formation  it 
may  probably  be  regarded  as  one  of  the  beds  of  sand  which 
separate  the  different  beds  composing  the  green  sand  proper; 
still,  no  opportunity  has  as  yet  been  furnished  me  to  see  what 
lies  beneath  it. 

The  foregoing  analyses  of  the  green  sand  furnish  all  the 
necessary  information  respecting  its  composition.  These  beds 
in  North-Carolina  are  deficient  in  potash,  an  element  which, 
in  New  Jersey  and  Delaware,  give  to  this  fertilizer  its  im- 
portance. It  is  possible  that  exposures  of  other  parts  of  this 
formation  may  come  to  light,  which  will  be  richer  in  potash. 
We  do  not  obtain  access  to  the  best  parts,  which  may  be 
richer  in  this  element.  Other  analyses,  therefore,  of  n^ew 
beds  may  result  in  better  success,  and  finally  furnish  a  fertil- 
izer equally  rich  with  those  of  New  Jersey. 


NORTH-CAROLINA  GEOLOGICAL  SURVEY.  101 


CHAPTEK  VIII. 

Eocene  or  white  marl — Quantity  or  per  centage  of  lime  variable,  but 
greater  usually  than  in  the  other  varieties. — The  "Wads worth  beds. — 
His  letter  and  remarks. — Beds  upon  the  Neuse. — Haughton's  marl. — 
Composition,  etc. 

§  71.  In  the  ascending  order,  the  next  series  of  marls  be- 
long to  that  division  of  the  formation  which  is  known  as  terti- 
ary, and  that  part  of  it  which  is  called  the  eocene.  This  part 
is  the  oldest  section  of  the  division,  and  hence,  reposes  upon 
some  part  of  the  cretaceous  system ;  either  the  green  sand, 
which  has  been  already  considered,  or  else  upon  the  chalk, 
as  is  the  case  in  Europe. 

Considered  as  a  marl,  it  is  readily  distinguished  from  the 
green  sand,  even  where  its  relations  are  concealed.  The 
color  is  white,  or  else  a  light  drab,  or  cream  colored,  and  is 
very  frequently  made  up  of  grains,  which,  when  examined 
under  the  microscope,  are  found  to  be  fragments  of  organic  re- 
mains, such  as  corals,  shells  and  echinoderms.  Some  beds, 
ten  feet  or  more  thick,  are  a  mass  of  small  fragments  of 
fossils,  mixed  with  sand.  Some  have  a  chalky  whiteness, 
others  take  a  brownish  tinge.  These  beds  are  frequently 
soft,  and  may  be  loaded  into  a  cart  like  dirt.  In  other  cases, 
consolidation  has  taken  place  in  part,  and  the  mass  is  known 
as  stone  marl.  This  variety  of  marl  is  more  calcareous  than 
the  green  sand  below,  or  the  shell  marl  above,  and  when  the 
mass  is  consolidated  it  makes  a  tolerable  lime  for  agricultural 
purposes.  But  sand,  which  is  a  constant  part  of  all  forma- 
tions in  the  eastern  counties,  exists  in  large  proportions  in 
some  beds,  and  usually  exceeds  fifty  per  cent.  But  some 
beds  have  seventy  or  eighty  per  cent  of  lime,  and  when  thus 
charged,  the  lime  is  well  fitted  for  mortar,  or  whitewashing, 
as  well  as  for  agriculture. 

§  72.  The  eocene  marl  occupies  a  narrow  but  an  ill-defined 
zone,  stretching  across  several  of  the  eastern  counties,  from 
the  lower  waters  of  the  Cape  Fear,  in  Hanover  county, 


102  NORTH-CAROLINA   GEOLOGICAL   SURVEY. 

through  a  part  of  Onslow,  Jones  and  Craven  counties,  cross- 
ing the  Neuse  twenty  miles  above  ISTewbern,  where  it  is 
either  lost  in  the  low  grounds,  or  may  be  discontinued  before 
it  reaches  Beaufort  county,  as  the  only  marls  of  the  lower 
waters  of  the  Tar  belong  to  the  shell  marl,  or  miocene  beds ; 
where  the  next  bed  below  is  visible,  it  is  known  to  belong  to 
the  upper  part  of  the  green  sand,  which  has  been  described. 

The  eocene  is  known  to  exist  at  Wilmington,  at  Pollocks- 
ville,  in  Jones  county,  and  underlies  the  whole  country  in  the 
vicinity  of  Newbern,  upon  the  Neuse.  In  this  formation  I  in- 
clude the  consolidated  beds  which  have  been  employed  for  mill 
stones,  and  which  consists  of  a  mass  of  the  casts  of  shells,  the 
most  common  of  which  is  a  small  species  of  clam.  Recently, 
this  variety  has  become  an  important  building  stone,  and  has 
been  employed  for  enclosing  the  cemetery  at  Newbern,  for 
which  it  is  more  suitable  than  any  other  rock  which  could 
have  been  procured. 

§  73.  It  will  be  seen  from  the  foregoing  remarks,  that  it 
occupies  a  less  area  than  the  green  sand,  and  it  will  also  prove 
to  be  more  limited  than  the  shell  marl,  though  the  latter 
never  forms  a  continuous  deposit  over  a  large  area.  When 
in  rocks,  or  consolidated,  it  is  also  broken  up  or  traversed  by 
fissures,  and  forms,  if  at  the  top  of  the  ground,  a  very  irreg- 
ular surface. 

§  74.  The  white  eocene  marl  has  been  used  as  a  fertilizer, 
and  probably  with  results  as  striking  as  the  common  shell 
marl.  It  would  seem  to  possess  some  advantage  over  other 
marls,  except  the  green  sand,  especially  as  it  is  fine  and  earthy. 
It  is  also  richer  in  lime.  For  analysis  I  have  selected  several 
specimens  from  the  central  part  of  the  region  where  it  is  un- 
derlaid with  it. 

The  marl  of  Wm.  Wadsworth,  Esq.,  of  Craven,  furnishes  a 
kind  which  represents  its  characteristics  in  as  much  perfect- 
tion  as  any  of  the  beds  of  the  county.  I  found  it  compos- 
ed of 


Sand,  26.60 

Water, 1.70 

Magnesia,  0.10 


NORTH-OAEOLINA   GEOLOGICAL   SURVEY.  103 

Carbonate  of  lime, 71.22 

99.62 

The  sand  is  in  the  form  of  white  grains,  often  coarse.  It  is 
a  soft,  earthy  marl,  and  is  made  up  of  fragments  of  corals, 
shells,  crinoid's  or  pentacrinites,  with  sand  mechanically  mixed. 

The  influence  of  this  marl  upon  vegetation  has  always  been 
favorable,  and  the  testimony  of  Mr.  Wadsworth,  whose  ample 
experience  qualifies  him  to  advance  an  opinion,  fully  sustains 
the  foregoing  statement. 

I  subjoin  an  interesting  letter  from  Mr.  Wadsworth  upon 
the  subject  of  marl  and  marling.  His  observations,  I  have 
no  doubt,  will  be  concurred  in  by  his  neighbors.  I  am  the 
more  desiro»us  of  making  his  letter  public  on  account  of  his 
experiment  with  marl  upon  his  premises  for  the  purpose  of 
counteracting  the  tendency  to  fever  and  ague  during  the  au- 
tumnal months.  If  farther  trial  should  confirm  the  opinion 
expressed  in  favor  of  the  use  of  marl  as  a  preventive  of 
fever,  the  importance  of  the  discovery  cannot  be  over-esti- 
mated : 

CORE  CREEK,  GRAVEN  COUNTY,  ) 
May  TO,     1857.      \ 

PROF.  E.  EMMONS — Sir: — The  marl,  (a  specimen  of  which  is  sent,)  I 
have  been  applying  since  1852.  I  have  now  marled  220  acres.  I  have, 
until  this  year  and  a  portion  of  the  last,  applied  100  bushels  to  the  acre. 
I  am  now  using  75.  The  weaker  parts  of  my  land  were  burned  with  the 
former  quantity.  My  land  varies  from  a  very  stiff  clay  to  a  soil  quite  light. 
Presuming  you  will  be  willing  to  be  troubled  with  it,  I  will  give  you  my 
mode  of  using  it,  and  the  results:  My  carts  are  made  to 'hold  just  five 
bushels.  I  have  the  land  checked  off  with  the  plough  into  as  many  squares 
to  the  acre  as  I  design  putting  on  bushels  of  marl.  One  bushel  is  put  into 
each  square.  The  first  four  bushels  is  pulled  out  with  a  hoe  from  the  tail 
of  the  cart,  and  the  last  one  is  dumped. 

By  this  method  I  am  enabled  to  have  the  material  much  more  equally 
spread,  which  I  think  is  a  full  equivalent  for  the  extra  trouble.  I  usually 
begin  to  haul  after  my  crop  is  "laid  by,"  and  it  remains  in  the  heaps  until 
about  the  following  February,  when  it  is  spread  and  ploughed  in.  1  have 
spread  some  and  let  it  lay  on  the  surface  twelve  months  before  it  was 
turned  under,  but  I  never  saw  any  advantage  from  it.  I  have  a  small  piece 


104:  NORTH-CAROLINA   GEOLOGICAL   SURVEY. 

of  very  poor  land  that  has  been  lying  in  that  condition  since  the  first  of 
the  year  1854.  It  was  designed  as  an  experiment.  The  groM7th  on  it  when 
it  was  marled  was  altogether  broom  straw ;  there  is  now  mixed  with  that 
growth  some  briars,  dog  fennel,  and  other  weeds.  I  have  consequently  in- 
ferred there  was  some  improvement,  but  whether  it  is  as  great  as  on  land 
that  was  marled  and  cultivated  I  shall  not  know  until  I  cultivate  it. 

The  land  I  have  marled  and  cultivated  has  very  considerably  improved. 
My  whole  crop  has  very  nearly  doubled,  notwithstanding  one-fifth  of  the 
land  I  crop  on  is  yet  unmarled. 

I  cultivated  the  land  every  other  year  in  corn,  and  it  rested  the  other, 
and  not  pastured.  Last  year  I  sowed  peas  on  a  portion  of  the  rested  land; 
what  will  be  the  result  I  am  now  unable  to  say.  I  have  used  plaster  on 
the  marled  land,  and  have  not  seen  any  beneficial  effect. 

I  fear  I  am  trespassing  too  much  on  your  time ;  I  will,  however,  say  a 
few  words  on  my  experience  of  the  effects  of  liming  on  the  health  of  the 
place.  Before  marl  was  used  on  this  plantation  it  was  uncommonly  sickly, 
so  much  so  that  I  was  compelled  to  carry  my  family  away  every  fall 
Scarcely  a  person,  white  or  black,  escaped  the  ague  and  fever,  if  he  had  no 
more.  All  the  land  around  the  house  has  been  marled,  and  the  yard,  under 
the  houses^  under  and  around  the  negro  houses,  I  keep  freshly  marled 
every  summer.  Last  summer  I  made  my  servants  use  it,  as  our  grand 
mothers  used  to  use  sand,  inside  of  the  houses.  Whether  it  is  owing  to 
this,  or  to  a  ditch  I  have  had  cut  through  the  yard,  or  whether  it  is  an  ac- 
cidental occurrence  I  can't  say,  but  fall  before  last  there  was  not  a  chill  on 
the  premises,  and  last  fall  there  was  but  one  case. 

I  will  trouble  you  with  one  more  result:  These  premises  were  infested 
with  ants  and  fleas,  now  such  animals  are  hardly  known  here. 

W.  B.  WADSWORTH. 
i 

§  75.  In  a  subsequent  letter  Mr.  Wadsworth's  remarks  go 
to  confirm  his  previously  expressed  opinions,  but  that  the 
reader  may  be  benefitted  by  Mr.  "WVs  experience,  I  subjoin 
his  remarks  in  his  own  language : 

CRAVEN  COUNTY,  N.  C.,  (NEAR  NEWBERNE,  ) 
October  12th,  1857.      f 

PROF.  E.  EMMONS — Dear  Sir: — The  fever  for  marling  is  spreading  in 
this  part  of  our  county  and  a  good  deal  of  land  will  be  limed  this  winter. 
I  have  given  some  of  mine  an  over  dose  with  only  one  hundered  bushels. 
Last  fall  and  winter  I  used  only  seventy  five  and  now  I  am  putting  on 
fifty.  My  experence  so  far  has  taught  me  to  begin  with  a  very  limited 
quantity  and  to  add  to  it  as  the  land  improves.  Where  I  have  not  burned 
my  land  the  improvement  is  very  satisfactory. 


NOKTH-CAROLINA   GEOLOGICAL   SURVEY.  105 

I  mentioned  in  my  last  letter  to  you  the  effect  that  marling,  or  ditching, 
or  both  combined,  had  had  upon  the  health  of  this  place.  I  told  you  that 
this  plantation  was  remarkably  sickly  previous  to  the  fall  of  1855 — so  much 
so  that  it  was  strange  for  even  one  to  escape  billious,  or  ague  and  fever.  I 
mentioned  that  in  1855  there  was  not  a  case  of  either,  in  1856  but  one, 
and  now  I  will  add  that  so  far  this  fall,  in  a  family  of  forty  persons,  there 
has  been  but  two  cases.  (I  happened  to  have  been  one  of  the  subjects.) 
These  three  falls  have  been  dry.  I  don't  know  how  a  wet  one  would  act 
upon  us.  I  have  kept  marl  plentifully  used  in  my  yard,  and  around  and 
in  my  negro  houses. 

I  shall  be  under  many  obligations  to  you  for  analysis  of  my  marl. 

Yours,  &c., 

W.  B.  WADSWORTH. 

§  76.  A  marl  belonging  to  the  same  epoch,  (eocene)  fur- 
nished by  J.  H.  Haughton,  from  his  plantation  in  Jones 
county,  gave  me  56.06  per  cent  of  carbonate  of  lime.  An- 
other specimen  gave : 

Silex  or  sand, 13.00 

Phosphate  of  peroxide  'of  iron  and  alumina,  1.10 

Carbonate  of  lime,    85.2C 

Carbonate  of  magnesia, 1.02' 

Potash,   0.02 

100.34 

I  have  found  in  these  white  marls  a  small  per  centage  of 
potash.  It  is  evidently  less  than  in  the  other  varieties.  This 
is  made  up  like  the  Wadsworth  marl,  of  fragments  of  fossils, 
in  which  certain  species  of  corals  and  a  crinoid  abound. 

A  variety  is  met  with  which  is  derived  from  the  disinte- 
gration of  a  large  species  of  oyster.  It  occurs  upon  the  plan- 
tation now  owned  by  L.  Haughton,  Esq.,  and  is  known  as 
the  Pollock  place,  in  Jones  county.  It  contains : 

Carbonate  of  lime,   34.54 

Sand, 63.46 

Peroxide  of  iron  and  alumina, 1.30 

99.30 

Large  grains  of  sand  are  distributed  through  the  marl.    It 


NOKTH-CAROLINA  GEOLOGICAL   SURVEY. 

follows  necessarily,  from  the  manner  in  which  these  marls 
have  accumulated,  that  they  should  vary  in  composition,  and 
that  the  substance  which  reduces  the  quantity  of  carbonate 
of  lime,  should  be  sand. 

A  ready  method  by  which  its  quantity  may  be  estimated 
is  by  washing  a  given  quantity.  It  will  be  seen,  that  by  agi- 
tating it  in  a  vessel  of  water,  there  is  a  considerable  quantity 
of  fine,  inpalpable  white  powder.  "Wash  it  until  the- water 
pours  off  clear,  and  the  sand  with  the  coarse  fragments  of 
fossils  remain.  The  existence  of  much  sand  is  not  suspected 
at  first,  but  as  washing  progresses,  it  will  be  found  to  prevail, 
in  some  cases,  over  the  carbonate  of  lime. 

§  77.  Upon  the  Neuse,  about  twenty  miles  above  E"ew- 
bern,  heavy  banks  of  the  marl  under  notice  occur,  which 
extend  continuously  for  more  than  a  mile.  This  exposure  of 
marl  is  upon  the  plantations  of  Samuel  Biddle  and  Benjamin 
Biddle.  It  is  accessible,  and  forms  steep  escarpments  on  the 
south  side  of  the  river.  On  account  of  the  accessibility  of 
this  outcrop  of  marl,  it  will  hereafter  become  an  important 
deposit  from  the  lime  which  it  is  capable  of  furnishing.  It 
is  consolidated,  and  may  be  quarried  for  the  kiln,  but  it  also 
furnishes  an  abundance  of  marl  in  a  fine  state  of  subdivision. 

It  has  been  tried  imperfectly  as  a  fertilizer,  but  while  the 
result  was  disastrous,  we  may  infer  from  it,  that  it  possesses 
as  valuable  properties  as  the  kind  used  by  Mr.  Wadsworth, 
which  has  been  described  already.  The  quantity  used  by 
Mr.  Biddle,  in  his  first  experiment,  was  600  bushels  to  the 
acre ;  consequently,  most  of  the  vegetation  was  killed,  and 
very  little  has  grown  upon  the  land,  thus  excessively  marled, 
for  six  years.  It  is  just  recovering  from  the  dose.  The  con- 
solidated part  of  this  outcrop  of  marl  contains : 

Sand,    20.00 

Carbonate  of  lime,  .  .• 78.60 

Oxide  of  iron  and  allumina,  1.70 

100.30 


NORTH-CAKOLINA   GEOLOGICAL   SURVEY.  107 

Another  specimen  of  consolidated  marl  from  Benjamin 
Biddle's  plantation  (Egypt)  gave  me  : 

Sand, 9.60 

Peroxide  of  iron  snd  alumina,  containing  phosphoric 

acid, 4.40 

Carbonate  of  lime, 85.00 

Magnesia,  trace, 

99.00 

A  few  grains  of  coarse  sand  were  visible  in  the  rock.  This 
mass  is  evidently  sufficiently  pure  for  burning  into  lime.  It 
would  be  adapted  for  the  various  purposes  for  which  lime  is 
required,  as  mortar,  whitewashing,  or  for  agriculture. 


CHAPTER  IX. 

FERTILIZERS CONTINUED. 

Shell  marl. — Heterogeneous  in  its  composition,  and  arrangement  of  its 
materials. — Chemical  constitution. — Application  of  marl. — Poisonous 
marl. — How  corrected. — Theories  respecting  the  operation  of  marl. 

§  78.  The  third  bed  of  marl  in  the  ascending  order  has 
been  appropriately  called  shell  marl,  from  the  great  abun^ 
dance  of  undecomposed  marine  shells,  of  which  it  is  mainly 
composed.  The  mass,  taken  as  a  whole,  is  formed  of  per- 
fect shells,  and  those  which  have  become  fragments,  and 
sand.  There  is  no  order  in  their  arrangement  in  the  bed. 
They  lie  as  if  they  had  been  washed  up  on  a  beach ;  hence, 
they  are  mixed  confusedly  together.  The  relative  position  of 
the  shell  marl  is  exhibited  in  the  sections  already  given.  It 
is  not  present,  however,  even  where  all  the  other  members 
of  the  sections  in  a  bluff  or  outcrop  exists.  Whether  its 


108  NORTH-CAROLINA   GEOLOGICAL   SURVEY. 

absence  is  due  to  denudation,  or  whether  the  beds  were 
formed  only  at  certain  points,  has  not  been  determined.  De- 
nudation, however,  has  taken  place  at  some  of  the  beds,  as 
they  still  preserve  the  gullies  which  were  cut  through  them, 
and  which  were  subsequently  filled  with  brown  earth. 

Although  it  is  not  possible  to  detect  an  orderly  arrange- 
ment of  materials,  still,  certain  parts  occupy  usually  a  com- 
mon position ;  for  instance,  the  large  pebbles,  coprolites-,  and 
certain  bones  and  teeth  lie  at  the  bottom  of  the  stratum. 
The  inference  which  may  be  deduced  from  this  fact  is,  that 
during  the  first  stage  of  its  formation,  there  was  considerable 
violence  in  the  movement  of  the  waters  in  which  the  stratum 
was  accumulating ;  and  that  probably,  prior  to,  and  during 
the  early  part  of  its  accumulation,  there  were  shiftings  of  the 
strata;  some  being  more  elevated,  others  depressed;  or  there 
was  a  change  of  level  of  the  sea  coast,  which  set  in  motion 
the  waters,  and  led  to  the  violence  which  collected  at  the 
bottom  the  large  and  less  destructible  fragments  to  which  I 
have  alluded. 

But  in  the  first  place,  I  propose  to  speak  of  the  use  of  this 
marl  stratum  as  a  fertilizer ;  and  as  it  has  a  more  general  dis- 
tribution, it  has  been  employed  more  extensively  than  either 
of  the  foregoing  which  I  have  described. 

The  beds  of  shell  marl  are  not  composed  uniformly  of  the 
same  elements  in  the  same  proportions.  It  is  as  heteroge- 
neous as  possible  in  this  respect.  Some  beds  contain  ninety 
per  cent  of  sand ;  in  others  it  is  reduced  to  twenty-five  per 
cent,  and  the  remainder  is  mostly  carbonate  of  lime. 

§  79.  The  most  important  subdivision  which  can  be  found- 
ed upon  composition,  is  that  into  a  gray  or  whitish  marl  in 
the  mass,  the  color  of  which  is  due  to  the  great  abundance 
of  marine  shells,  and  that  of  a  dark  bluish  green  marl,  which 
contains  grains  of  green  sand.  In  the  latter  there  is  a  no- 
table amount  of  potash,  while  in  the  former  it  exists  only  in 
very  small  proportions.  Some  recognize  a  red  or  brown 
marl.  This  color,  however,  is  due  merely  to  exposure  to  the 
atmosphere,  in  consequence  of  which  the  protoxide  of  iron 
has  changed,  or  is  changing,  by  the  absorption  of  oxygen 


NOETH-OABOLDSTA  GEOLOGICAL  SURVEY.  109 

into  the  peroxide.  This  change  is  indicative  of  a  valuable 
marl,  but  it  is  no  better  subsequent  to  this  change  than  be- 
fore it.  If  in  the  greenish  marl  green  grains  can  be  distin- 
guished, it  may  be  inferred  that  the  marl  contains  potash. 
The  presence  of  carbonate  of  lime,  as  is  usually  known,  is 
indicated  by  eifervescence  when  acids  are  poured  over  it, 
and  a  judgment  may  be  formed  by  its  continuance  and  vio- 
lence, whether  it  is  rich  in  this  substance.  If  it  is  prolonged, 
there  is  a  large  quantity  of  carbonate  of  lime  in  the  spe- 
cimen under  examination.  So  the  presence  of  sand  may  be 
detected  and  its  quantity  proximately  determined  by  simple 
washing. 

§  80.  The  shell  marl  upon  the  Cape  Fear  river  belongs 
usually  to  the  former.  A  bed,  however,  in  the  bluff  at 
Brown's  landing,  contains  the  green  grains  alluded  to,  but 
still  it  is  readily  distinguished  from  that  upon  the  Tar  river, 
which  is  usually  bluish  green,  and  belongs  to  the  latter  va- 
riety. 1  do  not,  however,  attach  much  importance  to  the 
subdivision. 

There  are  several  beds  of  shell  marl  immediately  upon  the 
banks  of  the  Cape  Fear,  or  within  a  mile  of  them  ;  and  when 
marine  shells  are  closely  packed  in  the  strata  their  several 
compositions  are  alike.  As  a  representation  of  the  compo- 
sition of  this  marl,  I  shall  select  Mr.  Cromarty's  marl  bed, 
near  Elizabeth  town.  It  consists  mainly  of: 

Sand, 52.50 

Carbonate  of  lime,    40.25 

Peroxide  of  iron  and  alumina,  7.20 

Magnesia,  , 0.75 

Potash  and  soda,  traces. 

I  have  always  found  phosphoric  acid  when  the  peroxide  of 
iron  and  alumina  are  tested  with  molybdate  of  ammonia.  It 
is  very  rare  for  the  carbonate  of  lime  to  amount  to  seventy- 
five  per  cent.  I  found  seventy-one  per  cent  in  Mr.  Mc- 
Daniel's  marl,  in  Nash  county.  The  bluish  green  marl  of 
Tar  river  is  quite  sandy,  and  yet  may  be  regarded  as  a  rich 
marl.  As  an  illustration  of  this  fact,  I  subjoin  an  analysis  of 


110  NOETH-OAKOLINA   GEOLOGICAL   SURVEY. 

the  marl  bed  owned  by  Col.  Clark,  three  miles  above  Taw- 
boro',  on  the  Tar  river.     It  consists  of: 

Peroxide  of  iron  and  alumina,  6.80 

Carbonate  of  lime,  16.10 

Magnesia, 0.436 

Potash,  0.616 

Soda,  1.988 

Sulphuric  acid,  0.200 

Soluble  silica, 0.440 

Chlorine,    0.030 

Phosphoric  acid,  0.200 

Sand, 72.600 

Of  one  hundred  parts,  only  about  twenty-six  can  be  re- 
garded as  available  matter,  and  yet  good  results  have  at- 
tended its  use. 

Immediately  above  the  shell  marl  of  the  Tar  there  is  a  bed 
of  clay  some  four  feet  thick.  This  clay  I  have  submitted  to 
analysis  for  the  purpose  of  ascertaining  the  quantity  of  potash 
it  contains.  The  results  show,  however,  that  as  a  fertilizer, 
it  is  of  no  importance.  It  gave  me : 


Sand, 84.00 

Peroxide  of  iron  and  alumina,  4.40 

Lime, 0.35 

Magnesia, 0.10 

Potash,  0.05 

Soda, 0.02 

Soluble  silica,  0.20 

Organic  matter  and  water,  10.50 


All  the  beds  except  the  upper  beds  of  sand  were  submitted 
to  analysis.  Only  two  in  this  bank  are  valuable  fertilizers, 
the  shell  marl  and  the  upper  bed  of  green  sand ;  both  con- 
tain potash,  soda  and  phosphoric  acid ;  and  there  is  no  neces- 
sity for  rejecting  the  latter  when  hauling  marl  for  the  plan- 
tation. If  some  method  could  be  devised  by  which  the  sand 
could  be  cheaply  separated  from  the  mass,  the  remainder 
would  form  a  marl  superior  to  the  richest  green  sand ;  the 


NORTH-CAROLINA   GEOLOGICAL   SURVEY. 


111 


sand  being  coarse,  presents  a  favorable  condition  for  effect- 
ing a  separation. 

§  81.  The  green  shell  marl  of  Mr.  Bridger's  plantation, 
upon  Fishing  creek,  I  found  to  possess  a  composition  similar 
to  Col.  Clark's.  There  is  a  greater  proportion  of  sand,  but 
the  available  part  is  almost  identical  with  the  Tar  river  marl. 

§  82.  The  application  of  marl  is  an  important  matter,  and 
requires  a  brief  discussion.  Notwithstanding  marl  has  been 
used  for  many  years,  still  there  is  much  disagreement  among 
planters  of  experience  as  to  the  best  mode  of  applying  it,  and 
the  quantity  to  be  applied  in  any  given  case.  Its  effects  are 
frequently  deleterious  if  a  large  quantity  is  spread  upon  a 
poor  soil,  and  yet  it  has  not  been  ascertained  how  its  injurious 
effects  may  be  obviated.  It  is  no  doubt  desirable  in  many 
instances  to  use  a  larger  quantity  of  marl  than  the  soil  will 
admit  of  when  it  is  in  its  natural  state. 

The  quantity  of  marl  which  is  usually  spread  upon  an  acre 
of  ground  is  from  150  to  200  bushels.  Three  hundred  bushels 
is  often  used.  But  certain  worn  out  lands  would  be  exceed- 
ingly injured  for  several  years  by  even  two  hundred  bushels. 
The  question,  I  have  no  doubt,  has  been  often  put:  Why  is 
marl  ever  injurious?  The  natural  conclusion  is  that  it  con- 
tains some  substance  unfriendly  to  vegetation.  This  substance 
is  no  doubt  in  certain  cases  an  astringent  salt,  formed  in  those 
marls  which  contain  iron  pyrites  which  is  prone  to  decompose 
on  exposure  to  those  bodies  which  contain  oxygen,  the  sul- 
phur thereby  is  oxidated,  and  slowly  acts  upon  the  iron  and 
forms  copperas,  or  upon  alumina,  which  is  present  in  the  marl. 
In  small  doses  copperas  will  not  fatally  injure  vegetation,  but 
operates  beneficially.  The  term  in  common  use  for  express- 
ing the  effect  of  injurious  marls  is,  burning.  Those  which 
are  decidedly  burning  marls  have  the  distinct  taste  of  cop- 
peras, sometimes  it  appears  upon  the  surface  of  those  marls 
in  dry  weather,  when  it  has  a  whitish  appearance.  But 
gypsum  sometimes  appears  also.  This  may  be  distinguished 
from  copperas  by  being  tasteless. 

§  83.  There  is  no  difficulty  in  treating  marls  in  which  cop- 
peras is  found.  It  is  readily  decomposed  by  lime.  Let  a 


112  NORTH-CAROLINA  GEOLOGICAL   SURVEY. 

compost  heap  containing  a  hundred  bushels  of  marl  be  form- 
ed, mixing  leaves  or  any  organic  matter  as  stable  manure, 
and  then  add  three  bushels  of  quick  lime  to  the  mass,  and 
incorporate  the  ingredients  together  by  shoveling  them  over 
twice.  Gypsurn  will  be  formed  by  combining  with  the  sul- 
phuric acid  in  combination  with  the  iron.  The  compost  is  all 
the  better  for  the  lime,  though  it  is  possible  the  gypsum  may 
not  in  all  instances  prove  itself  useful.  Astringent  marls, 
when  in  heaps  in  the  open  air,  lose  their  copperas  and  other 
soluble  salts  by  solution  in  rain  water  to  which  they  are  neces- 
sarily exposed,  they  undergo  a  leaching  process  by  which 
they  are  freed  of  their  injurious  properties.  Another  method 
may  be  resorted  to  when  it  is  found  that  vegetation  is  being 
injured,  or  has  been  by  the  experience  during  the  year  of  its 
application,  to  plough  deep  and  mix  the  marl  with  a  large 
quantity  of  soil ;  the  fertility  will  be  restored.  It  is  by  no 
means  difficult  for  any  farmer  to  test  his  marl  prior  to  its  use 
if  he  wishes  to  ascertain  whether  this  astringent  salt  is  pre- 
sent. To  do  this,  let  the  marl  be  boiled  in  rain  water ;  strain 
it,  or  let  the  turbidness  of  the  solution  disappear  by  rest; 
pour  off  the  clear  liquid,  and  if  sulphate  of  iron  and  alumina 
is  present,  it  will  turn  black  by  adding  a  solution  of  strong 
tea  to  it ;  it  will  become  a  dirty  white  by  lime  water  and  a 
solution  of  the  leaves  of  red  cabbage  change  it  to  red,  show- 
ing the  presence  of  an  acid  salt.  Most  of  the  marls  of  the 
State  contain  these  salts.  Where  they  are  abundant  unde- 
composed  pyrities  will  be  found  in  masses  adhering  to  por- 
tions of  petrified  wood  or  inseparate  concretions  in  the  marl. 

§  84.  Writers  upon  the  efficacy  of  marl  as  a  fertilizer,  have 
entertained  different  opinions.  As  the  progress  of  agricul- 
ture has  been  promoted,  and  observation  and  experiments 
multiplied  upon  the  effects  of  different  bodies  upon  vegeta- 
tion, these  opinions  have  become  more  consistent  and  reliable. 

Some  writers  have  maintained  that  lime  alone  is  the  effec- 
tive agent;  others  that  it  is  pyrites,  or  else  is  due  to  the 
presence  of  animal  matter,  which  has  been  derived  from  the 
fossils  of  the  beds ;  others,  still,  to  the  presence  of  phosphate 


NORTH-CAROLINA   GEOLOGICAL   SURVEY.  113 

\ 

of  lime,  while  others  have  maintained  that  it  is  due  to  the 
potash. 

§  85.  Now,  it  is  quite  possible  that  all  these  opinions  are 
right  as  far  as  they  go.     They  are  erroneous  in  being  re- 
strictive.    If  we  examine  the  composition  of  an  ash  of  any 
plant,  as  I  have  already  observed,  we  shall  find  all  these  ele- 
ments, and  we  may  well  suppose,  as  they  are  all  so  generally 
present,  that  they  are  all  required  ;  and  hence,  we  are  not  to 
attribute  the  efficacy  of  marl  to  one  of  its  elements  exclusive 
of  the  others.     It  may  be,  that  a  given   soil  is  notably  de- 
ficient in  potash,  while  the  other  elements  are  in  sufficient 
abundance  to  furnish  all  that  a  given  plant  requires.     In  such 
a  case  it  might  appear  that  fertility  was  restored  to  the  soil 
by  potash  alone.     Of  all  fertilizers,  wood  ashes  are  the  best, 
and  possess  a  more  general  application  than  any  other ;  being 
adapted  to  any  crop.     They  are  the  best,  because  they  con- 
tain all  the  elements  the  plant  needs ;  and  hence,  the  nearer 
a   marl  is  in  composition  to  wood  ashes,  the  better  it  is. 
Hence,  then,  the  efficacy  of  marl  is  due  to  its  potash,  soda, 
lime,  iron,  magnesia,  phosphoric  acid,  sulphuric  acid   and 
chlorine,  and  not  any  one  of  its  elements,  exclusive  of  the 
others.     The  only  modification  which  this  doctrine  requires, 
is  that  some  of  the  elements  are  more  important  than  others, 
and  it  may  be  true,  that  the  controlling  influence  is  to  be 
ascribed  to  the  alkalies,  alkaline  earths  and  phosphates ;  still, 
the  marl  is  better  with  the  less  essential  elements,  than  it 
would  be  without  them.     The  absolute  value  of  a  marl  is 
shown :  1.,  by  the  amount  of  soluble  matter  it  contains.     2., 
by  the  predominance  of  the  most  valuable  elements,  as  pot- 
ash and  phosphoric  acid.     Marls  which  contain  the  most  of 
these  bodies  are  the  quickest  and  the  most  durable  in  their 
effects ;  and  when  the  marl  is  rich  in  them,  a  full  dressing 
lasts  from  fifteen  to  twenty  years. 

§  86.  In  forming  a  theory  respecting  the  active  elements 
in  marl,  our  views  should  not  be  limited  to  the  nutrient  prop- 
erties they  possess,  or  simply  to  the  food  elements  which  con- 
tribute directly  something  to  the  weight  or  growth  of  the 
plant. 

9 


114:  NORTH-CAROLINA  GEOLOGICAL   SURVEY. 

Some  elements  perform  a  function  in  growth  or  nutrition  ? 
which  is  independent  of  nutrition  in  this  sense,  or  they  are 
nutritive  from  their  reactive  forces ;  they  are  not  taken  up 
by  the  plant,  but  furnish  or  provide  a  substance  by  their  re- 
actions upon  each  other,  which  is  nutritive  or  administers' 
to  its  growth. 

These  substances  perform  a  double  function  ;  they  are  really 
nutriments,  and  are  taken  up  into  the  vegetable  tissue ;  but, 
in  addition  to  this,  their  reactions  upon  other  matters  in  the 
soil  are  such  that  nutrient  matter  is  constantly  provided  with- 
out their  increase  or  diminution  in  the  soil  or  marl. 

The  substances  which  are  known  to  perform  a  double  office. 
are  the  oxides  of  iron  and  organic  matters.  To  enable  me  to 
give  a  brief  exposition  of  the  functions  of  the  oxides  of  iron, 
I  will  state  what  takes  place  in  the  soil  when  it  is  well  con- 
stituted for  the  growth  of  cereals,  and  other  plants  employed 
as  food.  It  will  be  observed  that  in  the  analysis  of  soils,  the 
iron  is  set  down  as  a  peroxide ;  this  is  the  state  in  which  the 
iron  is  obtained.  In  the  best  of  soils  the  iron  is  not  all  of  it 
in  this  state ;  but  that  of  a  mixture  of  the  two  oxides — the 
protoxide  and  peroxide.  Now,  the  protoxide  is  changed  in 
making  an  analysis  into  the  peroxide,  by  the  addition  of  a 
few  drops  of  nitric  to  the  hydrochloric  acid,  which  is  em- 
ployed for  effecting  a  solution,  for  the  purpose  of  obtaining 
an  exact  or  an  uniform  result.  The  nitric  acid  added  to  the 
solution,  is  deprived  of  so  much  of  its  oxygen  by  the  pro- 
toxide as  is  sufficient  to  change  it,  or  convert  it  to  a  peroxide. 
Now,  in  the  ordinary  course  of  nature,  this  change  takes 
place  when  the  soil  is  freely  exposed  to  the  action  of  water 
and  air.  The  protoxide  passes  into  a  peroxide  by  the  absorp- 
tion of  oxygen  from  the  water.  It  would  remain  in  this  state 
permanently,  if  the  soil  was  dry  and  free  from  vegetable  or 
organic  matter.  "When  soils  become  exhausted  of  these  mat- 
ters, it  remains  a  permanent  peroxide.  If,  however,  this  pe- 
roxide comes  in  contact  with  organic  matter,  it  robs  the  pe- 
roxide of  an  equivalent  of  oxygen,  and  passes  again  into  the 
condition  of  a  protoxide.  It  is  possible,  therefore,  for  these 
changes  to  take  place  at  all  times  when  the  needful  conditions 


NORTH-CABOLHTA   GEOLOGICAL   SUKVEY.  115 

exist.  But  this  is  not  all ;  the  water  of  the  soil  being  robbed 
of  its  oxygen,  its  hydrogen  is  set  free ;  and  being  in  its  nas- 
cent state,  it  is  ready  itself  to  combine  with  that  body,  for 
which  it  has  the  strongest  affinity.  That  body  is  nitrogen 
contained  in  the  air  diffused  in  the  soil ;  and  the  body  formed 
by  this  union  is  ammonia.  Now,  ammonia  is  one  of  the  most 
essential  bodies  in  the  list  of  nutrients.  Guano,  as  is  well 
known,  owes  its  fertilizing  properties  in  part  to  ammonia. 
But  I  need  not  dwell  upon  this  fact.  By  the  interchanges  oi 
oxygen  which  take  place  with  the  oxides  of  iron,  we  are  fur- 
nished with  an  explanation  of  the  origin  of  ammonia  in  the 
soil.  But  the  production  of  ammonia  is  only  one  of  the 
chemical  changes  which  take  place  in  a  soil  in  which  organic 
matter,  iron,  water  and  air  exists.  The  vegetable  matter, 
also,  undergoes  a  change,  for  the  oxygen  which  it  has  taken 
from  the  peroxide  of  iron  converts  it  into  organic  acids,  which 
are  known  by  the  names  of  crenic  and  apocrenic  acids.  These 
acids  being  one  of  the  series  of  changes  effected  through  the 
influence  of  the  oxides,  they  in  their  turn  become  active,  and 
unite  with  the  ammonia  and  form  crenates  and  apocrenates 
of  ammonia.  In  the  condition  of  a  salt,  this  compound  of 
ammonia  and  the  vegetable  acids  are  taken  up  by  the  root* 
of  plants,  and  become  their  food. 

§  87.  I  have  made  these  remarks  for  the  purpose  of  pre- 
paring the  way  for  farther  observations  upon  the  action  of 
marls  upon  vegetation.  The  condition  of  the  iron  in  a  large 
proportion  of  the  marls,  is  that  of  a  protoxide.  Thus  the  iron 
in  the  greenish  marl  upon  the  Tar  River,  is  a  protoxide.  In 
this  condition,  when  it  is  spread  upon  land  and  mixed  with 
the  soil  which  contains  vegetable  or  organic  matter,  change* 
first  into  a  peroxide,  it  is  then  in  an  active  state,  and  sezing 
upon  one  of  the  elements  of  water,  decomposes  it.  The  ox- 
ides of  iron  in  the  marl  undergo  the  same  changes  in  the  soil 
to  which  they  are  applied,  as  those  which  have  been  describ- 
ed as  taking  place  in  all  soils  which  have  not  been  exhausted 
o/  these  organic  matters.  It  will  therefore  be  expected  that 
marls  which  contain  a  large  percentage  of  iron,  are  more  val- 
uable than  those  which  are  destitute  of  it,  and  to  the  actioi. 


110  NO&THK)AfcOLINA   GEOLOGICAL   SURVEY. 

of  its  oxides,  we  are  indebted  for  one  of  its  most  important 
effects,  the  supply  of  the  salts  of  ammonia,  and  even  the  or- 
ganic salts  of  potash,  soda,  and  lime. 

These  facts  furnish  important  hints  relative  to  the  proper 
preparation  of  marl  for  the  plantation,  viz :  that  it  should  be 
Composted  with  organic  matters.  We  supply  in  this  way  the 
conditions  for  its  favorable  action  upon  vegetation.  With  a 
large  quantity  of  organic  matter,  a  large  amount  of  marl  may 
be  used  without  detriment  to  the  vegetation,  and  the  larger 
the  quantity  the  greater  the  amount  of  ammonia  which  will 
be  generated.  For  certain  crops,  this  practice  is  of  the  high- 
est importance.  It  has  been  proved  by  numerous  experiments 
with  wheat,  that  there  is  a  certain  yield  produced  by  the  use 
of  the  mineral  fertilizers  as  phosphates  of  lime,- — but  these 
will  not  increase  the  yield  beyond  a  certain  standard  when 
used  by  themselves.  But  if  a  larger  supply  of  ammonia  is 
furnished,  the  number  of  bushels  per  acre  is  increased  beyond 
that  standard.  So  that  in  order  to  bring  lands  to  their  full 
capacity,  ammonia  must  be  supplied  also  directly,  or  indirect- 
ly. A  compost  of  marl  properly  made,  is  one  of  the  best  fer- 
tilizers for  wheat,  and  there  is  little  doubt,  that  the  favorable 
influence  is  due  in  part,  to  the  chemical  changes  which  I  have 
described  by  which  ammonia  is  one  of  the  products  of  change. 

To  estimate,  therefore,  the  value  of  marl  by  the  number  of 
pounds  of  phophoric  acid  and  potash  which  is  contained  in  a 
ton,  does  not  give  its  true  value.  All  marl  contains  a  small 
amount  of  organic  matter,  but  it  is  improved  by  adding  more, 
and  thus  preparing  it,  we  provide  for  a  continuance  of  those 
changes  by  the  instrumentality  of  iron  until  the  organic  mat* 
ter  is  consumed,  and  when  ammonia  will  cease  to  be  genera* 
ted.  It  will  be  understood,  therefore,  that  organic  matter  is 
necessary  to  effect  these  changes  which  produce  the  salts  of 
ammonia ;  in  its  total  absence,  it  is  true,  ammonia  is  produced  ; 
still,  in  the  state  of  simple  ammonia,  it  is  not  fit  for  nutrition  ; 
it  requires  a  union  with  some  acid,  and  therefore  the  great- 
importance  of  providing  all  the  conditions  for  the  full  action 
of  marl  upon  the  crops  to  which  it  is  applied. 

§  88.  If  the  foregoing  views  are  correct,  it  will  be  admitted 


NORTH-CAROLINA   GEOLOGICAL   SURVEY.  117 

that  the  simple  application  of  the  oxides  of  iron  and  organic 
matter  may  become  the  best  of  fertilizers.  Experience  has 
proved  that  the  scales  of  black  oxide  of  iron,  or  the  oxides  and 
other  refuse  matter  obtained  from  a  smith's  forge  are  excellent 
fertilizers  for  the  pear  and  other  fruit  trees ;  and  they  are  no 
doubt  equally  valuable  for  wheat  and  Indian  corn.  Iron  itself 
is  always  present  in  the  ash  of  a  plant.  It  is  no  doubt  an  im- 
portant element  in  its  organization,  giving  it  tone  and  strength. 
But  as  we  have  attempted  to  explain,  it  is  equally  an  essen- 
tial element  in  soils  and  marls,  for  its  influence  in  effecting 
those  changes  which  finally  result  in  the  production  of  the 
vegetable  salts  of  ammonia,  potash,  soda  and  lime.  It  is  in 
this  state  that  they  are  taken  up  by  the  roots  of  plants  and 
become  thereby  the  effective  agents  of  growth. 

When  the  functions  of  iron  in  a  soil  or  marl  are  known,  it 
does  not  appear  improbable  that  it  is  as  important  and  as  val- 
uable as  phosphoric  acid  or  potash.  In  some  marls  it  is  easy 
to  recognise  the  change  which  the  iron  has  already  undergone 
by  its  having  become  brown  or  reddish.  This  change  does 
not  probably  affect  its  qualities,  though  some  maintain  that 
the  red  marl  is  better  than  the  blue.  The  only  difference  be- 
tween them  is,  that  the  protoxide  of  the  blue  has  passed  into 
peroxide  ;  the  latter  may  be  changed  back  to  the  protoxide 
in  a  soil  charged  with  organic  matter,  and  though  I  have 
omitted  to  state  the  fact,  the  organic  acids  are  capable  of  act- 
ing also  upon  the  oxide  of  iron  and  forming  with  them  salts, 
in  which  state  they  become  fitted  for  reception  into  the  cir- 
culation of  the  plant. 

§  89.  I  have  dwelt  somewhat  at  length  upon  the  importance 
of  the  oxides  of  iron  and  organic  matter  in  the  soil.  This 
subject  is  especially  interesting  to  planters  in  this  State,  1st, 
from  the  fact  that  so  large  a  proportion  of  the  best  soils  of  the 
eastern  counties  consist  of  vegetable  matter  in  the  main,  and 
2d,  from  another  fact  that  the  soil  in  the  midland  counties  is 
deficient  in  organic  matter,  it  having  been  consumed  by  long 
cultivation,  aided,  in  a  considerable  degree,  by  climate.  In 
1847,  I  prepared  an  article  for  the  American  Journal  of  Sci- 
ence and  Agriculture,  the  object  of  which  was  to  set  forth  in 


118  NORTH-CAROLINA   GEOLOGICAL   STJKVEY. 

as  a  clear  a  light  as  possible,  the  functions  of  the  vegetable 
matter  in  the  soil,  and  having  seen  no  reason  for  changing 
the  views  I  then  entertained,  and  still  believing  them  to  con- 
tain important  principles,  I  shall  transcribe  them  as  they  were 
then  printed.  It  should  be  remarked,  however,  that  the  more 
scientific  details  of  the  paper  belong  to  the  celebrated  Mulder, 
who  has  taken  a  widely  different  view  of  the  importance  of 
organic  matter  in  soil  from  Liebig.  I  made  just  an  allusion 
to  the  doctrines  inculcated  in  a  previous  communication,  which 
is  contained  in  the  following  extract : 

"Supplying,  then,  the  soil  with  decomposing  organic  matter,  and  several 
important  results  follow ;  the  rocks  are  dissolved  and  the  plants  may  be 
supplied  with  the  necessary  carbon,  ammonia,  and  other  essential  inorganic 
matter."  The  doctrine  contained  in  this  extract  is  important,  and  may  be 
drawn  out  more  in  detail.  The  opinion  has  generally  prevailed  that  mould 
or  the  black  matter  of  soil,  was  eminently  useful.  Many,  and  perhaps  all, 
at  one  time  entertained  the  idea  that  it  was  the  principal  food  of  plants. 
The  idea,  it  is  true,  was  crude,  and  it  will  not  offend  any  one  at  the  present 
time  to  say  that  the  early  notions  of  farmers  and  chemists,  who  had  turned 
their  attention  to  the  subject,  were  crude,  and  probably,  if  we  insist  upon 
it,  were  really  erroneous.  Still,  even  error,  in  toto,  is  rare,  and  some  truth 
at  least  is  usually  mixed  with  it ;  that  it  was  a  valuable  composition  in  the 
soil,  and  performed  some  function  serviceable  to  vegetation,  was  a  common 
belief.  The  error  consisted  in  the  misapprehension  of  the  truth,  and  was 
not  so  broad  or  fatal  as  that  which  maintains  that  it  is  of  no  use  at  all.  It 
is  by  no  means  a  fatal  error  to  maintain  that  a  substance  is  important,  and 
yet  mistake  its  function  or  office.  It  is  one  of  those  errors  which  belong 
to  theory,  and  does  not  necessarily  exist  in  practice.  A  farmer,  for  instance, 
believes  that  barn  yard  manure  is  useful.  His  belief  will  lead  him  to  save 
it,  and  employ  it  upon  his  corn,  and  this  he  may  do  notwithstanding  his 
theory  of  its  action  is  misapprehened,  or  may  be  totally  false.  The  main 
thing  is  to  be  right  as  to  the  fact.  Still,  a  correct  view  of  the  whole  sub- 
ject, how  the  organic  matter  acts,  in  what  way  it  is  beneficial,  and  how  it 
is  related  to  the  inorganic  matter,  will  undoubtedly  increase  our  power 
over  the  products  of  the  earth.  This  is  by  no  means  an  irrational  view  of 
the  subject.  If  we  apply  it  to  some  of  the  most  common  processes  of 
farming,  as  plowing,  it  is  evident  that  the  farmer  who  best  understands  the 
object  and  use  of  plowing,  will  derive  the  most  benefit  from  it.  All  agree 
that  it  is  useful,  and  hence  all  will  plow ;  still,  those  will  plow  the  best,  and 
adapt  the  work  better  to  the  end  in  view,  who  best  understands  its  use, 
than  the  farmer  who  has  only  this  naked  truth  at  his  elbow,  that  it  is  use- 
ful, but  knows  not  why  or  wherefore.  Theory,  then,  to  continue  the  line 


NORTH-CAROLINA  GEOLOGICAL   SURVEY.  119 

of  remark,  is  useful ;  and  correct  theory  eminently  useful.  At  the  same 
time,  the  fact  may,  and  usually  is,  more  important  practically ;  for  the  fact 
leads  to  the  right  action,  but  it  may  fall  short  of  the  benefit  it  is  calculated 
to  give,  when  fact  and  correct  theory  are  conjoined,  and  go  to  the  work 
together.  Theory  and  book  learning  are  often  ridiculed  by  the  matter  of 
fact  man,  and  yet  observation  often  bears  us  out  in  the  opinion  that  in  most 
instances  there  is  not  only  a  great  want  of  facts,  but  that  also  when  found 
they  are  often  greatly  perverted.  But  we  turn  now  to  the  subject  more 
immediately  before  us.  What  are  the  functions  which  the  organic  matter 
performs  in  vegetation?  Our  belief  is,  that  all  terrestrial  plants,  if  they 
do  not  absolutely  require  it,  are  at  least  benefited  by  it.  That  it  is  not 
taken  into  the  plant  in  the  condition  of  mould  or  humus,  is  proved  from 
the  fact  that  it  is  not  in  this  condition  sufficiently  soluble.  If  then  it  is 
useful,  it  is  necessary  to  maintain  that  it  undergoes  certain  changes  before 
it  becomes  the  food  of  plants.  It  may  minister  to  the  wants  of  vegetation 
in  several  ways,  without  its  becoming  the  food  itself.  It  ministers  to  the 
vegetable  by  its  presence,  procuring  thereby  an  open  state  of  the  soil,  by 
which  air  is  more  freely  conveyed  to  the  roots.  It  ministers,  also,  to  the 
wants  of  vegetation  by  its  absorbent  and  retentive  powers.  Indeed,  in  this 
respect  it  is  almost  indispensable  to  vegetation.  These,  then,  though  not 
all  the  uses  which  mould  exercises  in  vegetation,  still  are  sufficiently  im- 
portant to  merit  the  attention  of  the  agriculturist.  In  neither  do  we  find 
that  the  brown  or  black  matter  of  soil  becomes  the  nutriment  of  vegetables, 
and  yet  its  service  is  immense.  To  understand  the  nature  of  the  changes 
which  take  place  in  the  organic  matter  of  the  soil,  it  is  necessary  to  know 
what  agents  exist  there.  A  mixture  of  carbonate  of  lime  and  magnesia, 
silex  and  alumine,  and  organic  matter,  would  remain  without  change  for- 
ever, were  there  no  other  bodies  of  a  more  active  kind,  whose  affinities  be- 
come a  present  and  efficient  cause  for  action.  These  powers  or  forces  exist 
in  the  atmosphere  and  in  the  water  diffused  through  the  soil,  and  it  is 
proper  to  make  a  distinction  of  the  atmosphere  within  the  soil,  from  that 
above  or  without  it.  The  atmosphere  is  composed  of  two  elements,  oxygen 
and  nitrogen,  in  the  proportion  of  79  nitrogen  to  21  oxygen.  The  latter  is 
free  and  uncombined  with  the  nitrogen,  or  is  merely  dissolved  in  it,  just  as 
sugar  or  salt  is  dissolved  in  water.  The  consequences  which  follow  from 
this  condition  or  state  of  the  elements,  is,  that  both  are  free  to  unite  with 
other  bodies,  that  is,  so  far  as  attraction  for  each  other  is  concerned  there 
is  no  hindrance  or  force  to  be  overcome  to  bring  about  a  separation. 
Hence,  in  the  respiration  of  animals,  the  oxygen  of  the  atmosphere  which 
is  inhaled  combines  readily  with  the  carbon  suspended  in  the  return  or 
venous  blood.  So  in  the  soil,  there  is  the  same  independence;  the  oxygen 
or  nitrogen  is  not  hindered  from  uniting  with  other  bodies  by  any  affinity 
existing  between  themselves.  The  final  end  or  cause  of  this  is,  the  ulti- 
mate union  of  the  oxygen  with  certain  bodies  in  the  soil,  especially  with 
the  organic  part.  The  other  agent,  water,  undergoes  chemical  changes  of 


120  NORTH-CAROLINA  GEOLOGICAL   SURVEY. 

a  different  kind.  In  this  the  elements  are  chemically  combined,  and  hence 
they  are  not  so  readily  .separated  from  each  other,  and  hence,  too,  its  action 
is  constant,  and  that  which  is  proper  to  it  in  its  state  of  integrity — it  is  the 
solvent  power  so  necessary  to  bring  all  particles  to  a  state  of  fineness  that 
they  may  pass  into  the  organism  of  vegetables ;  for  solution  is  merely  that 
separation  of  particles  to  that  degree  of  minuteness  that  they  are  capable 
of  being  suspended  in  the  medium.  They  are  merely  farther  apart,  and 
they  are  brought  thereby  into  a  condition  to  undergo  farther  and  more 
thorough  changes  than  they  were  previous  to  their  solution  or  suspension 
in  the  medium  itself.  But  certain  bodies  can  and  do  decompose  it,  the  final 
end  or  cause  of  which  is  to  supply  ammonia  or  rather  nitrogen  to  the 
growing  plants.  Air  and  water,  then,  contain  the  elements  which  make  it 
possible  for  the  organic  matter  of  the  soil  to  return  once  more  to  that  vital 
state  in  which  it  exists  in  living  vegetables,  or  in  other  words,  to  become 
the  food  of  plants. 

If  we  now  trace  the  changes  which  decaying  wood  undergoes  from  the 
time  when  it  first  ceases  to  be  a  living  body  to  that  last  change  by  which 
it  is  fitted  for  the  function  of  nutrition,  we  shall  be  able  to  see  its  use  in 
this  part  of  the  economy  of  nature.  Wood,  when  it  has  lost  its  vitality, 
goes  to  decay,  but  the  progressive  changes  which  it  passes  through  are  not 
analagous  to  putrefaction.  Rotten  wood,  as  it  exists  in  decayed  trees,  is  a 
neutral  substance ;  neither  acid  or  alkaline  at  first.  But  in  progress  of 
time,  several  definite  substances  are  formed  from  it,  which  possess  activity 
and  belong  mainly  to  the  class  of  acids,  and  are  capable  of  combining  with 
the  alkalies  and  alkaline  earths  which  are  soluble  salts,  and  in  this  state 
minister  to  the  growth  of  plants.  Of  the  substances  which  are  formed  by 
decaying  wood,  and  by  peat  or  muck,  ulmine  is  one,  which  is  also  a  neutral 
body,  and  is  quite  insoluble,  and  hence  is  not  useful  as  a  nutriment.  This 
substance  is  called  ulmine  from  the  fact  that  it  was  first  prepared  from  the 
wood  of  the  elm ;  but  it  is  found  in  all  other  kinds  of  vegetable  matters 
which  are  undergoing  the  changes  already  alluded  to.  Ulmine  is  formed 
from  wood,  or  fibrous,  vegetable  matter  of  any  kind,  as  leaves,  twigs,  &c., 
by  the  absorption  of  oxygen  from  the  air,  or  contained  in  the  moist  earth. 
By  a  simultaneous  action  carbonic  acid  is  liberated.  The  substance  formed 
may  be  represented  by  C33,  H27,  0^;  33  equivalents  of  carbon,  27  of  hydro- 
gen, and  24  of  oxygen.  The  substance  represented  by  this  formula  is  a 
white,  friable  substance,  found  in  the  interior  of  hollow,  decaying  trees, 
and  is  produced  by  the  oxidation  of  the  woody  fibre.  Lignine  also  pro- 
duces other  bodies  by  combining  with  oxygen.  Thus,  4  atoms  of  lignine,* 
048,  H32,  032,  with  14  of  oxygen,  produce  80.  02  with  18H.  0. ;  and  an  atom 
of  ulmine,  C40,  H]4,  Oi2.  Other  products  of  an  analogous  kind  are  formed 
from  wood  by  union  with  oxygen.  Of  these,  humus  and  humic  acids  are 


*  Kane's  Chemistry,  edited  by  Draper,  p.  638. 


NORTH-CAROLINA   GEOLOGICAL   SURVEY.  121 

among  the  most  remarkable.  The  first  is  represented  by  the  formula  C^ 
Hi4,  Oi2;  the  latter  by  C40,  Hj5,  016.  These  two  acids,  which  are  spontane- 
ously formed,  and  are  common  in  peat  and  other  earths,  differ  from  each 
other  in  their  relations  to  ammonia ;  the  first  having  no  affinity  for  it,  wrhile 
in  the  latter  it  is  so  strong  that  it  is  difficult  to  separate  them.  In  conse- 
quence of  this  affinity,  it  no  doubt  forms  an  important  element  in  produc- 
tive soils. 

Another  class  of  vegetable  acids,  which  are  also  produced  by  the  action 
of  oxygen  on  organic  matter,  is  called  the  azotized,  from  the  fact  that  they 
contain  nitrogen.  These  acids  are  the  crenic  and  apocrenic  of  Berzelius. 
Both  are  soluble  in  water  and  alcohol;  the  apocrenic  less  so  than  the 
crenic.  They  form  with  alkalies  and  alkaline  earths,  soluble  and  insoluble 
salts ;  some  of  which  are  essential  constituents  of  a  rich  and  productive 
soil. 

By  the  continued  absorption  of  oxygen  from  the  atmosphere,  wood  and 
other  organic  matters  are  converted  into  a  nutriment  for  vegetables.  The 
crenic  and  apocrenic  acids  are  products  of  bodies  which  are  nitrogenous 
themselves;  the  nitrogen  of  which  is  retained  through  all  the  changes 
which  the  organic  matters  pass. 

It  seems  to  be  established,  then,  that  organic  matter  may  be  useful  to 
plants,  and  may  promote  their  growth  in  various  ways.  This  conclusion 
might  be  made  almost  a  priori,  subsequent  to  the  determination  of  the  na- 
ture of  the  bodies  under  consideration ;  for  it  is  well  known  that  many 
bodies  require  nitrogen;  and  it  is  ascertained  that  some  of  the  organic 
bodies  contain,  and  others  absorb  and  retain  ammonia  obstinately.  And 
each  of  these  classes  of  bodies  are  soluble,  and  in  a  condition  to  be  receiv- 
ed into  the  vegetable  system. 

If  the  foregoing  considerations  are  true,  why  should  farmers  be  taught 
that  the  organic  matter  of  decaying  leaves  and  of  their  barn  yards  is  use- 
less? that  it  is  a  bad  economy  to  spread  it  upon  their  fields,  or  plow  it  into 
their  soil  ?  We  have  sometimes  wondered  why  it  is  that  many  intelligent 
farmers  hold  book  farming  in  such  low  repute.  We,  however,  have  been 
satisfied  as  to  the  cause;  when,  for  instance,  doctrines  are  taught  so  con- 
trary to  their  experience;  and  when  they  are  told  that  they  had  better  burn 
their  barn  yard  manure  rather  than  carry  it  out  to  their  meadows,  we  are 
not  at  all  surprised  that  they  lose  confidence  in  books,  and  hence  often  re- 
fuse to  receive  many  things  which  are  really  sound  and  valuable ;  and  this, 
on  account  of  the  erroneous  doctrines  which  come  apparently  from  a  re- 
sponsible source. 

But  to  return  to  the  consideration  of  ammonia  in  the  soil.  Chemists 
are  not  agreed  as  to  the  processes  by  which  ammonia  is  supplied  to  the  soil. 
That  it  exists  there,  and  that  it  is  provided  for  by  certain  chemical  changes 
is  admitted.  We  have  stated  in  a  former  article  in  this  journal,  that  one 
of  the  means  by  which  it  is  restored  to  the  soil  is  through  the  mutual  in- 
fluence of  water  and  the  protoxide  of  iron ;  the  latter  substance  having  the 


122  NOETH-CAEOLINA  GEOLOGICAL  SURVEY. 

power  of  decomposing  the  former  and  taking  to  itself  its  oxygen ;  the  hy- 
drogen being  liberated  instantly  combines  with  the  nitrogen  of  the  air  in 
the  soil,  and  forms  with  it  ammonia.  Humic  acid,  too,  by  its  strong  affinity 
for  ammonia,  rapidly  absorbs  it  whenever  it  is  freed  from  its  combinations. 
Other  modes  undoubtedly  exist  by  which  the  nitrogenous  compounds  are 
supplied  with  this  essential  element.  Ammonia,  too,  has  been  proved  to 
be  present  at  all  times  in  the  atmosphere,  though  only  in  small  proportions. 

One  of  the  forms  in  which  ammonia  is  found  in  the  soil  is  that  of  apo- 
crenate  of  ammonia;  a  compound  which  is  formed  from  humic  acid  by  its 
continued  oxidation;  the  apocrenic  acid  being  merely  a  higher  state  of 
oxidation  of  the  same  substance.  In  the  chain  of  causes  by  which  apo- 
crenic acid  is  formed,  nitric  acid  is  also  generated,  according  to  Mulder — 
this  acid  acts  with  great  vehemence  upon  humic  acid.  Admitting  the  fact 
of  the  formation  of  nitric  acid,  and  its  subsequent  action  on  humic  acid 
follows  necessarily ;  and  furthermore,  we  can  understand  how  the  humic 
acid  is  oxidated  and  changed  into  apocrenic  acid.  Mulder  says,  p.  166,  in 
his  Chemistry  of  Vegetable  and  Animal  Physiology,  when  apocrenic  acid 
is  found  in  the  soil  it  is  accompanied  with  the  production  of  carbonic  acid; 
the  ammonia  of  the  soil  produced  in  it  from  the  atmospheric  air  it  has  ab- 
sorbed, may,  by  the  influence  of  decaying,  organic  substances  and  water, 
be  converted  into  nitric  acid ;  and  no  doubt  is  so  when  the  bases  required 
for  nitrification  are  present.  Saltpetre  was  long  extracted  from  the  soil 
exclusively,  as  in  many  places  in  Egypt,  India,  &c.  By  the  oxygen  of  the 
atmospheric  air  contained  in  the  soil,  the  hydrogen  and  nitrogen  of  ammo- 
nia produced  from  the  constituents  of  the  air  are  oxidized;  water  and  nitric 
acid  as  soon  as  it  is  formed,  meets  with  a  substance  in  the  soil,  humic  acid 
and  humin,  which  by  its  influence  is  converted  into  apocrenate  of  ammonia, 
and  at  the  same  time  produces  carbonic  acid.  This  change  of  humic  acid 
into  apocrenic  acid  takes  place  in  minute  quantities ;  as  is  the  case  with  the 
formation  of  ammonia  which  precedes  it.  Thus,  to  form  one  equivalent  of 
apocrenic  acid,  there  are  required  two  equivalents  of  humic  acid  and  one 
equivalent  of  ammonia  and  seventy-six  equivalents  of  oxygen.  In  this 
production  of  apocrenic  acid,  the  ammonia  from  the  humate  of  ammonia  is 
not  only  transferred  to  the  apocrenic  acid,  but  it  performs  an  intermediate 
part,  namely,  the  fixing  of  oxygen.  Through  the  tendency  of  ammonia 
to  form  nitric  acid,  the  oxygen  of  the  atmospheric  air  contained  in  the  soil 
is  combined  with  the  constituents  of  the  humic  acid ;  the  ammonia  itself 
remaining  unchanged;  neither  leaving  the  soil,  nor  being  oxidized  into 
nitric  acid.  If  there  be  not  an  abundance  of  organic  matter,  and  if  the  air 
be  moist,  and  lime,  magnesia  or  potash  be  present,  ammonia  is  first  pro- 
duced, and  afterwards  nitric  acid.  If,  on  the  contrary,  instead  of  these 
leaves,  organic  substances  are  in  excess,  humic  acid  is  formed  by  their  de- 
cay; at  the  same  time,  ammonia  is  produced  from  the  nitrogen  of  the  at- 
mosphere; and,  finally,  apocrenate  of  ammonia,  carbonic  acid  and  water." 

This  long  extract  seemed  to  be  required  in  order  to  put  the  reader  in 


NORTH-CAROLINA   GEOLOGICAL   SURVEY.  123 

possession  of  the  views  of  Mulder  on  this  important  subject;  from  which 
it  is  well  established  that  organic  matter  in  soil  is  of  the  highest  moment ; 
and  that  it  not  only  ministers  indirectly  to  the  growth  of  plants,  as  stated 
in  the  early  part  of  this  article,  but  also  becomes  food  itself  in  the  form  of 
apocrenate  of  ammonia.  So,  also,  that  important  substance,  carbonic  acid, 
is  liberated  and  furnished  to  the  roots ;  a  substance  which  many  suppose 
is  taken  up  by  the  leaves  only.  The  apocrenates  are  continually  forming; 
not  only  the  apocrenate  of  ammonia  but  also  those  of  potash,  lime  and 
magnesia. 

Through,  then,  the  action  of  the  organic  acids  the  inorganic  bodies  are 
received  also  into  the  circulation  of  vegetables ;  and  this  gives  us  an  idea 
of  its  importance,  namely,  as  a  medium  by  which  lime,  magnesia  and  pot- 
ash are  supplied  to  the  vegetable  kingdom.  The  carbonates  of  lime  and 
magnesia  are  rather  insoluble  bodies,  though  the  carbonate  of  soda  and 
potash  are,  as  is  well  known,  highly  soluble. 

We  should  take  an  unsafe  course  in  practice,  then,  in  rejecting  the  or- 
ganic part  of  manures ;  and  how  truly  important  lime,  potash,  soda,  mag- 
nesia, &c.,  are;  still,  soils  cannot  be  and  are  not  fertile  if  they  contain 
only  these ;  and  thp  highest  and  most  valuable  soils  are  those  in  which  a 
due  balance  is  preserved  between  the  organic  and  the  inorganic  parts. 

§  90.  Unfortunately  for  the  best  interests  of  agriculture, 
the  marls  of  North  Carolina  are  too  sandy  to  bear  transport- 
ation to  distant  points;  and  hence,  their  use  is  now  limited 
to  the  plantations  upon  which  they  are  found.  If,  however, 
a  method  could  be  devised  by  which  the  sand  could  be  sepa- 
rated cheaply  from  their  useful  parts,  they  would  then  be  re- 
duced in  weight  and  bulk  sufficiently  to  bear  transportation 
on  those  railroads  which  pass  within  three  or  four  miles  of 
the  beds  in  which  they  lie,  and  those  especially  upon  the 
Cape  Fear  and  the  Neuse  might  be  transported  very  cheaply 
by  water.  The  quantity  of  sand,  it  will  be  perceived,  is  often 
as&igh  as  80  per  cent.  The  remainder  twenty  per  cent  con- 
tains all  the  fertilizing  matter.  This  20  per  cent  is  a  concen- 
trated manure,  and  compares  very  favorably  with  the  super- 
phosphate of  lime,  especially,  considering  that  its  cost  would 
be  very  much  less,  or  according  to  its  actual  cost,  it  would  be 
worth  quite  as  much  as  the  superphosphate. 

By  aid  of  suitable  machinery,  it  is  highly  probable  the  sand 
may  be  separated  rapidly  from  the  valuable  parts  which  com- 
pose it.  If  so,  the  interests  of  agriculture  would  be  greatly 


124:  NORTH-CAUOLINA   GEOLOGICAL   SURVEY. 

promoted,  and  the  revenue  upon  the  railroads  increased ;  and 
in  the  end,  it  might,  and  invariably  would  supplant  guano, 
which  is  a  drain  upon  the  pockets  of  planters. 

§  91.  In  order  to  free  the  sand  from  adherent  marl,  it  might 
be  passed  through  a  cylinder,  the  inside  of  which  had  many 
projecting  angles,  and  within  which  another  cylinder  studded 
with  angular  rods  should  be  made  to  revolve  rapidly,  while 
the  marl  and  water  was  passing  through  them.  The  sand, 
after  issuing  from  the  machine,  would  subside  almost  imme- 
diately, while  the  lighter  marl  would  pass  forward  and  be 
allowed  to  subside  in  vats.  With  a  machine  properly  con- 
structed, a  hundred  tons  of  marl  might  be  washed  in  a  day, 
and  though  all  the  sand  might  not  be  removed  from  it,  yet  a 
very  large  proportion  would  be.  Some  of  the  marls,  as 
analysis  proves,  contain  seventy-five  per  cent  of  sand.  The 
concentration  consequent  upon  its  removal  would  convert  it 
into  a  fertilizer  which  would  contain  three  or  four  times  its 
amount  if  it  was  in  its  natural  state.  The  washed  marl  would 
then  possess  the  following  composition : 

Phosphate  of  lime,  2.50 

Peroxide  of  iron  and  alumina,  25.00 

Carbonate  of  lime, 44.17 

Magnesia,  1.71 

Potash,  2.35 

Soda,  2.50 

Sulphuric  acid 0.72 

Chlorine,    0.52 

Organic  matter,    1 6.12 

Soluble  silica,  0.78 

Water,    3.75 

The  commercial  value  of  marl  of  this  description  will  be 
from  8  to  9  cents  per  bushel.  A  bushel  of  dry  rnarl  weigh- 
ing eighty  pounds,  and  twenty-five  bushels  weighing  two 
thousand  pounds,  it  will  be  worth  from  $1  60  to  $1  80  per 
ton.  Fifty  tons  of  marl  might  be  washed  per  day,  which 
would  give  about  twelve  tons  of  concentrated  marl  in  the 
vats.  The  cost  of  raising  and  washing  may  be  performed  at 
from  37-J  to  50  cents  per  ton,  and  perhaps  less  than  the  low- 
est figure. 


NORTH-CAROLINA   GEOLOGICAL   SURVEY.  125 

§  92.  The  washing  of  the  marls  should  not  be  confined  to 
the  green  sand  marls,  the  white  eocene  marls  upon  the  Neuse 
in  Craven  county,  may  also  be  profitably  subjected  to  the 
operation.  It  would  at  any  rate  improve  it  much,  for  agri- 
culture, and  serve  to  create  a  demand  for  it  in  the  midland 
counties.  Besides,  when  it  has  been  subjected  to  this  opera- 
tion, it  becomes  an  excellent  material  for  burning  into  quick 
lime.  Being  in  a  tine  incoherent  state  after  washing,  and 
also  wet  or  a  calcareous  mud,  it  might  be  pressed  at  once  by 
means  of  moulds  into  the  form  of  large  bricks,  and  when  al- 
lowed to  dry,  put  up  in  kilns  for  burning.  In  western  New 
York,  the  white  fresh  water  marl  is  treated  in  this  way,  with 
the  exception  that  it  does  not  require  washing.  But  it  is 
moulded  into  the  form  of  bricks  and  burned.  It  is  highly 
esteemed  for  its  whiteness,  and  is  used  mostly  for  white-wash- 
ing. 

The  foregoing  hints  are  thrown  out  without  having  had 
time  and  opportunity  for  testing  their  value.  They  are  sug- 
gested  in  consequence  of  the  scarcity  of  limestone  in  the  mid- 
dle counties  of  the  State,  and  the  consequent  high  price  of 
lime.  There  is  lime  enough  in  the  eastern  counties,  but  its 
intermixture  with  sand,  which  diminishes  its  value  in  a  com- 
mercial point  of  view,  except  in  the  case  of  a  few  banks, 
which  have  been  designated. 

§  93.  To  show  that  green  sand  and  other  marls  may  be 
transported  over  railroads,  I  propose  to  quote  what  has  trans- 
pired already  in  New  Jersey,*  thus,  there  was  transported 
over  the  Freehold  and  Jamesburg  Agricultural  Eailroad  du- 
ring 1856,  270,982  bushels  of  marl,  all  of  which  found  a  mar- 
ket out  of  the  marl  district,  and  some  of  it  out  of  the  State ; 
and  as  an  evidence  of  the  estimation  of  the  marl  and  the  ready 
sale  it  finds  along  the  road,  it  requires  only  to  witness  the 
high  cultivation  of  the  lands  along  the  whole  route  of  the 
road.  Monmouth  county,  and  other  parts  of  New  Jersey  ^ 
were  as  barren,  or  as  much  exhausted  by  cultivation,  as  any 


*  Third  Annual  Report  of  the  Geol.  Survey  of  the  State  of  New  Jersey,  for  the 
year  1856,  p.  53. 


126  NOKTH-CAKOLINA   GEOLOGICAL   SURVEY. 

parts  of  this  State.  The  use  of  marl  has  renovated  the  coun- 
try, a  profitable  trade  has  sprung  up  which  will  not  only  ben- 
efit the  owners  of  rnarl  pits,  but  that  part  of  the  agricultural 
community  who  avail  themselves  of  this  substance,  when  it 
can  be  brought  from  a  distance  to  their  doors. 

§  94.  The  mode  of  calculating  the  money  value  of  a  marl, 
is  founded  upon  the  fact,  that  the  percentages  represent  the 
absolute  weights  in  the  compound, — thus  one  per  cent,  of 
phosphate  of  lime  is  equivalent  to  one  pound  in  a  hundred. 
This  number,  one,  or  one  pound  multiplied  by  20,  and  then 
estimated  by  the  value  per  pound  of  the  substance,  gives  its 
value  in  100  Ibs.  of  marl ;  or,  if  there  is  2,16  phosphoric  acid, 
the  product  is  4,32,  which  multiplied  by  5  cents,  the  value  per 
pound  of  phosphoric  acid  gives  $2.16,0,  or  two  dollars  and 
sixteen  cents,  the  value  of  this  substance  in  a  hundred  pounds 
of  marl.  The  object  to  be  secured  in  washing  the  marl,  is  to 
raise  the  percentage  of  phosphoric  acid  sufficiently  to  make 
it  a  merchantable  substance,  and  thereby  benefit  the  agricul- 
tural community  far  and  wide. 


CHAPTER  Y. 

Animal  manures — Fish — Crabs — Cancerine  composition  of  fish  before  and 
after  drying — Compost  of  Crabs- — Preservation  of  the  offal  of  fish  at  the 
large  fishing  establishments. 

§  95.  The  best  interests  of  agriculture  require  a  ready 
and  cheap  supply  of  manure.  Its  prosperity  depends  upon 
it.  Without  fertilizers,  it  would  be  impracticable  to  sustain 
this  branch  of  business,  except  in  some  highly  favored  districts 
where  the  supply  has  been  prodigally  provided.  A  source 
from  whence  an  immense  supply  in  some  localities  may  be 
obtained  is  the  ocean.  The  myriads  of  fish,  for  example, 


NORTH-CAROLINA   GEOLOGICAL   SURVEY. 


127 


which  resort  to  the  shores  of  North  Carolina,  might  be  turned 
to  an  immense  profit.  The  use  of  fish,  employed  for  this 
purpose,  has  been  practiced  for  a  century  upon  and  near  the 
coast  where  they  can  be  readily  procured.  Both  Connecticut 
and  Massachusetts  have  experienced  the  benefit  of  their  em- 
ployment. Recently  in  New  Jersey  a  more  systematic  at- 
tempt has  been  made  to  furnish  agriculturists  with  a  supply 
of  this  kind  of  manure.  In  the  old  wray  of  employing  fish 
they  were  put  whole,  if  small,  into  a  hill  of  corn  or  spread 
over  the  field.  In  this  mode  they  become  highly  useful,  but 
were  very  offensive.  The  moss-bonkers  have  been  principal- 
ly used  in  New  Jersey,  and  are  regarded  as  a  powerful  ma- 
nure. Prof.  Cook  has  given  an  analysis  of  this  fish  for  the 
purpose  of  ascertaining  the  amount  of  fertilizing  matter  which 
it  contains  and  its  comparative  value  when  dried  as  a  ma- 
nure.* 

In  the  fresh  state,  it  consists  of 

Water,    77.17 

Oil,    8.90 

Dry  substance,  19.93 

The  dry  substance  is  composed  of 

Lime, ,  8.670 

Magnesia,    670 

Potash,  1.565 

Soda, 1.019 

Phosphoric  acid 7.784 

Chlorine,  678 

Silica,  1.333 

Organic  matter, 78.301 

100.000 
Ammonia, 9.282 

The  fish  were  taken  in  the  fall  at  the  season  when  they  are 
fat.  At  this  season  they  weigh  nearly  a  pound.  Substances 
which  abound  in  oil  always  make  powerful  fertilizers.  Tbe 


Third  Annual  Report  for  1856,  of  the  Geol.  Survey  of  New  Jersey,  p.  68. 


128  NOBTH-CAEOLINA    GEOLOGICAL    SURVEY. 

Cotton  seed  is  a  well  known  substance,  whose  reputation  as  a 
fertilizer  is  based  in  part  upon  its  oil.  But  fish  are  rich  in 
oil,  phosphoric  acid  and  ammonia,  and  hence  they  form  a  con- 
centrated manure.  If  the  analysis  is  compared  with  those 
which  have  been  given  in  the  foregoing  pages,  it  will  be  seen 
that  the  constituents  offish  are  admirably  adapted  to  the  pur- 
poses for  which  they  have  been  employed. 

§  96.  The  same  remark,  however,  applies  equally  well  to 
all  animal  matters — flesh,  bone,  the  hoofs,  horns  and  hair,  all 
are  active  fertilizers,  their  speedy  influence  being  dependent 
upon  the  state  and  condition  in  which  they  are  applied.  Bone 
ground  finely  is  much  more  active  than  when  it  is  coarse. 
To  obtain  speedy  action  it  must  be  soluble.  But  fish  manure 
occupies  an  intermediate  position — it  is  more  speedy  in  its 
action  than  bone  dust,  but  it  is  more  transient  in  its  effects, 
in  which  case,  it  has  a  close  resemblance  to  guano. 

§  97.  Crabs  and  fish  of  the  same  class  have  also  been  pre- 
pared for  a  like  purpose.  The  king  crab  resorts  at  seasons  of 
the  year  to  parts  of  our  coast  in  immense  numbers.  These 
on  being  taken  are  dried  and  ground  when  it  is  prepared  for 
use.  It  has  been  sold  under  the  name  of  Cancerine  from 
cancer,  a  crab.  When  compared  with  guano,  it  is  found  quite 
similar  in  composition.  As  guano  is  supposed  to  owe  its  value 
mainly  to  its  ammonia  and  phosphate  of  lime,  it  may  be  com- 
pared with  fish  or  cancerine  to  determine  their  relative 
values.* 

Thus  Peruvian  Guano  contains  of 


Ammonia,    15.00 

Phos.  acid,    14.75 

Cancerine  ammonia,  10.75 

Dry  fish         do          9.27 

Phosphoric  acid,    7.78 

Phosphoric  acid  in  cancerine,    4.05 

An  immense  amount  of  fertilizing  matter  is  lost  which  might 
be  saved  in  the  offals  of  fish.     If  they  were  dried  or  preserved 


*  Geol.  Survey  of  New  Jersey,  p.  61,  for  1856. 


NORTH-CAROLINA   GEOLOGICAL    SURVEY.  129 

in  a  mode  which  should  free  them  from  offensive  odor,  they 
would  be  equally  valuable  for  a  manure.  All  the  large  es- 
tablishments upon  the  extended  coast  of  this  State  and  upon 
its  bays  arid  rivers,  would  furnish  as  much  fertilizing  matter 
as  is  now  imported  into  the  State  in  guano — the  cost  of  which 
is  paid  to  foreign  merchants. 

At  the  present  time,  the  inducements  for  the  preservation 
of  the  offal  of  fish,  and  the  taking  of  those  fish  which  ar* 
not  used  as  food  are  very  great,  in  consequence  of  the  di- 
minished cost  of  transportation  by  railroad  and  the  increased 
demand  in  the  interior  for  fertilizers.  The  prepared  can- 
cerine  for  market,  and  which  is  mixed  with  charcoal  and 
plaster  for  the  purpose  of  removing  its  unpleasant,  odor,  is 
composed  of:* 

Ammonia,    25.57 

Organic  matter,  29.23 

Phosphate  of  lime,   5.90 

Sulphate  of  lime 10.32 

Silex, 1.20 

Water, 26.10 

98.32  Booth. 

The  king  crab  is  used  without  preparation  in  New  Jersey 
by  the  farmers  of  Cape  May,  though  many  are  in  the  habit 
of  composting  them  with  earth.  It  is  thus  prepared  as  a  ma- 
nure for  wheat,  and  it  is  stated  by  Prof.  Cook,  with  the  hap- 
piest effects ;  the  poorest  soils  on  being  dressed  with  from  two 
to  four  thousand  produce  from  twenty  to  twenty-five  bushels 
to  the  acre,  and  thirty  bushels  is  not  an  uncommon  crop.  As 
this  kind  of  manure  contains  but  little  inorganic  matter,  an 
improvement  of  it  may  be  effected  by  the  addition  of  ashes 
or  lime  to  the  compost  or  dirt  heap.  Such  an  addition  would 
fit  it  for  corn,  clover  or  grass. 

It  is  very  possible  the  king  crab,  and  fish  only  fit  for  ma- 
nures, are  not  to  be  obtained  in  sufficient  quantities  upon  the 
coast  of  N"orth  Carolina,  to  give  the  business  an  importance 


Second  Annual  Report  of  the  Geol.  Survey  of  the  State  of  New  Jersey  p 

10 


130  NOETII-CAKOLINA   GEOLOGICAL   SURVEY. 

+ 

in  a  commercial  point  of  view.  But  the  real  advantages  of 
their  employment  is  still  very  great,  for  the  profits  of  fishing 
theror  may  be  added  those  of  agriculture,  which  is  probably 
neglected  on  account  of  the  natural  sterility  of  the  lands  upon 
thysounds  and  rivers.  In  many  places  vegetable  matter  may 
'obtained  with  which  to  form  in  part  the  compost  heap,  a 
stance  which  is  well  adapted  to  preserve  the  ammonia 
fnd  other  vegetable  matters. 

5$  98.  Concluding  remarks  upon  fertilizers.  Husbandry  in 
none  of  its  branches  can  be  conducted  successfully  in  the  ab- 
sence of  fertilizers.  This  remark  is  applicable  only  to  those 
soils  which  have  been  under  cultivation  long  enough  to  ex- 
hibit indications  of  incipient  exhaustion.  There  can  be  no 
question  respecting  the  necessity  of  supplying  the  waste  of 
soils  consequent  upon  cultivation,  and  there  is  no  branch  of 
agriculture  which  does  not  demand  a  constant  supply  of  ma- 
nures; and  hence  the  great  importance  of  creating  enough 
from  the  immediate  premises  of  the  establishment.  While  it 
is  better  to  purchase  fertilizers  than  to  proceed  in  the  cultiva- 
tion of  the  great  staples  without  them,  yet  when  the  expen- 
diture has  to  be  made  in  cash,  it  is  better  to  make  composts, 
save  the  excrements  of  animals,  under  cover,  procure  leaves 
and  all  kind  of  offal,  which  being  placed  in  a  condition  where 
their  volatile  matters  may  be  absorbed,  than  to  expend  ready 
cash  for  those  which,  in  the  end,  are  no  better  than  those 
made  at  home.  To  obtain  the  basis  for  the  construction  of 
compost  heaps,  the  mud,  and  swamp  bottoms,  salt  marsh- 
mud,  wThen  it  has  had  time  for  discharging  its  saline 
matter,  the  dirt  under  buildings,  which  is  always  rich  in  nitro- 
genous matters,  and  many  other  sources  may  be  found  and 
used.  In  the  eastern  counties,  those  places  in  particular, 
which  lie  upon  the  sounds  and  rivers  where  fishing  establish- 
ments are  accessible,  must  furnish  an  important  source  of 
manures.  The  offals  of  fish  should  be  composted  with  dirt, 
leaves,  plaster,  or  fine  charcoal,  to  deprive  it  of  its  odor  and 
retain  the  ammonia.  But  one  of  the  most  valuable  resources 
will  be  found  in  the  decaying  wood  of  forests,  swamps  and 
bottoms,  which  should  be  burned  when  there  is  no  wind,  and 


NORTH-CAROLINA  GEOLOGICAL   SURVEY.  131 

the  ash  secured  under  cover  before  it  has  lost  a  part  of  its 
potash  by  rains.  In  this  latitude  it  is  doubly  necessary  that 
all  fertilizers  which  abound  in  volatile  substances  should  be 
secured  from  the  direct  heat  or  rays  of  the  sun,  for  observa- 
tion very  clearly  proves  that  a  great  loss  is  sustained  in  all 
animal  fertilizers,  where  they  lie  unprotected  upon  the  ground, 
and  especially  if  exposed  to  its  direct  rays.  To  increase  the 
quantity  of  fertilizing  matter  upon  a  plantation,  should  be 
regarded  as  a  business,  and  that  business  should  be  systema- 
tized. It  should  be  followed  up  with  the  same  regularity  and 
attention  as  that  which  is  bestowed  upon  th  raising  of  cotton 
or  corn.  A  rich  plantation  is  "agreeable  to  the  eye  ;  it  will 
not  wash  nor  become  chanelled  into  unseemly  gullies,  unless 
the  owner  ploughs  his  grounds  carelessly,  or  neglects  to  sup- 
ply the  immediate  wants  of  the  crop  under  cultivation.  Ex- 
posed soils  gully.  Hence  the  importance  of  providing  for  the 
growth  of  the  crop  to  save  the  soil  from  washing  by  furnish- 
ing it  a  sufficient  protection  in  the  crop  under  cultivation. 
There  are,  therefore,  t\vo  considerations,  either  of  which  is 
sufficient  to  induce  the  planter  to  provide  fertilizers,  viz:  a 
remunerating  crop  and  a  tillable  surface,  or  one  free  from 
gullies.  A  soil  as  soon  as  it  is  approaching  to  an  exhausted 
state,  will  begin  to  be  marred  and  cut  by  streams  which  cross 
it,  and  those  which  are  formed  by  rain.  The  better  part  is 
thereby  carried  away  and  lost.  The  tendency  is  to  reduce 
the  value  of  the  plantation  and  render  its  cultivation  more 
difficult  and  expensive. 

The  cure  for  all  these  incidental  as  well  as  direct  evils,  is 
to  provide  an  ample  supply  of  fertilizers. 


132 


NOKTH-CAROLINA    GEOLOGICAL    SURVEY. 


CHAPTER  XI. 

(•lay. — Characteristics  of  a  good  clay. — Composition  of  fine  clays. — Com- 
position of  a  clay  upon  Bogue  Sound. 

§  99.  Clay,  though  rarely,  if  ever,  a  constituent  part  *of  a 
vegetable,  is  still  an  important  substance  in  matters  pertain- 
ing to  agriculture.  It  is  one  of  the  most  important  substances 
in  construction.  It  is  also  employed  largely  in  the  manufac- 
ture of  articles  indispensable  in  the  economy  of  the  house- 
hold, and  is  the  principle  material  employed  in  the  draining 
tile. 

Clays  differ  widely  from  each  other;  some  are  fusible; 
others  are  very  refractory  in  the  fire,  or  scarcely  fusible  by 
the  highest  heat  of  a  furnace.  For  certain  purposes,  the  re- 
fractory clays  are  indispensable.  For  lining  stoves  and  fur- 
naces, this  property  should  exist  in  an  eminent  degree.  For 
household  utensils,  it  is  not  necessary  the  clay  should  be 
highly  refractory  in  the  fire.  As  different  properties  are  re- 
quired for  the  different  uses  to  which  clay  is  to  be  put,  it  is 
desirable  that  the  adaptedness  of  clay  for  a  special  purpose 
should  be  determined  by  methods  which  are  within  the  reach 
of  every  intelligent  individual ;  at  least  that  good  clay  may 
he  determined  by  some  simple  and  easy  experiment. 

In  the  first  place,  good  clay  is  homogeneous  ;  it  is  free  from 
lumps,  stones  and  other  foreign  matter.  In  the  second  place, 
it  should  have  an  unctuous  feel ;  this  property  implies  tena- 
city, and  an  ability  to  mould  readily  and  retain  forms  and 
shapes  which  is  given  to  it  by 'working. 

In  the  third  place  it  should  contain  sand.  Too  much  sand 
destroys  cohesion,  but  a  certain  proportion  of  sand  imparts  to 
clay  an  ability  to  dry  or  season.  Bricks,  tiles  and  all  utensils 
must  dry  through  before  they  can  be  burned,  else  they  will 
crack  when  exposed  to  the  heat  of  the  kiln.  Excess  of  sand 
renders  moulded  clay  weak  and  unfit  for  handling;  its  tena- 


NORTH-CAROLINA   GEOLOGICAL   SURVEY.  133 

city  will  be  so  far  diminished  that  it  cannot  be  carried  from 
place  to  place. 

Certain  clays  contain  so  little  sand  that  in  order  to  dry  or 
season  well,  it  must  be  added  ;  but  when  clay  is  to  be  worked 
by  a  machine,  less  sand  is  required  than  when  it  is  worked  by 
hand. 

Clay  that  cuts  smooth  is  probably  a  good  clay.  The  sur- 
face exposed  by  cutting  should  not  exhibit  ragged  lines,  or 
show  particles  of  coarse  sand  or  hard  spots. 

Good  clay  has  a  uniform  color,  and  is  not  spotted  with 
ochrey  matter.  A  clay  may  be  red,  blue,  brownish  or  pur- 
plish, and  yet  possess  excellent  properties. 

Clays  for  certain  purposes  should  not  effervesce  with  acids  ; 
this  phenomenon  denotes  the  presence  of  carbonate  of  lime, 
which  imparts  fusibility  to  the  compound.  This  tendency  to 
fuse  in  the  kiln  is  increased  when  iron  is  present.  All  such 
clays  will  require  very  great  care  in  burning,  and  when  burnt 
into  brick,  are  unfit  for  places  where  they  will  be  exposed  to 
great  heat.  Fire  clays  consist  of  alumina  and  a  fine  or  im- 
palpable sand.  For  withstanding  high  heat,  as  much  sand 
must  be  mixed  as  the  clay  can  bear  and  handled  without 
breaking.  Sand  increases  the  infusibility  of  the  mass. 

§  100.  A  bed  of  fine  clay  overlies  the  shell  marl.  At  cer- 
tain places  it  is  fine,  plastic,  cuts  evenly,  and  may  be  moulded 
readily  into  the  form  of  any  article  in  common  use.  On 
Bogue  sound,  it  is  purplish  and  extremely  fine,  and  is  an  ex- 
cellent potter's  clay. 

The  composition  of  the  infusible  clays  of  the  best  kinds 
have  been  determined  by  many  anatyses.  Thus,  the  celebra- 
ted Stourbridge  clay  consists,  according  to  the  late  Prof. 
Johnston,  of 

Alumina, 38.8 

Silex, 46.1 

Water, 15.1 

100.0 

The  Woodbridge  fire  clay  of  New  Jersey,  according  to 
Prof.  Cook,  is  composed  of 


134  NORTH-CAROLINA  GEOLOGICAL   SURVEY. 

Water, 14.640 

Alumina,    52.850 

Protox  iron  and  magnesia,    0.944 

Silica,   39.76 

Lime, 0.398 

Magnesia, O.G50 

It  is  one  of  the  best  fire  clays  in  this  country. 

The  fusible  clays  contain  lime,  iron,  potash  and  soda,  all  of 
which  vary  more  or  less  in  the  proportions  they  bear  to"  the 
alumine. 

The  bed  of  clay  which  has  been  refered  to,  as  forming  one 
of  the  strata  in  the  series  of  coast  deposites,  appears  to  exist 
in  an  uncommon  state  of  purity  upon  Bogue  sound.  It  is 
readily  moulded  and  forms  a  very  firm  mass  on  drying ;  its 
grain  and  texture  is  very  fine  and  is  free  from  irregular  lumps 
or  regular  concretions.  It  is,  therefore,  homogeneous,  and  is 
well  adapted  for  fire-brick,  tiles,  etc. — and  may  also  be  em- 
ployed for  door  knobs.  It  is  composed  of 

Water, 5.70 

Silex, 67.40 

Protoxide  of  iron,  3.70 

Alumina,    23.08 

Lime,    , 0.11 

Magnesia, 0.8 

Potash,  0.4 

Soda, : 0.5 

This  clay  contains  but  a  small  percentage  of  water  after 
being  exposed  to  the  atmosphere  for  several  months.  It  be- 
comes nearly  as  firm  as  a  rock.  This  bed  of  clay  extends 
over  a  wide  territory,  and  at  many  other  points  I  have  ob- 
served that  it  is  equally  fine  and  compact.  It  is  one  of  the 
most  persistent  beds  in  the  tertiary  series.  A  fine  variety  of 
it  occurs  near  Halifax. 

Clay  is  sometimes  employed  as  a  fertilizer;  those  only, 
however,  which  are  rich  in  lime  or  potash  can  be  regarded 
as  of  sufficient  importance  to  warrant  the  expense  of  hauling. 
Clays  of  a  composition  similar  to  the  foregoing  are  not  adapted 
to  this  purpose. 

The  late  Prof.  Johnston,  in  summing  up  the  qualities  of  the 


NORTH-CAROLINA   GEOLOGICAL   SURVEY.  135 

best  tile  clays,  remarks  that  the  adhesiveness  of  clay  depends 
mainly  upon  the  proportion  of  alumina.  Clays  of  an  average 
goodness  will  contain  about  85  per  cent  of  silex  and  alumina 
when  taken  together.  Much  depends  evidently  upon  the 
coarseness  of  the  sand,  for  when  the  sand  is  coarse  the  ten- 
acity of  the  clay  is  very  much  diminished.  Clays  again  in 
which  the  infusible  ingredients  is  greatest,  other  properties 
being  equal  and  favorable,  are  best  adapted  to  the  manufac- 
ture of  good  tile,  besides  in  this  case  they  admit  of  being 
moulded  lighter  and  thinner.  If  lime  and  oxide  of  iron  ex- 
ist in  large  proportions,  the  clay  is  rendered  more  fusible,  but 
in  that  case,  it  possesses  an  advantage  of  being  burnt  with 
less  fuel.  So  with  brick.  The  clay  of  the  tertiary  beds,  it 
will  be  perceived,  contains  but  a  small  proportion  of  lime  and 
iron,  or  other  elements  which  are  calculated  to  confer  fusibili- 
ty. Hence  it  will  probably  be  found  that  this  clay  will  rank 
with  the  most  infusible  of  the  clays,  except  the  porcelain 
clays,  and  being  extremely  fine  and  tenacious  is  well  adapted 
to  the  manufacture  of  many  fine  earthern  wares  which  are  so 
necessary  in  house  keeping. 


CHAPTER  XII. 

The  grasses  and  their  functions — Different  objects  attained  by  their  culti- 
vation— Chemical  constitution  of  the  grasses — Elementary  organs,  and 
parts  of  the  blossom. 

§  101.  The  grasses  serve  many  important  purposes.  They 
clothe  the  earth  in  green,  a  color  easy  and  agreeable  to  the 
eye.  They  protect  the  loose  earth  and  prevent  its  washing 
away  and  transportation  into,  the  streams,  or  being  cut  into 
gullies.  They  furnish  food  to  the  beasts  and  birds,  and  the 
most  important,  the  cereals,  sustain  the.  millions  of  the  human 


136  NOKTH-CAKOLINA   GEOLOGICAL   SURVEY.      % 

race  which  now  people  the  earth.  The  seed  of  all  grasses  are 
nutritious ;  the  smallest  are  only  fit  for  the  sustenance  of  birds 
and  insects.  Those  which  are  denominated  corn,  are  those 
which  are  specially  cultivated  for  their  albuminous  matters  for 
the  use  of  man.  The  latte/,  I  do  not  propose  to  speak  of  un- 
der this  head ;  the  former,  or  the  grasses,  which  cover  the 
earth  with  green,  and  whose  herbage  forms  the  nutriment  of 
cattle,  compose  the  family  upon  which  I  propose  to  treat. 

The  diversity  in  kind  is  worthy  of  notice.  Each  one  has  its 
place.  The  meadow  has  its  special  occupants  which  usually 
belong  to  the  noble  kinds.  The  marsh  and  bog  are  covered 
with  those  which  are  coarse  and  unnutritious ;  and  the  dry 
hill-side,  with  the  tough  and  wiry  ones  which  serve  merely 
the  protection  of  the  surface.  The  hill-side,  however,  has  a 
better  class  of  occupants ;  and  where  the  surface  is  moist  the 
most  nutritious  grow  luxuriantly,  and  supply  the  herds  and 
flocks  with  the  most  nutritious  food. 

It  is  in  the  temperate  latitudes  that  the  best  grasses  find 
their  home,  and  the  husbandman  the  best  reward  in  their 
cultivation.  It  is  in  the  region  of  the  best  grasses  that  man 
obtains  the  richest  food  ;  milk,  butter,  cheese,  beef,  pork  and 
mutton  are  supplied  at  the  least  expense,  where  these  are  the 
material  productions  of  the  soil.  Life  is  sustained  at  the  least 
expense  where  the  better  grasses  grow  spontaneously.  Some 
of  them,  however,  must  be  sown  and  cultivated,  and  like  the 
cereals  be  raised  by  the  skill  of  the  farmer.  The  poorest 
grasses  frequently  crowd  out  the  better.  Lands  which  be- 
come poor,  support  only  the  poorer  kinds,  and  if  the  farmer 
seeks  his  best  interest,  he  will  displace  the  latter  by  good  til- 
lage and  the  use  of  fertilizers. 

The  direct  objects  which  are  sought  to  be  obtained  by  the 
cultivation  of  grasses,  are  the  production  of  beef,  milk  and 
butter ;  a  greater  variety  of  food,  better  in  kind,  and  more 
abundant  in  quantity. 

The  indirect  benefits  of  the  grasses,  in  addition  to  the  sup- 
ply of  food  for  cattle,  are  for  furnishing  a  source  of  fertilizers 
for  the  cereals,  and  preserving  the  soil  in  a  good  condition. 
If  cattle  are  left  to  roam  at  will  through  the  ranges  of  forest 


0       NORTH-OAROLINA    GEOLOGICAL    SURVEY.  137 

and  wild  pasture,  the  latter  object  is  sacrificed,  though  it  may 
appear  that  less  work  or  labor  is  consumed ;  still,  in  the  long 
run,  where  lands  are  sold  by  measure,  and  their  limits  restric- 
ted by  lines  and  corners,  the  losses  directly  and  indirectly 
sustained  more  than  counterbalance  the  gains  accruing  from 
the  use  of  indefinite,  uncertain  ranges. 

Another  consideration  bearing  upon  the  cultivation  of 
grasses,  may  be  regarded  somewhat  in  the  light  of  a  duty. 
Stock  require  a  variety  of  food.  The  benefits  of  variety  are 
numerous.  Health  is  one.  The  appetite  is  cloyed  by  con- 
finement. Unman  experience  is  a  sure  criterion  by  which 
to  determine  the  wants  of  the  beast.  Bacon  is  excellent  food. 
But  who  is  not  better  satisfied  with  his  diet,  if  a  beefsteak 
and  a  fowl  help  make  up  the  routine  of  meals  during  the 
week?  Watch  the  feeding  of  a  herd  of  cattle  or  a  flock  of 
sheep,  and  it  will  at  once  satisfy  the  close  observer,  that  they 
seek  variety,  and  doing  so  they  but  follow  the  promptings  of 
instinct.  Grasses  differ  in  value  ;  while  the  majority  of  them 
are  of  the  greatest  importance  to  animals,  some  rank  much 
higher  in  the  nutritive  scale  than  others.  The  most  nutritive 
grow  upon  the  best  soils,  the  least  either  upon  wet,  cold  soils, 
or  upon  worn  out  ones.  Let  an  intelligent  planter  see  the 
grass  of  a  field,  and  he  will  tell  you  whether  the  soil  is  rich 
or  poor,  cold  or  wet.  They  stand  as  indices  of  thrift  or  pov- 
erty, industry  or  laziness,  intelligence  or  ignorance. 

§  102.  In  the  cultivation  of  grasses  different  objects  are 
had  in  view.  Most  grasses  are  particularly  desired  for  their 
nutritive  properties,  but  some  fulfil  other  functions.  They 
may  be  demanded  for  their  ability  to  grow  in  sand,  when 
they  perform  the  important  office  of  confining  it  in  its  place. 
Some  make  a  good  turf,  and  their  strong  matted  roots  protect 
the  soil  and  clothe  the  suface  in  a  carpet  of  green. 

That  the  earth  may  be  covered,  and  the  marshes  and 
swamps  productive  in  something  useful  to  the  lower  forms, 
there  are  coarser  grasses  created  which  are  specially  fitted  for 
such  places.  The  Pheleum  pratense,  Poa  trivialis,  and  indeed 
most  of  the  rich  and  nutritive  ones  are  constitutionally  unfitted 
for  the  marsh.  A  rich,  sweet  grass  with  nutritive  seeds,  the 


138  KORTH-OAROLINA   GEOLOGICAL   SURVEY.       %  * 

Glyceria  fluctam,  flourishes  in  the  sluggish  waters  of  streams ; 
and  what  is  singular,  the  carnivorous  trout  feed  and  fatten 
upon  them.  The  broom  grass,  worthless  as  it  is  for  stock, 
clothes  the  worn  out  soil  and  protects  it  from  washing.  It  is 
better  it  should  be  covered  even  with  broom  grass,  than  burn 
in  the  sun  and  be  washed  away  by  the  showers.  Like  these, 
all  great  classes  or  divisions  of  natural  productions,  the  dif- 
ferent families  and  groups  have  special  duties  assigned  to 
them,  which  they  assiduously  fulfil,  whether  it  be  a  higher 
and  more  honorable  function,  that  of  supplying  nutritive  food 
for  cattle,  or  the  lower  and  humbler  ones,  to  protect  a  barren 
soil.  The  first  perform  a  double  office,  as  they  protect  equally 
well  the  soil  beneath  them;  the  latter  is  simply  protective  or 
passive.  As  grasses  have  their  preferences  for  certain  soils, 
as  the  wet,  or  dry,  or  one  moderately  wet,  so  they  also  re- 
quire a  particular  climate.  The  Timothy  grows  but  indiffer- 
ently in  North-Carolina.  It  requires  a  cooler  temperature^ 
or  a  less  scorching  sun.  Upon  the  mountains  constituting  the 
Blue  Ridge,  and  the  adjacent  ranges,  it  grows  as  well  as  in 
New  England,  where  it  is  the  most  important  of  the  grasses, 
and  a  source  of  wealth  to  the  inhabitants.  The  north  may 
have  a  few  species  which  are  restricted  by  climate  ;  the  south 
also  has  a  climate  which  is  suited  to  many  which  find  the 
north  incongenial  to  their  constitutions.  But  most  species  of 
grass  have  wide  ranges ;  they  are  less  restricted  when  they 
are  considered  only  as  to  ability  to  live,  but  do  not  grow 
freely ;  they  appear  under  restraint  and  fail  to  make  them- 
selves of  much  importance. 

A  moist  atmosphere  favors  development,  and  the  produc- 
tion of  a  juicy  tissue.  A  dry  and  cool  atmosphere  favors  a 
dense,  dry  and  wiry  tissue,  a  hard  outside,  and  a  tendency  to 
form  woody  fibre.  Animals  avoid  the  latter  and  seek  the 
former.  They  are  not  only  sweeter  and  more  palatable,  but 
require  less  effort  to  masticate,  and  less  wear  of  the  teeth,  in 
consequence  of  the  smaller  quantity  of  silex  in  the  dermal 
tissue. 

The  great  variety  in  the  constitution  of  grasses  secures  a 
succession  of  kinds  for  the  seasons.  The  early  spring  has  its 


.    NORTH-CAEOLINA   GEOLOGICAL   SUEYEY.  139 

kind,  and  a  succession  follows  till  late  in  autumn.  Some  are 
found  fitted  for  food  just  as  the  snows  are  about  to  cover  the 
ground.  The  farmer  will  not  fail  to  profit  by  this  succession. 
The  early  and  late  pasturage  shortens  a  winter  two  weeks  or 
more.  The  end  is  attained  by  mixing  the  seed  of  the  plants 
we  wish  to  cultivate.  The  advantage  is  not  confined  wholly 
to  a  successive  supply  of  food,  but  a  greater  quantity  grows 
upon  a  given  area  than  if  it  was  cultivated  with  one. 

§  103.  The  grasses  proper  consist  of  many  genera,  contain- 
ing each  many  kinds  or  species.  They  constitute  a  very  nat- 
ural family  of  plants  resembling  each  other  in  their  external 
characteristics,  and  also  in  their  internal  organization  and 
chemical  constitution. 

I  have  had  occasion  to  speak  of  the  chemical  constitution 
of  plants,  and  have  called  some,  as  the  clovers,  lime  plants, 
and  others,  potash  plants.  The  grasses  differ  from  these ;  in- 
stead of  lime  or  potash,  they  contain  silica,  though  potash  is 
sometimes  present  in  large  proportions,  and  must  necessarily 
be  present  to  a  certain  extent  in  combination  with  silica,  for  no 
doubt  it  is  required  to  give  it  solubility. 

The  design  and  construction  of  the  grass  plant,  as  it  was  to 
be  deficient  in  woody  fibre,  required  some  hard  substance  to 
sustain  its  slender  and  delicate  frame.  This  frame  work  is, 
in  a  portion  of  the  family,  a  hollow  cylinder,  or  several  hol- 
low cyinders  connected  by  impervious  solid  joints,  sometimes 
called  nodes.  Others  are  provided  with  a  pith  as  the  corn 
stalk.  Their  leaves  are  always  formed  upon  one  plan,  being 
long  and  tapering,  or  lanceolate  with  ribs  running  parallel 
with  each  other  their  entire  length  and  never  anastomosing. 
The  middle  one  is  stronger  than  the  rest,  and  more  prominent. 
The  leaf  terminates  in  a  sheath  below,  which  grasps  or  en- 
closes the  stem.  The  root  is  usually  fibrous,  sometimes  bul- 
bous, and  creeping ;  it  frequently  becomes  troublesome  to 
extirpate  as  it  emits  roots  from  the  numerous  joints  with 
which  it  is  provided.  The  flowers  or  blossoms  are  small  and 
never  showy.  They  are  simple,  having  envelopes  which  are 
in  keeping  with  the  family  characteristics.  Thus,  there  are 
provided  two  grassy  outside  leaves,  answering  to  the  calyx 


140  NORTH-CAROLINA   GEOLOGICAL   SURVEY. 

of  other  plants,  called  glumes,  and  two  more  delicate  inner 
ones,  answering  to  the  coral,  called  paleae.  In  the  centre 
stands  the  germ,  surmounted  by  two  feathery  sessile  anthers  ; 
and  beneath  and  around  the  germ,  there  issues  two  or  three 
filaments,  or  threads  bearing  anthers,  which  are  little  boxes 
containing  the  fertilizing  matter,  called  pollen.  The  indian 
corn  and  several  other  kinds  of  grasses  deviate  from  this  ar- 
rangement in  having  the  filaments,  bearing  the  pollen -boxes 
in  a  distant  part,  as  the  tassels;  while  the  pollen  receiving 
organs,  the  silks,  or  pistils  are  connected  with  the  germs 
lower  down  upon  the  stalk.  Wheat,  rye,  and  oats,  or  the 
hollow  stemed  grasses,  have  all  the  floral  organs  in  a  single 
blossom  together. 

The  floral  organs  are  borne  sometimes  upon  a  spike,  a  good 
example  of  which  is  furnished  in  the  Timothy  grass,  or  wheat 
head,  or  upon  a  panicle,  as  in  the  oat,  red  top,  bent  grass,  &c. 

The  grasses  contain  nutriment  in  their  stalks,  roots,  leaves 
and  seeds.  The  important  part  considered  as  food  for  beast, 
is  the  herbage,  the  stem  with  its  "leaves  and  head,  or  panicle 
of  flowers.  The  seed,  except  in  the  class,  cereals,  is  not  re- 
lied upon  as  an  article  of  diet.  The  nutiment,  so  called,  is 
divided  into  two  kinds:  1,  that  which  contributes  to  the  for- 
mation of  flesh  and  muscle.  2,  that  which  supplies  heat  to 
the  system,  and  which  is  capable  of  accumulating  in  different 
parts  of  the  body  in  the  form  of  fat.  It  is  designed  to  be 
burned  in  respiration  by  combining  with  oxygen,  while  the 
flesh  producing  matters  supply  and  renew  the  wasting  fibre. 

§  104.  The  value  of  grasses  for  feeding  stock  depends  upon 
the  quantity  of  flesh-forming  and  heat-generating  bodies 
which  they  contain.  The  first  are  known  under  the  names  of 
albuminous  substance ;  albumen,  the  white  of  an  egg,  represents 
the  first,  and  sugar  or  starch  the  second.  These  two  classes 
are  totally  unlike  each  other,  and  cannot  be  converted  one 
into  the  other  by  any  known  process.  All  substances  which 
are  used  for  food  contain  both  classes,  but  in  different  propor- 
tions. Flesh  of  animals  is  the  extreme  of  one  class  and  fat 
the  extreme  of  another.  In  the  potatoe  there  is  a  large  quan- 
tity of  heat-generating  matter,  and  a  small  quantity  onlv  of 


NORTH-CAROLINA    GEOLOGICAL    SURVEY.  141 

flesh-forming.  Milk  contains  these  two  classes  probably  in 
the  best  proportions  for  young  and  growing  animals.  The 
cheesey  matter  or  curd  is  the  flesh-forming  and  the  butter  or 
oil  the  heat-generating. 

In  all  cases  it  is  worthy  of  note,  that  water  is  a  very  large 
constituent  of  bodies  which  are  nutrient,  even  in  lean  meat 
the  highest  form  of  flesh-forming  matter,  about  four-fifths  is 
water. 

In  vegetables,  especially  the  seed,  these  two  classes  are 
concentrated  more  than  in  the  leaf  or  stem.  The  same  bodies 
exist  in  the  stem  and  leaves,  but  in  less  proportion.  The  con- 
stitution and  structure  of  domesticated  animals  undoubtedly 
require  that  the  flesh-forming  and  heat-generating  bodies 
should  be  so  combined  and  diluted  with  neutral  ones,  that  in 
order  to  satisfy  the  appetite  and  fulfil  the  designs  of  nature, 
they  should  take  in  a  bulky  aliment.  Hence  the  adaptation 
of  grasses  and  herbs  to  satisfy  the  requirements  of  their  sys- 
tems. The  nutritive  and  heat-generating  substances  do  not 
differ  in  kind  from  those  of  the  seed  or  even  from  flesh.  One 
of  the  questions  to  be  determined  then,  with  respect  to  grass- 
es, is  the  proportions  in  which  these  important  bodies  exist  in 
them.  This  question  is  easily  settled  by  an  analysis  of  the 
plant.  The  starch,  gum,  sugar  and  fat  represent  the  heat- 
sustaining  bodies,  the  albumen  the  flesh-forming.  A  grass 
will  be  valuable,  all  things  being  equal,  in  proportion  to  the 
latter  substance,  or  any  substance  which  performs  a  similar 
office.  Grasses  which  are  composed  mainly  of  silica,  as  the 
broom-sedge,  are  never  nutritious.  Those,  however,  which 
are  rich  in  potash  and  the  phosphates  of  the  alkalies,  are  nu- 
tritious, and  rank  high  as  flesh-forming  grasses.  As  grasses 
differ  among  themselves  in  these  particulars,  so  they  differ 
in  their  constituents  at  their  different  stages  of  growth.  The 
stem  particularly  loses  its  nutritive  properties  as  the  seed  be- 
gins to  form.  At  this  stage  its  woody  fibre  is  more  dense,  it 
is  less  palatable,  and  indeed  is  passed  over  entirely  by  stock, 
and  the  softer  vegetables  consumed  in  its  stead.  Hence  it  is 
necessary  in  forming  pasturages,  to  provide  a  variety  of  grasses 
which  ripen  their  seed  at  different  times,  and  thus  furnish  a 


14:2  NORTH-CAROLINA   GEOLOGICAL   SURVEY. 

juicy  food  during  the  time  of  pasturage.  For  hay.  a  similar 
rule  should  be  observed,  to  supply  hay  which  has  been  cut 
before  its  stalk  has  become  woody  and  unnutritious.  Hence, 
too,  a  meadow  which  is  designed  for  a  permanent  mowing 
lot  should  be  sown  with  grasses  which  reach  the  proper  stage 
nearly  at  the  same  time.  It  has  been  common  to  sow  Timo- 
thy and  red  clover  together.  They  are,  however,  incompat- 
ible with  each  other,  as  the  clover  comes  to  maturity"  before 
or  in  advance  of  the  Timothy.  Some  grass  then,  as  a  general 
rule,  should  take  the  place  of  Timothy,  where  it  is  wished  to 
sow  clover. 

§  105.  Grasses  grow  singly  or  in  clusters  and  tussocks ; 
both  frequently  increase  at  bottom,  or  spread  out  so  as  to 
form  a  turf,  a  matting  or  net  work  of  roots  woven  together  so 
as  to  form  a  coherent  mass,  somewhat  analogous  to  the  epi- 
dermis ;  it  is  a  protecting  surface,  spread  over  the  loose  soil  so 
as  to  confine  it  and  prevent  its  washing  away.  If  grasses  are 
mown  frequently,  they  are  more  tender  and  soft,  and  under 
a  moist  sky  assume  the  delicacy  of  a  green  velvety  lawn. 
The  grassy  surface  exerts  an  important  influence  over  tem- 
perature, maintaining  it  more  uniformly  than  if  it  were  earthy. 
It  prevents  wide  fluctuations  which  take  place  when  the  sur- 
face is  sand,  which  becomes  hot  and  burning  during  the  day, 
but  cold  and  uncomfortable  during  the  night.  The  stability 
of  the  earth's  surface  is  maintained  by  the  grasses. 

If,  then,  we  take  a  proper  view  of  the  offices  which  the 
grasses  perform  for  us  and  the  earth,  we  shall  set  a  high  value 
upon  them.  We  generally  think  of  them  simply  as  food  for 
cattle,  and  it  is  true  that  in  this  light  alone,  they  are  of  the 
utmost  importance.  But  this  is  not  all;  indeed  it  is  but  a 
small  item  in  consideration  of  the  good  they  do  and  the  ser- 
vices they  perform.  Though  humble  in  their  appearance  and 
pretensions,  they  serve  an  important  office  in  the  turf,  in  the 
temperature,  and  in  the  stability  and  permanence  of  the  earth's 
surface.  To  be  impressed  deeply  with  these  facts,  we  have 
only  to  witness  the  moving  sands  of  the  sea-shore  and  the 
sand-storm  of  the  desert. 

Important  as  I  have  represented  them,  it  is  probable  that 


NORTH-CAROLINA   GEOLOGICAL   SURVEY.  143 

other  forms  of  food  for  cattle  will  excel  them  in  profit  as  food. 
Roots  and  grain  outreach  them  on  this  score  for  special  pur- 
poses at  least,  though  cultivated  at  a  much' greater  expense 
than  the  grasses.  But  as  nature  demands  variety,  and  as  the 
system  must  have  food  large  in  bulk,  the  place  which  grasses 
occupy  cannot  be  filled  by  the  more  concentrated  nutrients. 
Disease  would  follow  if  cattle  were  fed  exclusively  upon 
grains. 

§  106.  The  valuable  grasses  belong  to  several  genera,  in 
each  of  which  there  are  several  species. 

Although  grasses  form  one-fifth  part  of  the  flora  of  a  coun- 
try, still  the  number  which  are  cultivated  or  domesticated  is 
comparatively  very  small, — cattle  consume  and  fatten  upon 
plants  which  are  not  grasses,  the  most  important  of  these  be- 
long to  the  leguminous  plants,  the  pea  family,  among  which 
are  ranked  the  clovers.  Of  these,  the  red  and  white  clover 
are  the  most  important.  The  red  clover  is  a  tender  plant 
when  young,  and  difficult  to  cultivate  in  a  hot  dry  climate,  as 
many  planters  have  experienced  in  the  eastern  part  of  the 
State. 

Grasses  or  Graminge,  are  subdivided  into  two  great  natural 
orders,  which  are  known  under  the  names  of  Cyperacem  and 
G-raminacece.  In  the  former,  the  flowers  are  monecious  or 
perfect,  consisting  of  imbricated  solitary  bracts.  They  com- 
prehend the  coarse  swamp  grasses,  but  few  of  which  are  es- 
teemed for  fodder  or  food  for  cattle.  They  are,  however, 
eaten  in  the  spring  when  young  and  tender.  The  latter,  have 
usually  perfect  flowers,  sometimes  monoecious  or  polygamous. 
The  external  envelopes  are  called  glumes  as  already  stated. 

The  southern  genera  comprehended  in  the  family  of  the 
true  grasses,  are  as  follows : 

Zizania,  Rottboellia, 

Leersia,  Cenchrus, 

Oryza,  Setaria, 

Mulenbergia,  Tripsacum, 

Agrostis,  Zea, 

Aristida,  Festuca, 

China,  Danthonia, 


144  NORTH-CAKOLINA    GEOLOGICAL    SURVEY. 

Calamagrostis,  Uralcpis, 
Stipa,                                       .     Bromus, 

Oryzqpsis,  Anthoxanthum, 

Spartina,  Aira, 

Manisurus,  Arena, 

Paspalum,  Phalaris, 

Cynodon,  Melica, 

.       Phleum,  Uniola, 

Alopecurus,  Briza, 

llordeuui,  Poa, 

Erianthus,  Arundinaria, 

Andropogon,  Eleusine, 

Oplismenus,  Dactylis, 

Panicum,  Elymus, 

Chloris,  Monocera. 

Many  of  the  genera  in  the  foregoing  list  belong  to  the  un- 
cultivated or  wild  kinds,  which,  though  they  are  eaten  by 
stock,  yet  are  supposed  to  be  unworthy  of  an  attempt  to  in- 
troduce them  into  our  system  of  husbandry. 

The  following  list  includes  the  cultivated  species : 

Botanical  names.  Common  names. 

Alopecurus  pratensis,  Meadow  foxtail, 

Phleum  pratense,  Timothy  or  herds  grass, 

Agrostis  vulgaris,  Red  top, 

"         alba,  English  bent, 

44        stolonifera,  Florin, 

"        dispar,  Southern  bent, 

Dactylis  glomerata.  Orchard  grass, 
Glyceria  nervate, 

Poa  pratensis,  June  grass, 

44    compressa.  Blue  grass, 

44    trivialis,  Rough  stalked  meadow 

41    serotina,  Fowl  meadow, 

Festuca  ovina,  Sheep  fescue, 

44         loliacea,  Slender  fescue, 

C-ynosurus  cristatus,  Crested  dog's  tail, 

Bromus  secalinus,  Willards  bromus, 

Lollium  perenne,  Perennial  rye  grass. 

44         italicmn,  Italian  rye  grass, 

44         multifloruin,  Many  flowered  darnel. 


NORTH-CAROLINA    GEOLOGICAL    SURVEY.  14ri« 

Avena  Sativa.  Oat. 

Avena  flavescens,  Yellow  oat  grass, 

Zea  mays,  Indian  corn, 

Phalaris  canariensis,  Commop  canary  grass, 

Anthoxanthum  odoratum,  Vernal  grass,  , 

Setaria  italica,  Bengal  grass, 

Oryza  sativa,  Rice, 

Sorghum  vulgare,  Indian  -millet, 

"        saccharatnm,  Chinese  sugar  cane.* 

Panicum  germanicum,  Hungarian  millet 

" '       sanguinalis.  Crab  grass. 

Cultivated  Leguminous  Plants, 

Trifolium  pratense,  Red  clover, 

"          repens,  White  clover, 

Medicago  Sativa,  Lucern, 

Hedysarum  onobrychis,  Sainfoin. 

Grasses  cultivated  for  confining  Mowing  tawi*. 

Ammophila  arundinacea,  Beach  grass, 

Elymus  arenarius,  Upright  sea  lyme  grass. 

§  107.  The  foregoing  list  of  cultivated  plants  have  been 
divided  into  the  following  natural  families  or  TRIBES: 

TRIBE    I. ORYZEAE,  TRIBE    III. PANICEAK. 

Oryza  sativa,  Panicum  germanicum, 

Leersia  oryzoides.  "        sanguinalis, 

TRIBE  IT. — PHALARIDEAE.  (Includes  38  species  of  paiiifiini. i 

Zea  mays,  Setaria  italica, 

Phalaris  arundinacea,  TRIBE  v. — AGROSTIUKAK. 

Phalaris  canariensis,  Agrostis  vulgaris, 

Anthoxanthum  odoratum,  "         alba, 

Alopecurus  pratensis,  "         stolonifera, 

"          geniculatus,  u        dispar. 

Phleum  pratense.  TRIBE  vn. — AVENACKAK. 

Avena  flavescens, 
"      sativa. 

TRIBE    VIII. FESTUCINEAE. 

Poa  pratense,  Bromus  secalinus, 

"     compressa,  Elymus  arenarius, 

"    trivialis,  (Triticum,  wheat, ) 

"     serotina,  Hordeum,  barley, 

Festuca  ovina,  Lollium  perenne, 
loliacea,  "       italicum, 

11 


146  NORTH-CAROLINA   GEOLOGICAL    SURVEY. 

Festuca  pratensis,  Lollium  multiflorun, 

Dactylis  glomerata,  Cynosurus  cristatus. 


GRAMINACEAE.— THE  GEASSES. 

TRIBE   I. — ORIZEA.         , 

Containing  those  grasses  whose  spikelets  are  one  flowered^ 
and  whose  flowers  are  often  monoecious  in  branched  panicle*. 

§  108.  Oriza  Sativa  is  cultivated  only  for  its  grain.  LEERSIA 
oryzoides,  rice  grass,  cut  grass,  false  rice.  The  rice  grass 
grows  with  a  procumbent  stem  and  an  erect  panicle,  having 
rough  slender  branches  and  long  narrow  leaves,  with  sheaths 
very  scabrous.  It  grows  from  two  to  three  feet  high  in  wet 
swampy  places.  Its  spikelets  are  flat,  and  the  florets  of  an 
oval  form  and  triandrous,  imbricate.  Where  other  grasses 
are  scarce,  this  may  be  cultivated  to  advantage,  as  it  makes 
a  good  hay,  and  may  be  cut  twice  or  three  times  in  a  season. 
It  flowers  from  October  to' November. 

TRIBE   II. PHALARIDEAE. 

The  spikelets  are  one  flowered,  and  perfect  /  if  more  than 
one  flowered,  polygamous  or  monoecious. 

£EA  mays. — INDIAN  CORN. 

Probably  no  plant  passes  into  or  forms  so  many  varieties 
as  Indian  corn,  or  furnishes  so  much  sustenance  for  man  and 
beast.  It  grows  within  the  limits  of  latitude  4*2°  south  and 
45°  north,  and  on  plains  and  mountains.  The  varieties  ripen 
at  different  times,  some  producing  in  forty  days  from,  plant' 
ing.  Others  require  six  months.  The  common  eight  rowed 
corn  cultivated  in  the  middle  and  northern  States,  comes  to 
maturity  in  about  ninety  days.  The  stalk  of  Indian  corn,  if 
deprived  of  its  tassel  and  silk,  furnishes  a  large  amount 
of  sugar,  but  it  does  not  possess  qualities  so  agreeable  as 
those  of  the  sugar  cane.  Its  ability  to  adapt  itself  to  climate 
is  of  immense  importance,  as  this  property  enables  it  to  be- 
come widely  distributed  over  the  earth's  surface. 


NORTH-CAROLINA   GEOLOGICAL    SURVEY.  147 

GENUS    PHALARIS. 

Its  glumes  are  two,  membranaceous,  equal,  keeled  and  one 
flowered ;  paleae  coriaceous,  shorter  than  the  glumes  and  jni- 
besent  at  base ;  flowers  in  compound  spikes, 

PHALARIS  ARUNDINACEA. REED  CANARY  GRASS. 

It  has  a  round  stem  which  is  smooth  and  erect,  with  five 
or  six  broad  leaves  of  a  lightish  green,  and  rough  on  both 
sides.  The  central  rib  is  prominent.  It  grows  on  wet  ground, 
and  attains  a  height  of  from  two  to  seven  feet.  The  ribbon 
grass  is  a  variety  of  this  species.  The  P.  arundinacea  is 
scarcely  worth  cultivating  for  its  fodder ;  its  yield,  however, 
during  the  season  is  quite  large,  but  cattle  are  not  fond  of  it, 
even  when  cut  early  and  well  cured.  They  eat  it  from  ne- 
cessity, when  nothing  better  is  furnished  them.  It  ranks  low 
in  the  nutritive  scale.  Phalaris  canariensis  is  cultivated  for 
its  seed  for  the  Canary  bird. 

ANTHOXANTHUM. 

Its  glumes  are  from  two  to  three  flowered ;  lateral  florets 
imperfect,  with  one  paleae,  bearded ;  intermediate  florets  per- 
fect, shorter  than  the  latteral  ones.  PALEAE  OBTUSE,  PANICLE 

CONTRACTED. 


148 


NORTH-CAROLINA    GEOLOGICAL    SURVEY. 


(FiG.  6.) 

best  on  moist  lands. 


A.  odoratum.  Sweet  scented  ver- 
nal grass. — (tig.  6)  Its  stem  is  erect, 
rough  at  the  summit,  leaves  hairy, 
sheaths  striate,  pubescent  at  the 
throat.  Glumes  are  acute,  hairy  and 
membranaceous.  Flowers  in  ap- 
pressed  panicles,  root  perennial, 
grows  from  twelve  to  fifteen  "inches 
high — flowers  in  May  and  June. 

This  grass  owes  all  the  importance 
which  it  possesses  to  its  fragrance. 
It  is  true,  that  it  is  an  early  grass ; 
and  hence,  may  be  eaten,  still  it  is 
not  much  relished.  It  appears,  how- 
ever, that  it  is  consumed,  with  the 
other  grasses  among  which  it  grows, 
and  imparts  to  the  milk  of  cows  a 
pleasant  taste,  which  is  more  partic- 
ularly given  to  the  butter. 

PHLEUM  PRATENSE — TIMOTHY,  OR  HERDS 
GRASS  OF  NEW  ENGLAND — CATS-TAIL 
GRASS  OF  NEW  ENGLAND. 

The  flowers  are  arranged  in  dense 
cylindrical  spikes.  It  has  two  equal 
mucronate  glumes,  which  are  longer 
than  the  paleae's,  they  include  two 
truncate,  boat  shaped  paleae,  without 
awns. 

This  species  has  aW  erect  smooth 
stern,  with  flat  linear-lanceolate 
leaves,  whose  sheaths  are  longer 
than  the  joints  ;  glumes  equal,  ciliate 
and  hairy  root  fibrous,  often  bulbous. 
Flowers  in  June  and  July,  and  grows 
It  grows  to  the  height  of  two  and  a  half 


feet.     It  was  introduced  into  Maryland  by  Timothy  Hanson, 
from  whom  it  derived  its  name.     This  grass  is  difficult  to 


NORTH-CAROLINA   GEOLOGICAL  SURVEY.  149 

cultivate  in  all  that  part  of  the  Southern  States  which  is 
known  as  the  low  country,  or  the  whole  of  the  Atlantic  slope. 
The  difficulty  in  its  cultivation  arises  from  the  dry  summers. 
In  the  months  of  August  and  September  it  dwindles  away 
and  finally  dies  out,  even  when  protected  by  many  large 
shading  trees  and  grown  upon  new  bottoms. 

In  mountainous  ranges,  however,  it  may  be  cultivated  suc- 
cessfully, and  as  it  is  one  of  the  best  of  grasses,  it  is  worthy  of 
the  attention  of  farmers.  It  should  also  succeed  in  the  higher 
grounds  of  the  middle  region. 

The  soil  required  for  timothy,  is  one  which  is  cool  and  moist, 
and  composed  of  a  vegetable  mould,  and  a  stiffish  base  of  clay. 
On  dry  upland  it  flourishes  well.  On  such  situations  it  often 
yields  two  tons  to  the  acre.  It  is  not  at  all  adapted  to  the 
sandy  soil  of  the  Atlantic  border.  The  seed  may  be  sown  at 
two  seasons:  in  the  fall,  immediately  after  the  sowing  of 
wheat,  or  in  March  when  the  ground  is  in  an  open  porous 
state  from  the  effects  of  a  frost. 

The  quantity  of  seed  required  for  an  acre,  is  from  a  peck- 
to  twelve  quarts.  Some  farmers  sow  only  from  four  to  six 
quarts.  It  yields  in  good  seasons,  from  ten  to  fifteen  bushels 
of  seed  to  the  acre,  and  has  produced  thirty,  weighing  46  Ibs. 
to  the  bushel,  and  it  is  worth  one  dollar  and  fifty  cents  per 
bushel.  Timothy  hay  is  preferred  oveif  all  others,  for  horses  ; 
it  is  also  a  superior  hay  for  working  cattle  in  the  spring. 

As  this  species  of  grass  gives  a  large  product,  it  will  be  in- 
ferred at  once  that  it  exhausts  the  soil — especially  where  it  is 
allowed  to  stand  and  ripen  its  seed. 

The  time  for  cutting  timothy  is  when  it  has  fully  blossom- 
ed. At  this  period  it  possesses  a  larger  percentage  of  nutri- 
ment than  when  its  seeds  are  ripening.  When  it  has  stood 
until  the  seeds  are  ripe,  the  stem  is  hard  and  coarse,  and  is 
not  relished  so  well  for  horses ;  besides,  it  is  less  nutritive, 
though  many  farmers  affirm  that  it  spends  better  and  goes 
farther.  Much  seed  may  be  saved  from  this  hay,  even  if  cut. 
early,  as  all  the  seed  does  not  ripen  at  the  same  time. 

The  old  practice  in  the  New-England  States,  and  which  is 


150  NORTH-CAROLINA   GEOLOGICAL   SURVEY. 

prevalent  still  to  a  great  extent,  is  to  sow  timothy  and  clover 
seed  together  in  stocking  down,  after  wheat  or  oats.  This 
practice,  however,  is  less  common,  as  it  is  evident  from  the 
period  at  which  the  two  plants  ripen,  that  one  is  too  imma- 
ture, and  if  allowed  to  stand,  the  other  has  passed  its  prime. 
Clover  is  too  early  for  timothy,  and  if  the  cutting  is  delayed 
till  the  timothy  is  ready,  the  clover  has  gone  to  seed,  and 
much  of  its  foliage  has  dried  too  much  to  be  of  any  value — 
its  stalk  alone  remaining  green  and  fresh. 

Wherever  this  grass  is  wished  to  succeed,  it  is  highly  ne- 
cessary that  it  should  not  be  fed  too  close  in  the  fall,  winter, 
or  spring  months.  Hogs,  if  allowed  to  run  in  meadows 
where  it  is  growing,  will  root-up  and  consume  its  bulbous, 
farinaceous  toot,  and  thereby  entirely  destroy  the  crop.  If 
cut  very  close  to  the  ground,  even  in  tke  northern  States,  it 
may  suffer  from  a  drought  which  frequently  occurs  about  this 
time  of  the  year ;  and  a  week  or  two  of  dry,  hot  weather  suc- 
ceeding immediately  its  removal  from  the  field,  is  very  liable 
to  injure  it.  Although  in  a  moist  climate  which  prevails  in 
mountainous  regions  generally,  it  is  very  easy  to  cultivate, 
yet  these  liabilities  to  fail  from  drouth  are  a  drawback  upon 
its  value — though  it  is  probably  the  best  stock-grass  which 
grows. 


NORTH-CAROLINA    GEOLOGICAL    SURVEY. 


151 


XLOPECURUS  PRATENSIS MEADOW    FOXTAIL    GRASS. — (Fig.  TV) 

Its  blossoms  are  arranged  in  dense  cyl- 
indrical spikes,  quite  similar  to  the  timo- 
thy, but  may  be  distinguished  from  it 
by  having  one  paleas.  Its  stem  is  erect, 
smooth,  and  from  two  to  three  feet  high. 
The  spike  is  shorter  than  the  spike  of  the 
phleum  pratense,  and  is  also  softer. 

This  grass  has  received  but  little  atten- 
tion in  this  country.     It  is  esteemed  in 
England,  where  it  is  a  native,  though  it 
is  indigenous  to  nearly  every  country  in 
Europe.     This  grass  is  specially  adapted 
to  pasturage,  as  it  vegetates  with  great 
luxuriance,  and  starts  up  vigorously  when 
eaten  off  by  sheep  or  cattle.     It  produ- 
ces seed  abundantly,  and  hence  stocks 
itself;  moreover,  it  bears  forcing  and  ir- 
rigation.    It  is  late  in  arriving  at  matu- 
rity— requiring  full  three  or  four  years 
to  come  to  perfection — and  hence  is  not 
well  adapted  to  an  alternate  husbandry. 
In  one  or  two  respects  it  is  more  valua- 
ble than  timothy,  as  it  yields  a  large  af- 
ter-math, whereas  the  timothy  yields  but 
a  small  one,  unless  it  is  growing  under 
the  most  favorable  circumstances.     Mea- 
dow foxtail  forms  a  good  sward  and  hence 
for  permanent  pasturage  it  is  eminently 
adapted. 

This  grass  too,  is  better  adapted  to  gen- 
eral cultivation  than  the  timothy  as  it  ear- 
ly grows  rapidly,  and  thrives  well  on  all 
soils,  except  on  very  dry  sands.     It,  how- 
ever, thrives  best  on  rich,  moist,  strong 
soils,  and  its  nutritive  matter  increases 
in  proportion  to  the  strength  of  soil  on  which  it  is  grown.     It 
grows  in  the  New  England,  the  Middle  States,  Ohio  and  Ma- 
ryland and  it  is  believed  that  it  will  grow  well  in  the  South- 


152 


NORTH-CAROLINA    GEOLOGICAL    SURVEY. 


ern  States,  because  it  grows  well  in  the  warm  climate  of  Italy. 
It  flowers  twice  in  the  season,  and  the  second  crop  exceeds 
the  first.  Sheep  are  fond  of  it,  and  when  it  is  mixed  with 
white  clover,  an  acre  it  is  said  will}rieldan  abundant  pasturage 
for  ten,  even  with  their  lambs.  An  acre,  therefore,  would 
grow  grass  for  one  cow.  London  observes,  that  it  affords 
more  bulk  of  hay  and  more  pasturage,  than  any  other  grass. 
This  remark,  however,  may  be  applicable  only  to  the  climate. 
Another  grass  belonging  to  this  genus,  grows  very  generally 
in  the  South ;  it  is  the  Floating  Foxtail,  Alopecurus  genicu- 
latus.  Its  stem  is  ascending,  but  bent  at  the  lower  joints, 
forming  knees,  smooth  and  glabrous ;  the  sheaths  are  shorter 
than  the  joints,  and  it  has  a  panicle  composed  of  cylindrical 
spikes  ;  the  glumes  are  pubescent,  but  the  paleas  are  glabrous, 
with  an  awn  at  base.  It  grows  from  12  to  18  inches  high, 
and  is  common  in  the  rice  fields.  It  may  flower  as  early  as 
March.  It  grows  in  water,  upon  which  the  upper  part  of  the 
ijtem  floats.  It  is  not  so  much  relished  by  stock  as  to  encour- 
age its  cultivation.  Its  early  growth  furnishes  green  and  fresh 
food  when  cattle  need  it  the  most,  but  still  it  is  not  sought  for 
with  avidity. 

TRIBE   III. PANICEAE. 

§  109.  Spikelets  two  flowered  ;  inferior  flowers  incomplete. 

Panicum  has  two  unequal  glumes,  the  lower  very  small : 
the  lower  florets  also,  are  usually  staminiferous.  Paleae  con- 
cave, equal,  beardless ;  seed  coated  with  the  paleae ;  flowers 
in  loose  scattered  panicles. 

PANICUM   GERMANICUM. HUNGARIAN   MILLET. 

The  testimony  which  has  come  to  hand  respecting  this 
species  of  millet  as  a  fodder,  is  favorable,  so  far  as  southern 
cultivation  is  concerned,  as  it  bears  a  drought  well,  and  re- 
vives speedily  on  the  occurrence  of  rain,  and  is  tolerably  pro- 
ductive on  dry  light  soils.  It  becomes,  however,  luxuriant, 
only  on  soils  which  are  well  manured. 

The  plant  is  leafy  and  remains  green  until  its  seed  are  ma- 
tured. In  France  its  cultivation  has  become  extended.  As 
a  green  fodder,  it  is  said  to  be  relished  by  stock  of  all  kinds. 

It  is  sown  broadcast  and  cultivated  like  other  kinds  of  mil- 


NORTH-CAROLINA    GEOLOGICAL    SURVEY. 


153 


let,  and  comes  to  maturity  in   about  the  same  time.     It  was 
introduced  into  this  country  through  the  Patent  Office. 

PANICUM    SANGUINALIS — COMMON  CRAB    GRASS. (Fig.  8.) 


(FiG.  3.) 


154  NOKTH-CAKOLINA    GEOLOGICAL    SURVEY. 

It  has  a  procumbent  ,assurgent,  geniculate  stem,  which  root? 
at  the  joints;  the  leaves  are  hairy,  with  spikes  shorter  than 
the  joints.  Spikes  digitate,  spreading,  from  4  to  6.  Annual, 
grows  through  the  summer;  common  in  cultivated  fields. 
This  grass,  though  by  no  means  so  valuable  as  orchard  grass 
or  redtop,  still  as  it  grows  luxuriantly,  and  is  moderately  nu- 
tritious, it  might  justly  be  cultivated  to  a  greater  extent  than 
it  is  at  present.  Cattle,  horses  and  mules  eat  it  with  consider- 
able relish  here,  and  it  is  frequently  saved  for  fodder.  But  as 
it  is  pulled  up  from  the  cornfields,  it  is  foul  with  sand  and 
dirt,  and  its  value  probably  diminished.  It,  however,  cannot 
take  the  place  of  the  better  grasses.  It  grows  from  one  to  two 
feet  high  in  waste  places,  in  gardens,  corn-fields  and  yards, 
and  is  frequently  a  troublesome  weed. 

The  panicum  (Oplismenus)  crusgalli  is  common  about  barns* 
and  waste  places  where  the  soil  is  rich,  and  some  attempts 
have  been  made  to  cultivate  it.  It  is  rich  and  nutritious,  and 
is  relished  tolerably  well  by  stock,  though  it  must  be  regarded 
as  coarse  fodder.  There  is  no  difficulty  in  cultivating  this 
grass  in  this  State,  as  it  grows  spontaneously  in  many  places, 
and  attains  a  height  of  4  feet.  It  is  better,  and  contains  more 
nutriment  than  the  crab  grass.  Its  ash  is  composed  of: 

Silica,   17.325 

Phosphate  of  iron, 0.425 

Phosphate  of  lime,   0.625 

Phosphate  of  magnesia, 2.831 

Phosphoric  acid,   6.894 

Silica  acid,    0.626 

Carbonate  of  lime,  3.060 

Magnesia,  2.613 

Potash,   36.656 

Sada,  1.885 

Chloride  of  sodium,  5.723 

Sulphuric  acid, 8.524 

Coal 1.850 

One  hundred  parts  of  the  plant,  nearly  dry,  gave : 

Water 4  737 

Dry  matter, 95.'J63 

Ash,    ...  .  11.479 


NOKTH -CAROLINA    GEOLOGICAL,    SURVEY.  155 

Amount  of  inorganic  elements  removed  in  a  ton  of  hay, 
235  pounds. 

TRIBE    IV. STIPACEAE. 

Spikelets  one  flowered  ;  inferior  paleae  awned  ovarium  stip- 
itate.  This  tribe  contains  only  wild  plants. 

TRIBE    V. AGROSTIDEAE. 

Spikelets  one  flowered. 

Agrostis;  glume  naked  and  beardless;  two  valved;  one 
flowered ;  valves  longer  than  the  paleae ;  paleae  two,  mem- 
branaceous ;  stigmas  longitudinally  hispid. 

AGROSTIS    VULGARIS. RED    TOP FINE    TOP. DEW-GRASS,    HERDS- 
GRASS,    OF   THE    SOUTHERN    STATES. 

Spikelets  one  flowered,  glume  naked,  beardless,  2  valved, 
valves  longer  than  the  paleae,  paleae  membranaceous. 

It  grows  erect,  slender,  with  round  smooth  stems,  wearing 
an  oblong  panicle ;  the  roots  are  creeping.  This  grass,  with 
many  others  of  the  genus  agrostis,  has  received  the  name  of 
'bent-grass^  by  the  English  ;  here  it  is  always  called  herds-grass. 
It  is  one  of  the  most  common  of  the  field  grasses,  and  is  not 
so  particular  in  its  selection  of  the  soil  in  which  to  grow,  as  it 
is  found  growing  spontaneously  in  wet  and  dry  meadows,  as 
well  as  upon  the  dry  hill  side.  It  is  regarded  as  possessed  at 
least  of  medium  qualities.  There  is  probably  no  well  cured 
hay  which  spends  better  than  red  top,  and  it  is  relished  by 
stock. 

The  soil  best  suited  to  red  top  is  one  which  is  moderately 
moist.  This  grass  is  comparatively  small,  and  hence  does  not 
yield  so  much  hay  to  the  acre,  but  it  forms  a  dense  bottom, 
and  if  fed  close,  it  makes  an  excellent  pasturage ;  if  allowed 
to  grow  up  to  stalk,  cattle  do  not  crop  the  stems,  or  do  not 
seem  to  relish  them.  Its  average  height  is  about  16  inches, 
but  on  rich  soils  it  is  twenty,  and  even  thirty  inches,  and  col- 
ored with  a  strong  tinge  of  purple.  On  poor  soils,  it  is  found 
as  low  or  short  as  six  or  eight  inches,  and  is  lighter  colored. 
Some  regard  this  dwarfed  variety  as  distinct  from  the  large  red 
top  of  rich  soils,  and  it  frequently  goes  under  the  name  of 
fine  top. 

It  flowers  here  in  June,  and  in  Massachusetts  in  July.     In 


156  NOBTH-CAROLINA   GEOLOGICAL   SURVEY. 

stocking  soils  after  oats,  or  corn,  the  red  top  forms  an  excel- 
lent addition  for  mixing  with  clover  and  timothy.  As  the 
timothy  diminishes  the  red  top  takes  its  place,  and  particular- 
ly does  it  fill  the  places  left  by  the  red  clover  as  it  gradually 
disappears. 

It  forms  a  close  or  dense  sward,  or  grows  thickly  at  bottom, 
and. hence  covers  and  protects  the  ground  when  the  timothy 
fails  to  grow  in  consequence  of  a  continued  drouth.  -  This 
grass  should  also  be  more  extensively  cultivated  in  this  State 
as  it  is  evident  on  examining  moist  meadows,  it  grows  very 
well,  spontaneously  and  without  much  attention 

AGROSTIS    ALBA WHITE   TOP. 

It  has  an  erect,  round,  smooth,  polished  stem,  which  is  sup- 
plied with  four  or  five  leaves,  whose  sheaths  are  roughish  and 
striate ;  joints  numerous,  from  which  roots  are  sent  off  when 
in  contact  with  the  ground.  It  is  distinguishable  from  red 
top  by  its  rough  sheaths  and  the  large  glume  toothed  only  at 
the  upj)er  part.  It  grows  in  wet  places. 

AGROSTIS     DISPAR SOUTHER^    BENT. 

The  stem  is  large,  erect  and  smooth,  surmounted  by  a  loose 
many  flowered  panicle,  somewhat  verticillate  and  pyramidal ; 
exterior  glume  largest.  It  is  a  native  of  the  United  States. 
It  has  been  commended  both  in  England  and  France,  but  is 
now  discarded.  The  hay  is  rather  coarse,  but  it  yields  a  heavy 
crop  on  good  sandy  bottoms  which  are  overflowed.  It  tillers 
out  and  becomes  strongly  rooted  in  the  soil,  and  hence,  is  a 
good  pasture  grass.  It  grows  well  in  the  low  country  of  the 
South,  where  it  appears  to  be  at  home. 

TRIBE    VI CHLORIDES. 

§  111.  Spikelets  in  unilateral  spikes  from  1  to  many  flow- 
ered, digitate  or  paniculate ;  rachis  not  articulated.  It  con- 
tains only  wild  grasses. 

TRIBE     VII AVENACE^E. 

Spikelets  two,  to  many  flowered,  panicled ;  the  lower  palese 
bearing  upon  its  back  a  bent  or  twisted  awn. 


NORTH-CAROLINA    GEOLOGICAL    SURVEY.  157 

A-VENA — (OAT.) 

Its  glumes  are  from  2-7  flowered,  longer  than  the  florets  ; 
palese  bifid,  toothed  with  a  twisted  awn  upon  the  back. 

The  common  oat  is  susceptible  of  cultivation  in  high  lati- 
tudes, where  it  is  the  most  profitable  grain.  In  warm  climates 
bears  a  lighter  grain.  The  stem  of  the  oat  is  quite  nutritious, 
and  forms,  with  meal,  an  excellent  feed  for  horses. 

The  oat  plant  when  sun-dried, 

Contains  water,   9.58 

Ash,    2.37 

Calculated  drj-,  2.61 

.  The  ash  of  the  straw,  consists  of 

Silica,    13.399 

Earthy  and  alkaline  phosphate,   8.902 

Carbonate  of  lime,  7.254 

Magnesia, 0.448 

Potash 60.035 

Soda,  3.622 

Sulphuric  acid,  5.754 

Chlorine,    0.581 

This  analysis  was  calculated  without  carbonic  acid  or  or- 
ganic matter.  These  amounted  to  in  carbonic  acid  6.140 ; 
organic  matter  2.400. 

In  a  ton  of  straw  there  will  be  removed  from  the  soil  in, 

Silica,   21.907  Ibs. 

Phosphates, 14.555 

Carbonate  of  lime,   11.868 

Magnesia, 0.732 

Potash,  98.157 

Soda,  5.921 

Sulphuric  acid, 9.408 

Chlorine, 0.950 

163.498  Ibs. 

The  amount  of  ash  in  an  unripe  straw  is  greater  than  after 
it  has  ripened,  which  is  undoubtedly  owing  to  the  transfer  of 
matter  from  it  to  the  grain.  The  ash  of  an  unripe  straw 
amounted  to  3.15,  which  calculated  from  a  perfectly  dry 
straw,  amounts  to  3.48. 


158 


NORTH-CAROLINA    GEOLOGICAL    SURVEY. 


The  oat  is  an  exhausting  crop  to  soil,  but  for  that  reason  it 
should  be  widely  cultivated  where  the  climate  suits  it.  It  is 
for  this  reason  that  it  is  so  valuable  for  food,  both  for  man 
and  beast. 

In  this  family  we  find  the 

AVENA  (DANTHONIA)  SPICATA. — WILD  OAT  GRASS. — (fig.  8.) 


It  has  an  erect  pubescent 
stem,  and  tubular  pubescent 
leaves,  with  sheaths  bearded  at 
the  throat.  Glume  usually  six 
flowered,  longer  than  the  spike 
margins  membranaceous.  Pa- 
leae  two;  exterior  one  lanceo- 
late villous,  the  sides  terminat- 
ing in  two  awns,  with  the  spi- 
ral one  upon  the  back.  Com- 
mon in  the  middle  country 
from  Carolina  to  Georgia. 

It  grows  in  dry  sunny  pas- 
tures, aud  attains  a  height  of 
twelve  to  eighteen  inches.  It 
is  of  but  little  value  for  pas- 
turage or  hay. 


NORTH-CAROLINA   GEOLOGICAL    SURVEY. 


.UK  A  FLBXUO6A WOOD  HAIR-GRASS—COMMON  HAIR-GRASS.— (fig.  9.) 


(Fio.  9.) 


160  NORTH-CAROLINA    GEOLOGICAL    SURVEY. 

It  has  an  erect,  terete,  glabrous  stem,  with  setaceous  leaves 
and  a  diffuse  panicle,  whose  branches  are  somewhat  verticil- 
late  ;  glumes  unequal ;  palese  equal,  exterior  one  pubescent 
at  base,  and  bearing  also  an  awn.  The  grain  is  oblong  and 
smooth.  It  flowers  in  August  and  September.  Figure  taken 
from  the  grass  when  in  fruit.  In  high  dry  pastures,  it  grow? 
remarkably  well,  and  is  eaten  freely  by  sheep.  It  is  poor  \r. 
nitrogen,  and  is  worth  nothing  for  cultivation. 

TRIBE   VIII. — FESTUCINEAE. 

Spikelets  two  to  many  flowered  ;  panicles  sometimes  race- 
mose, and  generally  without  awns. 

POA. — (MEADOW  GRASS.) 

The  poas  have  two  glumes,  and  usually  many  flowered, 
^pikelets  compressed  ;  paleae  sometimes  woolly  at  base;  scales 
Mn-v>ih  ;  panicle  more  or  less  branching  or  scattered. 

J\ 

-- 

TOA   COMPRESSA  —BLUE-GRASS WIRE-GRASS. 

Stem  decumbent  and  compressed,  ascending  and  surmount- 
ed with  a  dense  compressed  panicle,  somewhat  onesided,  and 
provided  with  short  bluish  green  linear  leaves.  Spikelets 
flat  ovate  oblong,  and  from  four  to  nine  flowered,  which  are 
rather  obtuse,  and  hairy  below  the  keel.  It  rarely  exceeds 
14:  inches  in  height.  It  has  a  creeping  root  and  a  geniculate 
stem,  and  much  compressed,  and  under  favorable  conditions 
grows  to  the  height  of  17  or  18  inches. 

The  blue  grass  varies  much  in  its  appearance.  On  dry  soils 
it  grows  m  tufts  with  rigid  culmlike  or  wiry  stems  ;  it  is  also 
short,  and  has  small  compressed  panicles,  and  the  whole  plant 
lias  a  bluish  green  color.  It  is  solid  and  heavy,  and  also  te- 
naceous  of  life  as  might  be  suspected  from  its  growth  upon 
very  dry  knowles,  and  in  wheat  fields  is  frequently  regarded 
as  a  pest.  It  is,  howrever,  a  very  nutritious  grass,  and  is  eaten 
freely  by  stock.  It  is  valuable  as  a  pasture  grass. 


NORTH-CAROLINA   GEOLOGICAL   SURVEY. 


161 


PRATENSIS SPEAR-GRASS GREEN   MEADOW-GRASS JUNE- 
GRASS KENTUCKY  BLUE-GRASS. (Fig.    11.) 

Stem  smooth,  erect, 
terete,  surmounted  by  a 
rather  spreading  crowd- 
ed panicle,  and  whose 
spikeletsare  ovate,  acute 
and  crowded  on  the 
branchlets,  from  two  to 
five  flowered.  Glumes 
unequal,  sharply  accu- 
minate,  lower  paleae  five 
nerved. 

This  grass  is  a  native 
of  Europe,  but  has  be- 
come extensively  natu- 
ralized in  the  United 
States,  both  north*  and 
south.  It  is  particularly 
at  home  in  some  of  the 
south-western  States,  as 
Kentucky  and  Tennes- 
see. It  extends  through 
the  Atlantic  States  as 
far  south  as  Charleston, 
where,  according  to  El- 
liott, it  grows  to  the 
height  of  18  inches, 
where  it  also  makes  a 
fine  winter  grass,  re- 
markable for  its  deep 
green  color,  and  soft  suc- 
culent leaves.  It  bears 
the  summer  heats  in 
(FIG.  11.)  close,  rich  soils-,  and 

wants  only  size  to  render  it  one  of  the  most  valuable  acquisi- 
tions to  the  farmer.     It  is  perennial,  and  hence  deserves  the 
special  attention  of  the  southern  planter,  as  there  is  a  great 
12 


162  NORTH -CAROLINA   GEOLOGICAL   SURVEY. 

want  of  good  perennial  pasture  grass.  ISTor  is  there  the  least 
doubt  but  that  it  can  be  generally  cultivated  in  the  eastern 
and  midland  counties  of  the  State.  As  for  the  western  coun- 
ties, no  farther  proof  is  required  than  what  is  already  known 
of  its  ability  to  thrive  there.  This  grass  continues  green  and 
fresh  in  Western  New  York,  frequently  as  late  as  December, 
it  is  probable,  therefore,  that  in  a  large  portion  of  Western 
Carolina,  it  will  continue  growing  most  of  the  winter.  Al- 
though it  continues  to  grow  during  a  long  period,  yet  it  sends 
up  its  spike  of  flowers  but  once  in  the  year,  which,  in  this 
climate  is  from  about  the  first  of  June  to  July.  It  continues 
afterward  to  spread  at  the  bottom  and  furnish  a  thick  mat  or 
growth  of  leaves.  It,  therefore,  makes  a  good  turf.  It  is  not 
so  particular  in  its  selection  of  soils  as  it  grows  on  dry  knowles 
as  well  as  moist  places.  But  still  it  flourished  best  in  a  good 
soil,  but  here  it  is  important  to  obtain  a  grass  which  will  en- 
dure a  drought  and  grow  on  poorish  soils. 

The  produce  is  ordinarily  small,  but  it  is  of  a  fine  quality. 
For  lawns  and  door  yards,  it  is  probably  better  adapted  than 
any  grass  in  cultivation.  One  of  the  difficulties  it  has  to  con- 
tend with  in  this  State  is  its  consumption  by  the  hog.  This 
would  not  be  so  formidable  to  surmount  if  it  attained  perfec- 
tion at  an  earlier  period,  requiring  two  or  three  years  to  get 
perfectly  set. 

As  it  requires  time  to  attain  perfection,  it  is  not  well  adap- 
ted to  an  alternate  system  of  husbandry,  or  when  land  is  to 
be  ploughed  every  two  or  three  years.  Shaded  pastures  fur- 
nish the  best  examples  of  this  grass  in  Kentucky  where  it 
ripens  its  seed  about  the  tenth  of  June.  In  August  it  takes 
another  vigorous  shoot  and  continues  to  grow  till  stopped  by 
the  cold  of  winter.  When  it  dries  up  in  the  drought  of  sum- 
mer, it  is  still  nutritious.  It  continues  to  furnish  under  the 
snow  pasturage  for  mules,  horses  and  sheep. 

If  designed  for  hay,  it  should  be  cut  late  in  flower,  and  if 
mixed  with  clover,  the  yield  will  be  at  least  midling  in  quan- 
tity. It  is  eaten  and  relished  by  all  kinds  of  stock.  It  seems, 
however,  to  flourish  best  on  what  are  called  limestone  soils, 
similar  to  those  uf  the  Kentucky  limestone  belt.  It  is  main- 


NORTH-CAROLINA   GEOLOGICAL   SURVEY.  163 

tained  by  several  writers  that  the  June  grass  is  deficient  in 
nutritive  properties,  that  it  is  far  inferior  to  timothy;  yet 
cattle  do  fatten  upon  it,  and  so  far  as  observation  goes,  the 
cattle  that  are  raised  and  prepared  for  market  in  Kentucky, 
are  equal  to  any  grass-fed  animals  seen  in  market.  Prof. 
Way,  whose1  analysis  of  this  grass,  have  led  to  the  unfa- 
vorable opinions  respecting  its  deficiency  in  flesh-forming 
elements,  may  have  analyzed  specimens,  which,  growing  in 
England,  may  not  have  been  as  nutritive  as  those  commonly 
growing  in  our  climate.  It  is  certain  that  the  composition  of 
plants  are  very  variable  under  different  circumstances,  soils, 
etc. ;  variable  also  at  the  different  periods  of  growth. 

In  Kentucky  farmers  sow  in  September  or  February.  Some 
prefer  a  late  winter  or  early  spring  sowing  to  save  the  tender 
plant  from  frost.  It  is  sown  both  in  open  ground  and  wood- 
land. If  sown  in  woodland  it  should  not  be  grazed  until  it 
matures  seed.  The  seed  is  often  mixed  with  timothy  and 
clover,  and  half  a  bushel  of  the  seed  of  June  grass  is  suffi- 
cient for  an  acre.  By  mixing,  the  field  may  be  fed  at  an 
earlier  day.  Ultimately,  the  June  grass  takes  full  possession 
of  the  field. 

POA   TRIVIALIS. ROUGH    MEADOW  GRASS. 

Stem  or  culm  somewhat  scabrous ;  leaves  smooth ;  narrow 
with  scabrous  sheaths  ;  panicle  equal  and  diffuse,  somewhat 
verticillate.  Spikelets  three  to  four  flowered;  glumes  une- 
qual ;  scabrous  at  the  apex;  lower  paleae  obtuse  ;  pubescent 
at  base ;  culm  from  two  to  three  feet  high. 

In  England  this  grass  is  highly  esteemed,  and  according  to 
the  opinion  of  Mr.  Curtis,  an  English  writer,  it  is  one  of  tht- 
most  valuable,  both  for  hay  and  pasturage.  In  this  country, 
however,  it  does  not  stand  so  high  in  the  estimation  of  agri- 
culturists, but  it  is  probable  that  it  has  not  been  so  fairly 
tested  as  the  blue  grass.  Mr.  Sinclair  recommends  it,  and 
says  of  it  that  it  is  superior  in  produce  to  many  other  grasses ; 
it  is  nutritive,  and  oxen,  horses  and  sheep  exhibit  a  marked 
partiality  for  it.  It  grows  vigorously  only  on  moist  situations ; 
when  upon  dry  pastures  it  is  only  inconsiderable  in  quantify. 


v  KOKTH-CAROLINA   GEOLOGICAL   SURVEY. 

He,  (Sinclair,)  remarks  that  it  should  be  mixed  with  other 
grasses,  when  it  will  nearly  double  itself,  which  is  in  conse- 
quence of  being  partially  sheltered.  Where  spots  in  pastures 
are  closely  eaten  down  it  will  be  found  the  places  were  occu- 
pied with  this  grass,  proving  thereby  the  fondness  of  stock 
for  it.  It  is  not  so  widely  diffused  as  the  June  grass,  but  it  is 
found  in  Kentucky,  from  which  it  may  be  distinguished  by 
its  rough  sheaths.  It  has  a  fibrous  root  and  is  an  annual.  It 
should  be  cut  when  in  seed.  It  has  more  nutriment  in  its 
aftermath  than  when  cut  in  seed.  In  a  specimen  which  I 
submitted  to  analysis,  I  found  : 

Water,  ....................................  ......  77.874 

Dry  matter,  ....................................  .  22.626 

Ash,  ........  ................................  ______     2,073 

This  was  cut  the  8th  of  June,  was  thirty  inches  high.,  and 
in  flowers,  having  attached  its  radical  leaves. 

Another  species  which  was  younger  and  cut  May  13,  just 
heading  out,  gave  : 

Water,    ......  ..........................  .........  81.564 

Dry  matter,  .................  ..  ............  ...  .......  18.436 

Ash,  ......  .............  ---------  ,......,,  .........     2.267 

Another,  at  about  the  same  stage  of  growth,  cat  May  20, 


Water,  .  .  .......................  ..................  80.75 

Dry  matter,  ......................................  17.91 

Ash,    ............................................     1.34 


The  analysis,  however,  was  confined  to  the  stalk  ;  the  leaf 
of  the  stalk  gave  : 

Water,  ...........................................  75.50  A 

Dty  matter,  ......................................  2l.5t> 

Ash,  .............................................     2.84 

lu  three  trials  for  the  quantity  of  ash  in  plants  growing  in 
this  country  the  quantity  exceeds   that  obtained  from  the 


,; 


y. 


NORTH-CAJROLINA   GEOLOGICAL   SURVEY.  165 

plant  growing  in,  the  climate  of  England.  Prof.  Way  ob- 
tained ash  1.95.  The  June  grass  contains,  according  to  Prof. 
Way; 

Alluminous  or  flesh  forming  elements,  10.35 

Falty  matters, 2.63 

Heat  producing  elements,  consisting  of  starch,  sugar 

and  gum,  43.06 

Woody  fibre,  38.02 

Mineral  matter,  or  ash, 5.94 

The  latter  is  calculated  from  the  dry  substance.    The  ash  of 
the  June  grass  which  I  submitted  to  analysis,  gave : 

Silex, 56.320 

Earthy  and  alkaline  phosphates,   14.980 

Carbonate  of  lime, 3.540 

Potash,  15.624 

Soda, 6.828 

Magnesia, 1.996 

Sulphuric  acid, 200 

Chlorine,    863 

100.351 

The  plants  were  selected  from  well  made  hay. 


166 


NORTH-CAROLINA   GEOLOGICAL    SURVEY. 


POA     8EROTINA LATE     FLOWERING     MEADOW-GRASS FALSE     RED- 
TOP FOWL   MEADOW.— (Fig.    12.) 


(FiG.  12.) 

Stem  and  leaves  smooth.  Panicle  elongated  diffuse,  branch- 
es in  fives  or  sixes  whorled.  Spikelets  ovate,  accuminate 
three  to  four  flowered,  tinged  with  yellow  at  the  apex  ;  glumes 
long,  lanceolate,  very  acute ;  paleae  lanceolate,  rather  obtuse 
and  pubescent  at  base. 

The  leaves  are  2.63  lines  wide,  and  4  or  5  inches  long ;  root, 


NORTH-CAROLINA    GEOLOGICAL    SURVEY.  107 

perennial.  Flowers  in  July.  Ripens  about  the  first  of  Au- 
gust, and  becomes  drooping. 

It  grows  best  in  moist  places  or  meadows,  and  yields  abun- 
dantly. Its  hay  is  excellent;  sheep  and  other  stock  eat  it 
with  avidity  and  thrive,  especially  if  mixed  with  clover.  It 
is  highly  esteemed  in  Europe.  It  grows  well  in  the  south- 
western States.  Some  think  it  superior  to  Timothy  as  its 
culms  are  more  tender. 

It  grows  in  all  parts  of  New  England  and  New  York,  and 
is  esteemed  by  all  for  its  qualities.  It  is  quite  productive. 
It  grows  three  feet  high,  and  is  liable  to  lodge  or  fall  down  in 
consequence  of  its  slender  stalk. 

There  is  no  doubt  this  fine  grass  may  be  cultivated  in  the 
low  rich  grounds  of  the  eastern  counties,  particularly  in  parte 
of  Hyde  county. 

The  genus  Poa  contains  a  large  number  of  species  which 
inhabit  woods  and  woody  places,  or  high  and  mountainous 
regions.  Although  known  to  be  relished  and  eaten  by  cattle, 
they  do  not  yield  enough  to  make  it  an  object  to  introduce 
them  into  the  cultivated  fields.  Thus,  the  Poa  nemoralis, 
wood  meadow  grass,  is  a  good  grass  so  far  as  its  properties 
are  concerned.  It  has  been  recommended  for  cultivation  by 
Sinclair,  who  remarks  that,  although  the  produce  is  inconsid- 
erable, yet  its  early  growth  in  the  spring,  and  its  remarkably 
fine  succulent  herbage,  recommend  it  for  admission  into  com- 
pany with  others  which  form  good  pasture  grasses.  For  hay 
it  is  not  recommended  as  its  yield  would  be  too  inconsiderabe 
to  deserve  attention.  It  flowers  early  in  May. 


168 


NORTH-CAROLINA   GEOLOGICAL   SURVEY. 


POA  NEBVATA. — (Fig.    13.) — MEADOW   SPEAK   GKASS — FOWL, 
MEADOW   OF   SOME   FARMERS. NERVED   MANNA   GRASS. 


(Flo.  13:) 

The  stem?  is  slightly  compressed — bears  an  open  or  spread- 
ing panicle,  with  small  ovate,  oblong  and  green  spikelets — 
leaves  in  two  rows,  and  rough,  and  grows  from  two  to  three 
feet  high. 

This  American  grass  is  highly  nutritive.  The  ripening  of 
the  seed  does  not  diminish  the  nutritive  value  of  the  stem  and' 
leaves.  It  is  hardy,  grows  best  in  most  places.  It  is  eaten 
by  cattle  both  in  summer  and  winter,,  but  is  more  relished  in. 
the  latter  than  in  the  former  season,. 


GEOLOGICAL   SURVEY; 
FESTUCA. 

Glumes  two,  unequal,  many  flowered.  Palese  two  lanceo- 
late ;  outer  one  accuminate,  or  awned.  Panicle  usually  com- 
pound. 

FESTUCA   OVINA — SHEEP    FESCUE. (Fig.    14:.) 

Stem  slender,  surmounted  by 
small  panicale,  with  spikelets  from 
two  to  six  flowered  ;  awn  inconsid- 
erable ;  leaves,  bristle  shaped,  red- 
dish or  greenish.  It  grows  from  6 
to  10  inches  high,  in  dense  peren- 
nial rooted  tufts.  1 1  flowers  in  June 
and  July ;  grows  in  dry  pastures, 
and  makes  an  excellent  pasturage 
for  sheep. 

FESTUCA  PRATENSIS — MEADOW  FESCUE. 

(Fig.  15.) 

Its  panicle  is  branching,  nearly 
erect,  slightly  one-sided,  and  with 
linear  spikelets,  and  with  from  five 
to  ten  cylindrical  flowers ;  color  of 
the  leaves  of  a  glossy  green,  lower 
ones  broad  and  pointed  and  with 
roughish  edges,  root  creeping  per- 
ennial. Flowers  early  in  June.  It 
grows  in  rather  wet  open  grounds 
to  the  height  of  two  or  three  feet. 

The  qualities  of  this  grass  giveit  a 
tolerable  high  rank  among  the  pas- 
ture grasses.  It  has  long  tender 
leaves,  which  are  relished  by  cattle. 
It  sometimes  forms  a  good  turf  in 
old  pastures.  When  sown,  it  should 
be  mixed  with  orchard  grass,  June 
grass,  or  common  spear-grass. 
(FIG.  14.)  The  figure  was  taken  from  a  spe- 

cimen near  its  maturity,  and  past  flowering. 


170 


NORTH-CAROLINA   GEOLOGICAL   SURVEY. 


FESTUCA  LOLIACEA — SLENDER-SPIK- 
ED  FESCUE. 

Stem  erect,  slender ;  spikelets 
acute,  close  pressed,  rather 
crowded,  and  from  ten  to  twelve 
in  number.  It  grows  in  moist 
meadows  in  small  tufts,  root  per- 
ennial. It  is  a  nutritive  grass, 
and  would  form  good  pastures, 
but  it  is  too  rare  to  be  ranked 
among  those  worth  cultivating. 
The  fescue  grasses  are  com- 
mon in  most  meadows,  and  occu- 
py shady  as  well  as  sunny  places ; 
among  the  most  valuable  and 
common  of  the  tribe,  is  the  Festu- 
ca  pratensis.  Its  stem  is  round 
and  smooth,  and  from  2  to  3  feet 
high,  with  creeping  roots,  and 
surmounted  by  an  erect  branch- 
ed panicle,  and  somewhat  one- 
sided ;  spikelets  linear,  with  from 
live  to  ten  flowers.  The  leaves 
are  long  glossy  green  striated, 
and  have  rough  edges. 

Flowers  in  June  and  grows 
in  moist  pastures.  It  ripens  its 
seeds  early,  and  hence  takes  pos- 
session of  the  ground  before  oth- 
er grasses  are  matured.  It  is  a 
nutritive  plant,  growing  in  stiff 
moist  soils,  and  in  shaded  places. 
Darby  does  not  speak  of  it  as  a 
southern  grass. 
BROMUS. 

Glumes  two,  many  flowered,  and  shorter  than  the  florets ; 
florets  imbricate  in  two  rows ;  lower  palese  cordate  emargin- 
ate,  and  sometimes  armed  with  an  awn  below  the  summit ; 
scales  ovate  smooth. 


NORTH-CAROLINA    GEOLOGICAL    SURVEY. 


171 


BROMU8 


SECALINUS CHESSCHEAT. — (Fig.  16.) 

Stem  glabrous,  erect, 
swollen  at  the  joints,  leaves 
ciliate,  pubescent  on  the 
upper  surface.  Panicle 
branching  -jerect  or  nod- 
ding ;  spikelets  compressed 
oblong  ovate,  florets  about 
10  longer  than  the  bristles. 

The  remarkable  views 
which  are  entertained  of 
this  plant,  excuse  the  no- 
tice of  this  worthless  grass 
in  this  place.  It  has  been 
a  common  opinion  with  a 
very  large  proportion  of 
farmers,  that  wheat  chan- 
ges into  chess,  the  grass 
under  consideration.  This 
has  frequently  been,  in 
one  sense,  favored  by  the 
fact  that  when  wheat  has 
been  winter-killed,  chess 
has  sprung  up  in  its  place, 
therefore,  to  those  who 
have  not  been  careful  ob- 
servers, it  has  seemed  that 
the  wheat  itself  has  un- 
dergone the  change  which 
they  maintain ;  usually, 
this  view  seems  rational, 
because  chess  has  not  been 
observed  by  them  in  this 
particular  place  in  former 
times.  Notwithstanding 
this  apparent  support  to 
the  doctrine,  it  only  re- 
quires a  good  eye  to  detect 
chess  in  almost  any  corner 


(Fio.  16.) 


172  NORTH-CAROLINA   GEOLOGICAL   SURVEY. 

of  a  cultivated  field,  and  if  it  has  not  appeared  before  on  a 
particular  spot,  it  has  probably  been  owing  to  the  fact  that  it 
has  been  occupied  by  other  plants  and  grasses  which  ex- 
clude it. 

Facts,  when  properly  ascertained  and  sifted,  never  sustain 
the  doctrine  of  a  change  of  one  species  to  another.  There  is 
in  nature  no  transmutation  of  the  kind.  Northern  Indian  corn 
after  growing  in  the  south  for  a  few  years,  assumes  the  habits 
and  appearance  of  southern  corn,  which  is  a  thing  quite  dif- 
ferent from  the  one  under  consideration,  the  change  of  one 
species  into  another.  Chess,  though  it  possesses  some  nutri- 
ment, yet  it  is  too  low  to  encourage  its  propagation.  It  is 
rather  a  pest  which  should  not  be  allowed  to  mature  seed, 
and  thereby  propagate  itself  among  the  valuable  grains  and 
grasses.  It  is  an  annual  grass,  but  if  cut  early,  will  spring  up 
and  propagate  itself  the  succeeding  year. 


NORTH-CAROLINA   GEOLOGICAL   SURVEY. 


173 


COCKSFOOT    GRASS 


ORCHARD    GRASS,    DACTYLIS    GLOMERATA. 

(Fig.  17.) 

Flowers  in  --dense  tufts  or 
spikelets,  crowded  in  clusters, 
one-sided,  with  a  dens  branch- 
ing panicle  at  top.  It  grows 
erect  and  attains  a  height  of 
three  feet ;  not  perennial ;  it 
is  a  native  of  Europe,  but  has 
been  naturalized  in  many 
parts  of  this  country,  and  El- 
liott says  that  it  has  become 
naturalized  on  James  Island, 
near  Charleston,  South-Caro- 
lina. This  being  the  case,  fur- 
nishes sufficient  evidence  that 
it  is  an  important  grass  for  the 
South. 

The  orchard  grass  is  very 
widely  distributed.  It  is  well 
known  in  the  north  of  Africa, 
Europe,  Asia  and  America. 
It  is  said  that  it  was  introduced 
into  England  from  Virginia 
where  it  now  forms  one  of  the 
most  common  grasses  of  Eng- 
lish pastures,  is  highly  es- 
teemed among  cattle  feeders, 
being  exceedingly  palatable 
to  stock  of  all  kinds. 

This  grass  is  worthy  of  cul- 
ture from  its  rapid  growth, 
luxuriant  aftermath,  and  its 
endurance  of  close  cropping, 
and  when  fed  down  closely  it 
recovers  in  a  shorter  time  than 
any  other  grass  under  culti- 
!7.)  vatioik  It  forms  an  excellent 


NORTH-CAROLINA    GEOLOGICAL    SURVEY. 

grass  for  mixing  with  clover ;  it  is  free  from  the  objection 
which  applies  to  the  case  of  timothy,  as  it  reaches  its  mature 
state  about  the  same  time  as  clover.  Hence,  it  will  be  per- 
ceived that  it  is  an  earlier  grass.  The  time  for  cutting  it  for 
winter  food  is  when  it  has  blossomed.  If  delayed  until  the 
seeds  have  ripened,  it  is  far  less  valuable,  as  it  loses  at  this 
stage  its  juiciness.  Thick  tufts  of  it  form  in  pasture  lands. 
when  it  is  not  fed  close.  As  it  regards  resistance  of  drouth, 
it  is  well  known  that  it  bears  it  well,  in  which  respect  it  is 
quite  unlike  the  timothy.  Good  observers  declare  that  it 
produces  more  pasturage  than  any  other  grass.  On  this  point 
the  opinion  of  the  late  Judge  Buel,  of  Albany,  coincided  with 
other  eminent  agriculturists,  and  all  agree  in  two  other  im- 
portant points,  viz :  that  it  should  be  kept  fed  close  and  that 
when  it  has  had  only  five  or  six  days  to  recover,  it  acquires  a 
good  bite  for  cattle.  These  points  give  it  a  preference  again 
over  timothy.  Sheep  are  more  fond  of  it  than  any  other 
grass.  It  is  less  exhausting  to  the  soil  than  many  other  nu- 
tritive grasses,  which  arises  from  the  lightness  and  small 
amount  of  seed  which  it  produces.  A  bushel  of  seed  weighs 
only  twelve  or  fourteen  pounds.  This  grass  is  but  little  culti- 
vated in  New  England,  probably  from  the  preference  given 
to  timothy  and  red  top,  which  is  rather  remarkable,  seeing  so 
much  hay  and  pasturage  is  required.  One  of  the  finest  fields 
of  grass  the  writer  ever  saw  was  upon  the  plantation  of  Col. 
Capron,  at  the  Laurel.  Orchard  grass,  when  sown  spar- 
ingly and  upon  uneven  ground,  is  disposed  to  grow  in  tus- 
socks. This  fault  may  be  remedied  by  preparing  the  ground 
properly  and  sowing  a  snfh'cieut  quantityy  of  seed.  This 
grass,  however,  should  not  be  cultivated  by  itself,  unless  it  is 
wished  to  grow  it  for  seed.  The  celebrated  Sinclair  gives  ;i 
formula  for  the  formation  of  a  crop  for  pasturage.  lie  rnixe'.i 
the  seeds  of  certain  grasses  in  the  following  proportions : 

Doctylis  plomenata,   4  pecks. 

Festuca  pratensis,    3    do. 

Timothy,    %  do. 

Fiorin,  or  agrostis  stolonifera,    1     do. 

H  olcus  arenaceus,  2    do. 

Lolium  perenne,  8     do. 


NOKTH-CAKOLINA   GEOLOGICAL   SURVEY.  175 

Poterium  songuisorba,  (burnet)  2  pecks. 

Trifolium  pratense,  red  clover,  6  Ibs. 

"         repens,  white  clover,  8    do. 

This  mixture  was  regarded  as  sufficient  for  an  acre.  We 
see  in  this  prescription  a  love  for  variety  and  an  excessive 
amount  of  seed.  As  pasturage  is  one  of  the  great  desiderata 
in  this  State,  and  as  this  grass  stands  dry  weather  remarkably 
well,  it  will  probably  be  one  of  the  most  important  measures 
in  husbandry  to  encourage  its  cultivation.  "Whether  it  can 
be  shown  hereafter  that  it  will  give  as  much  profit  per  acre 
as  has  been  reported  for  a  field  near  Rochester,  N.  Y.,  can 
only  be  determined  by  experiment.  The  profits  reported  as 
having  been  reared  from  one  and  a  quarter  acres  of  ground 
were  given  in  the  Genesee  Farmer,  Vol.  Y,  p.  245  : 

There  were  obtained  17  bushels  of  seed,  $2  per  bushel $34  00 

Yielding,  also,  2  tons  of  hay,  $10  per  ton,   20  00 

for  the  first  crop. 
There  were  obtained  1}^  tons  for  the  second  crop,   15  00 


Amounting  to  $69  00 

Expense  for  gathering  crops : 

Cutting  and  shocking  seed,  one  hand  half  a  day,  0  50 

Threshing,     1  00 

Cutting  stuble,    ]  00 

Making  the  same  into  hay  and  overhauling, 1  50 

Cutting  and  making  hay  of  the  second  crop, 2  00 

Interest  on  the  value  of  land,  4  87 

$10  87 
Deducted  from  sales,  leaves  a  nett  gain  of  53  1 2 

To  save  the  seed  properly  requires  the  skill  of  a  good  cracl- 
ler,  who  cuts  the  tops  and  ties  them  in  bundles  to  dry  in  the 
field  for  eight  or  ten  days.  They  should  be  hauled  into  the 
barns  and  threshed  immediately  with  a  flail.  If  there  is  a 
large  quantity  of  seed  it  should  be  still  allowed  to  dry  upon 
the  floor,  as  when  retaining  moisture  it  is  apt  to  heat  in  the 
heap,  when  the  vitality  of  the  seed  is  destroyed.  The  seed 
us  above  stated,  is  very  light.  If  sown  with  clover,  one 
bushel  of  orchard  grass  to  ten  quarts  of  clover  seed  makes 


176  NORT^-CA'KOLINA   GEOLOGICAL   SURVEY. 

the  proper  preparation  ,per  acre.  If  sown  alone,  two  bushels 
are  required.  For  pasturage  alone,  a  mixture  of  the  white 
clover  will  form  an  excellent  addition.  Whatever  opinions 
may  prevail  with  respect  to  the  cultivation  of  the  grasses  in 
the  eastern  part  of  the  State,  or  even  the  middle,  there  can 
be  but  little  doubt,  that  when  the  attempt  is  made  to  intro- 
duce a  more  extended  pasturage,  this  grass  will  have  the 
preference  over  many  others. 

The  analysis  of  the  ash  of  the  orchard  gave,  Prof.  Way : 

'Silica,    26  65 

Phosphoric  acid,  8.60 

Sulphuric  a'cid,  , 3.52 

'Carbonic  acid,   2.09 

Lime, ,. .  5.82 

Magnesia,   2.22 

Per  oxide  of  iron,  0.59 

Potash, 29.52 

Chloride  of  potassium,  17.86 

^Chloride  sodium, 3.09 

Percentage  of  ash  furnished  by  the  dry  plant,   5.51 

The  nutritive  value  of  this  grass  is  exhibited  in  the  follow- 
ing analysis  of  Prof.  Way  : 

Water,    70.00 

Albuninous  matter,  (flesh  forming,) 4.06 

Falty  matters,  0.94 

Starch  gum  sugar,  (heat  producing  bodies,)  13.30 

Woody   fibre,  10.11 

Ash,  1.59 

ELYMUS WILD   RYE. 

It  has  two  or  more  spikelets  at  the  joints  of  the  rachis,  and  is 
from  3  to  9  flowered.  Glume  2,  nearly  equal,  sometimes 
wanting;  lower  paleae  entire  with  a  short  awn  ; — upper  one 
bifid.  Scales  ovate  hairy. 

ELYMUS    ARENARIUS. — UPRIGHT   SEA   LIME   GRASS. 

Stem  erect,  round,  smooth  from  two  to  five  feet  high,  and 
bearing  sessile  spikelets  ;  leaves  long,  narrow,  rolled  inward, 
and  rough  on  the  inner  surface  ;  root,  long,  perennial  and 
creeping. 


NORTH-CAROLINA    GEOLOGICAL    SURVEY.  IVl 

Resembles  beach  grass  in  its  mode  of  growth ;  it  is  also  a 
valuable  grass  for  confining  blowing  sands. 

In  England  it  is  called  the  sugar  cane,  from  the  quantity 
of  sugar  in  its  stem. 

The  E.  virginicus,  (wild  rye,)  E.  canadensis,  (Canadian 
lyme  grass,)  E.  striatus  (slender,  hairy  lyme  grass,)  grow  along 
the  banks  of  rivers  and  streams,  but  they  are  of  no  special 
value  for  cultivation. 

LOLIUM. 

Spikelets  many  flowered,  solitary  on  each  point  of  a  con- 
tinuous rachis,  placed  edgewise. 

LOLIUM    PERENNE. 

Stem  erect,  smooth,  leaves  flat,  acute,  smooth  on  the  outer 
surface,  roughish  on  the  inner,  glume  shorter  than  the  spike, 
flowers  from  six  to  nine,  awnless.  Flowers  early  in  June. 
From  15  to  24  inches  high.  Root  perennial,  creeping. 

This  is  regarded  as  valuable  grass  both  in  England  »pid 
France.  It  is  relished  by  stock  previously  to  its  blossoming, 
afterwards  it  becomes  hard  and  less  palatable. 

It  is  not  equal  to  the  orchard  grass  in  any  respect,  but  at 
the  same  time  it  must  be  admitted  that  it  could  not  have  stood 
its  ground  so  long  in  England  and  France  unless  its  merits  are 
considerable.  It  is  doubtful  whether  it  can  be  cultivated  in 
this  State  with  pron't.  It  seems  to  attain  perfection  in  a  more 
humid  climate  than  ours. 

LOLIUM   ITALICUM ITALIAN  RYE-GRASS. 

It  is  inferior  to  our  best  grass,  as  timothy,  orchard-grass, 
blue-grass,  etc.  In  some  points  of  view,  however,  it  is  supe- 
rior to  them,  as  it  may  be  cut  several  times,  when  sown  upon 
moist  rich  land.  It  grows  luxuriantly,  and  for  soiling  cattle 
it  is  an  excellent  addition  to  our  grasses,  as  it  bears  cutting- 
well.  Its  actual  value  to  us,  however,  is  still  to  be  determin- 
ed by  farther  experiments. 
13 


ITS  NOETH-CAEOLINA   GEOLOGICAL   SUE  YET. 

LOL1UM   MULTIFLOEUM MANY-FLO WEEED    DAENEL. 

This  grass  is  so  little  known  in  this  country,  that  it  may  be 
passed  over  without  remark. 

TKITICUM  — WHEAT. 

Flowers  in  spikes ;  spikelets  imbricate  sessile  ;  J  flowered. 
Glume  two,  nearly  equal  opposite  ;•  palese  lanceolate ;  the 
lower  concave  acaminate  awned ;  scales  two  ciliate.  . 

Wheat  is  supposed  to  have  been  indigenous  to  Central  or 
South-western  Asia.  It  is  known  to  have  been  cultivated 
from  the  earliest  times. 

Like  the  Indian  corn  its  varieties  are  numerous,  amounting 
at  the  present  time  to  about  S'OO,  which  are  known  to  be  un- 
der cultivation. 

The  characters  of  these  varieties  are  essentially  the  same. 
The  modifications  affecting  merely  its  appendages  without  ex- 
tending to  its  essential  characteristics.  The  character  of  the 
soil  influences  the  value  of  the  grain  ;  it  is  always  richer  and 
better  on  rich  substantial  soils.  When  grown  upon  those 
which  abound  in  vegetable  matter  its  grain  is  light. 

TEITICTJM      BEPENS COUCH-GEASS SWITCH-GEASS DOG— GEASS — 

DUTCH-GEASS — QTTACK-GEASS. 

It  has  an  erect  stem,  with  smooth  joints,  two  upper  most 
remote;  spikelets  close  pressed,  leaves  acute,  upper  one 
broadest ;  sheaths  striated,  roots  creeping  extensively.  Intro- 
duced from  Europe  ;  flowers  in  June,. 

This  grass  is  cut  in  blossom, — is  relished  by  cattle,  and 
makes  a  nutritious  hay.  In  gardens  and  other  cultivated 
grounds  it  becomes  a  great  pest,  from  the  difficulty  of  eradi- 
cating it.  Its  roots  are  short-jointed,  and  send  out  fibres  from 
all  of  them,  in  consequence  of  which  it  grows  and  maintains 
itself  when  a  single  joint  remains,  besides  it  is  tenacious  of 
life,  and  does  not  readily  die  when  left  upon  the  earth's  sur- 
face. 

This  grass  cut  in  May  13,  gave, 


NORTH-CAROLINA   GEOLOGICAL   8UKVEY.  179 


Water,    81.564 

Dry  matter, 18.436 

-Ash,    2.367 

A  second  specimen  from  the  same  bed,  cut,  June  8,  gave, 

Water,    '. 77.374 

Dry  matter, 22.626 

Ash, 2.073 

As  this  grass  approaches  maturity,  its  inorganic  matter  de- 
creases and  its  woody  fibre  increases.  A  third  specimen  ta- 
ken when  in  full  blossom,  gave, 

Water,  68.50 

Dry  matter, 30.50 

Ash,    1.00 

An  analysis  of  the  ash  of  this  grass,  gave  me, 

Silica,    27.150 

Phosphates  of  lime,  magnesia  and  iron, 19.250 

Lime, 0.112 

Magnesia,  trace 

Potash, 10.350 

Soda,  26.985 

Chloride  of  sodium,  8.990 

Sulphuric  acid, 4.811 

Carbonic  acid, 1.455 

The  same  change  takes  place  in  the  lolium  perenne.  These 
experiments  have  an  important  bearing  on  the  time  they 
should  be  cut  for  hay.  It  is  well  known  that  stock  relish  grass 
and  hay  while  it  is  succulent  and  juicy.  After  the  woody 
fibre  is  largely  formed  it  is  less  palatable  and  more  difficult  to 
masticate ;  besides,  it  wears  the  teeth  more,  and  less  nutri- 
ment is  taken  into  the  system. 

CYNOSURUS   CRISTATTJS- — CRESTED   DOG's-TAIL. 

Its  stems  are  about  one  foot  high,  stiff  and  smooth,  provided 
with  fibrous  perennial  root,  more  or  less  tufted.  Its  stem  being 
hard  and  wiry,  cattle  usually  refuse  to  eat  it.  In  dry  sheep 


ISO  NOKTH-CAEOLINA   GEOLOGICAL   SURVEY. 

pastures,  it  is  more  valuable  as  a  permanent  grass.     Its  stem 
is  used  in  the  manufacture  of  straw  plait. 

The  common  broom-sedge  is  another  grass  whose  stem  and 
leaves  become  hard  and  wiry  with  age,  and  still  more  unfit 
for  food  for  cattle  than  any  ot  the  preceding.  It  takes  posses- 
sion of  old  and  worn  out  fields,  and  imparts  to  them  a  look 
of  barrenness,  which,  in  many  instances,  they  do  not  deserve. 
Cattle  eat  this  grass  only  in  the  spring,  when  it  first  springs 
up,  and  wThen  it  is  comparatively  tender.  Although  almost 
worthless  for  fodder  when  mature,  it  is  still  better  for  the 
ground  to  be  covered  and  protected  by  this  grass  than  to  be 
naked  and  exposed  to  the  heat  of  the  sun  and  the  action  of 
rains.  This  grass  has  but  a  small  proportion  ot  nutrient  mat- 
ter ;  at  the  same  time  the  consideration  how  fields  should  be 
treated  when  covered  with  it,  is  worth  a  moment's  considera- 
tion. When  such  a  field  is  to  be  ploughed  for  a  crop  of  wheat, 
it  is  important  to  lay  it  under  while  it  is  still  green,  or  before 
it  has  reached  its  full  maturity.  At  this  period  it  is  more 
valuable  as  a  fertilizer ;  the  proportion  of  silex  in  the  stem 
being  relatively  less  and  the  more  valuable  elements  are 
greater.  When  mature,  it  contains  about  72  per  cent,  of  sil- 
ica, and  only  8  per  cent,  of  the  phosphates  of  lime  and  mag- 
nesia. The  only  grass  which  approaches  this  in  its  mature 
state  in  the  proportion  of  silica,  is  the  Italian  rye-grass,  which 
contains  60  per  cent.  In  burning  off  a  crop  of  broom-grass, 
a  large  proportion  of  this  silica  becomes  insoluble.  Hence  it 
should  be  ploughed  under  when  well  grown,  when  all  its  nu- 
tritive elements  are  in  the  best  condition  to  aid  the  growth  of 
the  succeeding  crop. 


NORTH-CAROLINA   GEOLOGICAL   SURVEY.  181 


CHAPTEE  XIII. 

Red  clover  belongs  to  the  Leguminosae — Organic  constitution — Composition 
of  its  ash — Differs  in  composition  from  the  grasses-— Failures  in  its  culti- 
vation— For  a  green  crop — Lucerne — Sanfoin. 

§  113.  In  the  northern  and  western  sections  of  the  United 
States  the  red  clover,  though  not  a  grass,  is  now  regarded  as 
one  of  the  important  resources  of  husbandry.  It  forms  of  it- 
self an  excellent  food  for  cattle.  It  is  one  of  the  most  speedy 
and  effectual  means  by  which  soils  may  be  brought  to  pro- 
duce remunerating  crops.  It  is  therefore  both  a  nutriment 
direct  for  cattle,  and  a  fertilizer  for  the  cereals.  It  is  in  vir- 
tue of  its  rapid  growth,  large  herbage  and  roots  that  it  occu- 
pies a  place  in  husbandry  so  important ;  besides,  it  derives  no 
inconsiderable  part  of  its  substance  from  the  air.  In  the  nat- 
ural classification,  it  belongs  to  the  family  leguminoscey  or  the 
game  family  as  the  bean  and  pea.  Its  common  name,  clover, 
is  most  in  use.  It  is  sometimes  designated  by  the  term  trefoil, 
three  leaved. 

It  scarcely  requires  a  description,  as  it  is  known  by  every 
farmer  and  planter.  Its  stem  is  inclined  to  be  prostrate  or 
ascending,  and  the  leaves  are  oval,  and  stand  in  threes  at  the 
termination  of  the  stem. 

The  red  clover,  after  many  years  cultivation,  has  developed 
a  number  of  varieties.  One  of  these  varieties  is  biennial  and 
another  is  perennial,  and  like  many  other  biennials  which  has 
become  so  in  other  families  of  plants,  it  frequently  lasts 
three  or  four  years,  provided  it  is  not  suffered  to  go  to  seed. 

Clover  is  a  very  easy  plant  to  cultivate  in  a  cool,  moist  cli- 
mate. In  one  similar  to  North  Carolina,  which,  perhaps,  is 
more  subject  to  droughts  than  New  England  or  New  York,  it 
is  more  difficult.  This  arises  from  the  tenderness  of  the  young 
plant.  In  its  early  stage,  if  exposed  to  a  burning  sun,  it  dies. 
But  it  is  not  difficult  to  protect  beneath  the  shade  of  another 
plant,  and  thereby  save  it  from  perishing. 

Clover  is  a  nutritious  fodder,  and  cattle  and  horses  are  very 


NORTH-CAROLINA  GEOLOGICAL   SURVEY. 

fond  of  it.  But  as  it  frequently  grows  very  rank,  it  is  not 
perfectly  cured,  and  in  a  green  state  it  moulds.  If  fed  to  a 
horse  in  this  condition,  which  is  at  all  inclined  to  the  heaves, 
it  will  certainly  produce  it. 

As  a  nutriment,  clover  takes  rank  with  the  best  of  grasses. 
According  to  Prof.  Way,  red  clover  contains, 

Water,    81.01 

Albumen,  4.27 

Fatty  matters,    69 

Gum,  starch,  sugar,  or  heat-producing  principles,  . . .     8.45 

Woody  fibre,  3.76 

Ash, 1.82 

Clover  is  a  lime  plant,  but  this  element  increases  with  its 
age.  In  the  young  plant  the  proportion  is  much  smaller  than 
in  the  old.  Thus : 


OLD.  TOTING. 

Silica,... 0.850  0.981 

Phosphates  of  lime,  and  magnesia,  etc., 20.600  30.245 

Carbonate  of  lime,  30.950  7.642 

Magnesia, ; 3.930  2.285 

Potash,    25.930  33688 

Soda,    14.915  7.164 

Chlorine, 1.845  3.642 

Sulphuric  acid,   0.495  6.723 

Carbonic  acid,  5.744 

The  upper  part  of  the  stem,  with  the  leaves  and  heads,  gave 
a  composition  varying  from  the  above,  thus : 

Silica 0.810 

Phosphates,  21.900 

Carbonate  of  lime,  32.333 

Magnesia,   0.200 

Potash,  27.940 

Soda,  6.7S3 

Chlorine,    3.780 

Sulphuric  acid,  3.366 


From  the  foregoing  analysis  it  will  be  perceived  that  clover 
differs  in  composition  from  the  grasses.     It  contains  only  a 


NORTH-CAROLINA   GEOLOGICAL   SURVEY.  183 

small  per  centage  of  silica-;  and  hence,  cattle  and  horses  mas- 
ticate it  easily.  Two  elements  exist  in  large  proportions,  lime 
and  potash;  and  hence,  it  must  exhaust  a  soil  as  much  as 
timothy  or  any  of  the  best  grasses.  For  this  reason,  clover 
makes  an  excellent  green  crop  to  precede  wheat.  Its  large 
roots  loosen  and  open  the  soil,  and  supply  by  their  decay  a 
large  amount  of  fertilizing  matter. 

I  have  already  remarked  that  clover  has  not  succeeded  well 
in  this  State.  In  many  instances  it  has  not  come  up,  and  in 
others  it  has  died  out.  In  some  instances  it  has  not  been  dif- 
ficult to  assign  a  reason  for  its  failure.  Where  it  has  failed  to 
grow,  I  found  on  enquiry  that  it  had  been  ploughed  in ;  buried 
too  deep.  The  seed,  in  these  cases,  was  not  in  fault.  Clover 
requires  only  a  shallow  covering,  and  especially  if  the  roller 
is  employed,  good  seed  will  come  up.  '  In  other  cases,  after 
It  had  corae  up,  the  planter  allowed  his  pigs  to  have  the  ben- 
efit of  the  young  and  growing  plant.  It  wae,  therefore,  fed 
or  crushed  out.  In  other  cases  it  was  sown  at  the  wrong  time 
and  was  exposed  without  protection  to  the  sun-rays. 

In  nine  cases  out  often,  a  good  stand  may  be  secured  un- 
der the  right  system  of  culture.  Atl  those  causes  of  failure 
which  I  have  named  must  of  course  be  avoided,  and  in  this 
climate  it  will  not  do  to  allow  cattle  and  hogs  to  feed  upon  it 
until  it  is  half  grown,  or  has  acquired  a  strong  root. 

For  a  green  crop  to  be  disposed  of  as  a  fertilizer,  clover  has 
one  advantage  over  the  pea ;  from  the  former,  a  good  crop  of 
hay  may  be  obtained,  and  at  the  same  time  its  stubble  and 
root  ploughed  in.  The  latter,  if  taken  oif  for  fodder  leaves  on 
the  ground  only  a  small  remnant  of  fertilizing  matter.  But  if 
the  whole  pea  is  allowed  to  remain,  it  is  more  valuable  than 
clover,  and  is  better  adapted  to  this  climate,  and  hence  requires 
much  less  care  in  its  cultivation. 

White  clover  is  a  more  hardy  plant  than  the  red,  but  being 
much  smaller,  it  is  not  useful  for  winter  fodder.  For  fine  pas- 
tures it  is  one  of  the  best  of  plants,  though  cattle  do  not  relish- 
it  quite  as  well  as  we  have  reason  to  expect  from  its  sweetness 
and  tenderness ;  yet,  is  eaten  freely  by  sheep,  and  the  meat, 
whether  of  cattle  or  sheep,  is  of  a  fine  quality.  It  is  also  re- 


184  NORTH- CAROLINA   GEOLOGICAL   SURVEY. 

lished  by  swine.  Its  root  being  creeping,  it  spreads  far  and 
wide,  and  makes  a  durable  pasture,  which  bears  close  feeding 
remarkably  well.  Butter  and  cheese  made  from  the  milk  of 
cows  whose  pastures  are  dotted  with  the  white  clover,  is  su- 
perior to  any  other,  all  things  being  equal. 

White  clover  contains,  when  fresh  and  healthy. 

Water,    81.50 

Dry  matter,  16.76 

Ash,    1.75 

In  one  ton  of  clover  there  are  234.08  Ibs.  of  inorganic  mat- 
ter.    The  ash  I  found  composed  of 

Silica,    28.075 

Phosphate  of  lime,  magnesia  and  iron,  19.325 

Carbonate  of  lime,  16.730 

Magnesia,  2.175 

Potash,  10.880 

Sulphuric  acid,   2.305 

Chlorine, 0.615 

Carbonic  acid, 4.234 


99.979 

The  white  clover  differs  from  the  red  in  the  composition  of 
its  ash  in  containing  a  much  larger  amount  of  silica.  It  may 
turn  out  that  the  foregoing  determination  is  erroneous  or  is 
too  large.  It  may  be  accounted  for,  perhaps,  by  supposing 
that  line  sand  adhered  to  the  stem  and  leaves. 

LUCERNE — (MEDICAGO  SATIVA.) 

§  114.  This  plant  belongs  also  to  the  leguminosse  or  pea 
tribe.  It  is  an  inhabitant  of  a  warmer  climate  than  red  clo- 
ver. It  has  been  cultivated  for  fodder  or  the  food  of  cattle 
for  twenty-three  centuries. 

Lucerne  requires  a  soil  especially  adapted  to  it ;  it  is  not 
therefore  so  easily  cultivated  as  clover.  It  requires  a  tolera- 
bly rich  soil,  and  one  that  is  mellow  and  permits  its  roots  to 
penetrate  deeply.  A  light  sandy  soil  does  not  suit  it,  neither 
does  a  stiff  subsoil  which  retains  moisture  strongly,  or  is  im- 


NORTH-CAROLINA    GEOLOGICAL    SURVEY. 


185 


pervious.  A  fair  proportion  of  sand,  clay  and  vegetable  mould 
will  be  found  a  suitable  mixture  for  the  growth  of  lucerne. 
The  climate  of  North  Carolina  is  well  adapted  to  its  cultiva- 
tion. It  would  undoubtedly  grow  well  and  vigorously  on 
many  of  the  pocosin  soils,  whose  composition  is  similar  to  that 
of  Hyde  county,  though  probably  a  better  drainage  may  be 
required.  Still,  a  soil  so  well  adapted  to  Indian  corn  may  be 
expected  to  grow  lucerne  equally  well.  It  sends  down  along 
tap  root,  provided  with  many  fibrous  off-shoots,  which  imbibe 
nutriment  from  a  wide  area.  Hence  its  vigor,  when  well 
located,  and  the  great  amount  of  food  it  furnishes.  Lucerne 
continues  to  produce  good  crops  from  5  to  10  years  in  suc- 
cession. Hence  its  value ;  when  once  thoroughly  rooted  or 
set,  it  is  as  permanent  as  the  best  pasture  lands.  It  would 
seem,  if  we  reason  from  the  effects  of  the  cultivation  of  other 
plants,  that  after  10  years  cropping  the  soil  would  be  perfect- 
ly exhausted.  This  is  not  the  case,  for  it  is  said  to  render  the 
soil  richer.  This  is  going  too  far.  For  though  leguminous 
plants  derive  a  large  portion  of  their  solid  matter  from  the 
atmosphere,  yet  the  inorganic  matter  comes  from  the  soil,  and 
just  as  much  of  it  as  is  removed  from  the  field,  just  so  much 
also  is  the  land  impoverished.  The  reason  of  the  anomaly 
claimed  for  lucerne,  is,  that  it  penetrates  much  deeper  than 
other  plants  and  takes  its  food  from  a  much  greater  space. 

The  best  time  for  cutting  lucerne  is  just  before  it  blossoms. 
If  cut  before  this  period  it  is  too  watery  to  dry  and  cure  well; 
if  later  or  after  blossoming  it  is  too  woody  and  contains  less 
nutriment.  This  is  probably  one  of  the  best  plants  for  soiling 
cattle.  When  cut  it  sprouts  vigorously  again,  and  in  a  climate 
like  that  of  North  Carolina,  it  seems  to  be  the  plant  which 
may  be  relied  upon  to  stand  the  sun  and  drought,  and  at  the 
same  time  furnish  a  forage  superior,  if  any  thing,  to  the  red 
clover.  The  seed  of  lucerne  are  yellow,  and  if  good,  glossy 
and  heavy.  The  first  year  it  should  not  be  cut  too  close  nor 
a  large  amount  of  forage  expected  from  it.  Time  should  be 
given  for  it  to  take  deep  root.  The  second  year  it  begins  to 
pay  and  may  be  relied  upon  for  several  succeeding  years.  It 
should  be  sown  early  in  spring. 


186         NORTH-CAROLINA  GEOLOGICAL  SURVEY. 

According  to  Prof.  Way,  the  proximate  elements  of  lucerne 
are  as  follows : 

Water, 69.95 

Albuminous  matter,    3.83 

Fatty                do         0.82 

Heat-producing  matter, 10.32 

Woody  fibre, 8.74 

Ash, 3.04 

When  the  plant  is  dried  in  a  water  bath  at  212°  Fah.,  the 
albuminous  matter  amounts  to  12.76,  and  the  heat-producing 
to  18.62  per  cent.  The  albuminous  matter  or  flesh-forming 
elements  of  the  Kentucky  blue-grass  are  10.35,  and  its  heat- 
producing  matter  to  43.06.  It  is  therefore  superior  in  flesh- 
forming  elements  to  this  favorite  grass. 

SANFOIN — (HEDYSARUM  ONOBRYCIIIS.) 

•  §  115.  Like  the  clovers  and  lucerne,  sanfoin  is  a  legumi- 
nous plant,  but  differs  from  the  latter  in  many  important  par- 
ticulars. It  has  many  long  leafy  stems.  The  leaflets  are 
smooth  and  pinnate,  or  in  pairs,  rather  oblong  and  pointed, 
and  slightly  hairy  on  the  under  side.  Flower  stalks  are  ter- 
minal and  extend  above  the  leaf  stalks,  and  arranged  in  the 
form  of  a  spike,  with  crimson  and  variegated  blossoms.  The 
stems  grow  from  two  to  three  feet  high.  The  pods  are  flat, 
hard  and  toothed  on  the  edge ;  root  perennial  and  woody ; 
flowers  in  July. 

According  to  the  opinion  of  an  experienced  English  agri- 
culturist, who  has  resided  many  years  in  this  country,  the 
sanfoin  will  prove  a  valuable  addition  to  the  artificial  grasses 
of  this  country.  The  following  remarks  containing  a  sum- 
mary of  his  opinions  I  propose  to  embody  for  the  considera- 
tion of  the  planters  and  farmers  of  this  State. 

In  the  first  place,  it  will  grow  well  on  light  soils,  sandy  and 
gravelly  loams.  It  may  be  sown  after  rye  or  barley,  and 
should  not  be  fed  the  first  year,  or  immediately  after  the  crop 
is  removed.  It  may  also  be  sown  with  grass  seed.  The  fol- 
lowing year  it  may  be  mowed,  and  then  it  is  in  a  condition 
to  be  fed  by  sheep. 


NORTH-CAROLINA   GEOLOGICAL   SURVEY. 


1ST 


This  plant  is  probably  better  adapted  to  horses  than  cattle, 
especially  milch  cows,  or  rather  horses  and  sheep.  Sheep 
consume  the  leaves  and  softer  parts  of  the  stems,  and  then 
horses  eat  readily  the  remainder.  Working  horses  do  well 
on  what  sheep  leave.  Sanfoin  has  been  mown  for  nine 
or  ten  years  in  succession,  and  has  produced  good  crops 
each  year  without  manure.  It  is  not  the  proper  food  for 
milch  cows,  as  it  imparts  a  bitter  taste  to  the  butter.  The 
sod,  after  it  has  been  growing  for  several  years,  is  full  of 
roots,  and  it  is  often  ploughed  and  then  burnt  over.  In  this 
climate  ploughing  and  burning  is  not^advisable. 

The  nutritive  value  of  sanfoin  does  not  differ  materially 
from  lucerne.  It  is  composed,  so  far  as  its  proximate  elements 
are  concerned,  of: 

Water,  76.64 

Alluminous  matter,  4.32 

Fatty  matter,   0.70 

Heat  producing,  10.73 

Woody  fibre, 5.77 

Ash,    1.84 

When  dry,  it  yields  of  alluminous  matter,  18.45,  and  heat 
producing,  45.96. 

CRIMSON   CLOVER (TRIFOLIUM   INCARNATUM.) 

In  some  parts  of  this  country  this  clover  would  no  doubt 
succeed.  It  however,  requires  a  climate  rather  cooler  and 
moister  than  that  of  the  eastern  counties.  But  in  the  moun- 
tainous section  of  the  Southern  States  it  can  hardly  fail  of 
being  received  with  favor.  The  advantages  arising  from  its 
culture,  are,  it  may  be  sown  after  potatoes  are  secured,  and 
produce  a  spring  crop  which  will  be  earlier  by  eight  or  ten 
days  than  lucerne  or  red  clover.  It  produces  two  good  crops 
in  one  year.  It  is,  however,  an  annual,  and  it  requires  as 
much  care  to  insure  success  as  the  red  clover.  For  soiling 
cattle  it  is  well  adapted,  in  consequence  of  its  early  growth. 
If  cut  for  hay,  it  should  be  gathered  as  soon  as  it  is  in  flower. 
The  seed  may  be  obtained  from  the  second  crop.  As  a  gen- 


188  NORTH-CAROLINA   GEOLOGICAL    SURVEY. 

eral  rule,  where  the  red  clover  succeeds,  it  may  also  be  ex- 
pected that  the  crimson  clover  will  succeed  also. 


CHAPTER  XIV. 

Methods  by  which  the  valuable  grasses  may  be  cultivated  successfully — 
Soiling,  and  its  advantages. 

* 

§  116.  In  this  State  it  is  important  in  the  first  place  to  se- 
lect the  proper  field  for  the  cultivation  of  grass  which  it  is  de- 
signed to  cut  for  winter  fodder.  It  appears  to  the  writer  that 
as  summer  heat  and  drouth  are  the  greatest  obstacles  to  the 
successful  cultivation  of  grass  and  hay,  that  such  fields  should 
be  selected  as  suffer  the  least  from  the  operation  of  these 
causes.  Hence  it  is  believed  that  the  meadows  and  low 
grounds  which  are  bordered  by  permanent  streams  and  which 
are  naturally  quite  wet,  but  may  be  laid  comparatively  dry 
are  the  most  suitable  for  grass  lands.  The  first  work  which  is 
required,  is  to  drain  the  field  thoroughly  by  ditching.  Fields 
of  this  description  are  invariably  supplied  with  a  rich  bottom, 
which  is  capable  of  furnishing  an  indefinite  amount  of  nutri- 
ment, or  sufficient  to  sustain  crops  of  hay  for  years  in  succes- 
sion, and  being  also  supplied  with  water  which  percolates 
through  the  lower  strata  of  earth,  are  little  liable  to  suffer 
from  summer  droughts.  Besides,  these  low,  fiat  meadows 
may  be  cheaply  irrigated  if  necessary.  Irrigation  is  also  one 
of  the  cheapest  and  most  effectual  means  by  which  nutriment 
may  be  conveyed  to  the  grass.  The  great  object,  however, 
to  be  attained  in  the  selection  of  such  field,  is  that  of  securing 
a  cool  and  moist  soil,  for  many  of  the  best  grasses  are 
found  flourishing  under  those  conditions,  though  they  by  no 
means  grow  in  wet  bogs  or  swamps.  Timothy,  one  of  the 
best  of  the  Northern  grasses,  grows  best  in  a  moist  soil. 


NOKTH-CAKOLINA    GEOLOGICAL    SURVEY.  189 

After  a  drainage  has  been  effected,  many  of  the  wild  and 
least  useful  grasses  will  die  out.  But  to  aid  the  process  of 
substitution  of  better,  for  the  poorer  grasses,  and  the  weeds 
which  always,  more  or  less,  take  a  joint  possession  of  such 
fields,  it  may  be  harrowed  with  an  instrument  provided  with 
sharp  teeth.  When  this  is  done,  a  proper  mixture  of  seed 
may  be  sown,  after  which  the  surface  is  swept  over  with  a 
heavy  brush. 

The  introduction  of  the  valuable  grasses  is  also  materially 
aided  by  a  top  dressing  of  compost,  which  puts  the  soil 
in  a  better  condition  to  receive  the  seed,  and  facilitates,  as 
well  as  quickens,  its  germination.  It  also  gives  more  strength 
to  the  newly  introduced  grass,  and  enables  it  to  contend  more 
successfully  with  those  which  are  already  in  possession  of  the 
premises.  As  in  law,  so  in  agriculture,  possession  gives  im- 
portant advantages  ;  and  the  new  claimant  which  we  desire 
to  put  in  possession,  must,  in  the  first  place,  oust  the  old  oc- 
cupant. Much  depends  upon  the  perfection  of  our  prelimin- 
ary steps.  If  we  have  thoroughly  under-drained  the  premi- 
ses, we  shall  be  enabled  to  starve  out  very  speedily  the  occu- 
pant we  wish  to  remove  ;  and  if,  in  addition  to  this,  we  sup- 
ply nutriment  to  our  favorite  intruder,  we  have  provided  or 
opened  more  than  one  way  by  which  we  hope  to  succeed. 
The  poor  grasses  are  generally  destroyed  by  high  cultivation, 
and  so  are  weeds,  and  the  process  which  so  evidently  favors 
the  disappearance  of  the  poorer  ones,  favors  the  introduction 
of  the  good.  One  of  the  most  substantial  reasons  why  grass- 
es are  so  difficult  to  grow  in  the  South,  is,  that  they  are  not 
manured.  They  are  sown  first  upon  soil  already  partially  ex- 
hausted, where  the  poor  grasses  are  taking  deep  root,  and 
hence  their  chance  for  life  is  very  small. 

If  a  grass  plat  is  to  be  formed  upon  upland,  the  proceeding 
should  be  somewhat  different.  After  the  land  is  made  even 
by  light  ploughing  and  harrowing,  winter  rye  should  be  sown, 
and  the  field  stocked  down  with  orchard  grass,  mixed  with 
herds  grass,  June  grass  and  red  and  white  clover.  The  rye 
makes  an  excellent  spring  fodder,  and  protects  the  grass  seed, 
which  in  due  time  will  germinate  and  replace  the  rye.  To 


190  NOETH-CAEOLINA   GEOLOGICAL   SURVEY. 

iosure  success,  let  the  seed  be  sown  thickly,  not  sparingly,  for 
the  writer  believes  that  in  the  climate  of  North-Carolina  more 
seed  is  required  than  where  the  climate  is  cooler.  Besides, 
there  is  no  more  effectual  means  to  guard  against  drouth,  and 
a  hot  sun,  than  to  cover  the  whole  surface  with  vegetation, 
and  the  supplying  this  vegetation  with  abundant  nutriment. 
In  support  of  this  view,  let  a  field  of  Indian  corn  be  ex- 
amined, a  part  of  which  has  grown  sufficiently  to  shade  the 
soil,  and  part  is  backward  from  any  cause,  and  does  not  shade 
it.  The  first  will  sustain  a  drought  without  material  injury, 
while  the  other  will  be  destroyed.  So  also,  where  clover  has 
taken  a  strong  and  vigorous  hold  and  covers  the  ground,  it 
stands  a  severe  drouth,  while  that  portion  of  the  field  which 
is  thinly  planted,  dries ;  the  soil  becomes  hard  and  cracks, 
and  the  plants  perish.  We  may,  therefore,  be  guided  to  suc- 
cessful results  by  observation.  What  frequently  takes  place 
naturally,  or  accidentally,  in  consequence  of  a  failure  in  our 
own  experiments,  will  furnish  safe  ground  to  go  upon.  We 
cannot  insist  too  strongly  in  this  climate  upon  the  use  of  much 
seed,  that  the  soil  may  be  covered  with  vegetation ;  and 
hence,  protect  it  by  preserving  the  surface  in  a  cool  con- 
dition. Moisture  is  always  condensed  from  the  atmosphere 
upon  such  a  surface  during  the  night,  and  evaporation  is  in  a 
great  measure  prevented  by  day,  if  a  thick  coating  of  veg- 
etation has  grown  upon  it.  We  should  not  forget  in  this  con- 
nexion that  early  planting  is  one  of  the  means  by  which  we 
may  secure  a  crop  from  the  effects  of  a  drouth. 

One  of  the  best  materials  for  grass  lands  is  ashes,  either 
leached  or  unleached.  The  latter  will,  of  course,  contain  less 
potash,  but  even  then,  they  are  highly  valuable.  In  the  ab- 
sence of  ashes,  fine  marl  sown  broadcast,  or  if  accessible, 
strewed  freely  upon  the  surface,  will  effect  important  results, 
either  ash  or  marl  bring  in  clover,  without  sowing  seed. 
Plaster  produces  the  same  effects.  Where  a  system  of  hus- 
bandry is  pursued  which  furnishes  barn-yard  manure,  it  sup- 
plies an  admirable  basis  for  composting.  Very  few  planta- 
tions in  the  eastern  section  of  the  State,  which  do  not  furnish 
muck  or  peat.  With  one  load  of  barn-yard  manure  and  two 


NORTH-CAROLINA   GEOLOGICAL   8UEVEY.  191 

loads  of  muck  or  peat,  three  loads  of  an  excellent  fertilizer 
may  be  made.  These  materials  should  be  well  incorporated 
and  receive  from  time  to  time  all  the  refuse  matter  of  the 
house,  yard  and  garden,  or  anything  which  will  ferment 
under  the  influence  of  the  necessary  conditions.  Wool,  hair, 
refuse  animal  matter  of  all  kinds,  become  of  the  utmost  im- 
portance in  composting.  One  important  addition  should  not 
be  neglected  ;  that  is  plaster  of  paris.  In  the  absence  of 
that,  dirt  sprinkled  with  copperas  water,  which  is  not  expen- 
sive, will  make  an  absorbent  of  the  gasses.  That  dirt  alone, 
or  earth,  has  strong  absorbent  powers,  we  have  sufficient 
evidence  in  the  fact,  that  very  little  odor  escapes  from  the 
carcass  of  a  decaying  animal  body  when  it  is  perfectly  cov- 
ered. But  additional  earth  should  be  added  from  time  to 
time,  as  the  first  becomes  saturated  with  the  effluvia.  The 
matter  which  escapes  under  these  circumstances,  is  ammonia, 
which  is  one  of  the  active  principles  paid  for  in  guano,  which 
makes  the  difference  in  the  price  of  Peruvian  and  Mexican 
guano.  Compost  heaps  require  a  small  proportion  of  lime, 
but  wherever  animal  matters  or  excrements  are  concerned, 
there  should  be  a  large  intermixture  of  muck  or  peat.  No 
good  farmer  adds  lime  to  his  barn-yard  manures ;  it  may  be 
done  only  where  nndecomposed  vegetable  matter  is  ready  to 
absorb  the  disengaged  ammonia. 

SUMMER  SOILING. 

One  of  the  most  important  measures  for  carrying  on  a  suc- 
cessful and  profitable  scheme  of  husbandry,  is  to  incorporate 
with  the  general  plan  or  system,  that  of  soiling  cattle.  Its  value 
has  been  fully  established,  both  in  this  country  and  Europe. 
Apparently,  it  is  objectionable  from  the  amount  of  labor  it 
requires ;  but  this  objection  vanishes  when  it  is  put  in  prac- 
tice, and  becomes  the  every-day  business  of  those  appointed 
to  superintend  it.  Cattle,  when  soiled,  must  be  confined  to  a 
yard,  at  least,  and  fed  on  mown  grass,  lucerne,  clover,  or  corn 
sown  broadcast.  A  large  stock  may  be  kept  on  five  acres  of 
ground,  or,  it  may  be  made  to  yield  that  of  thirty  acres  of 
pasture  lands.  After  being  fed  in  stables,  they  may  be  driven 


192  NORTH-CAROLINA    GEOLOGICAL    SURVEY. 

to  a  pasture  for  the  purpose  of  exercise,  and  returned  again 
at  night,  and  fed  on  fresh  mown  fodder  in  the  morning. 
Soiling  is  no  doubt  as  well  adapted  to  the  South  as  in  the 
North.  By  this  system,  cattle  are  protected  from  a  burning 
sun  during  the  day, — a  protection  which  is  almost  as  impor- 
tant as  protecting  them  from  the  cold.  Most  farmers  appear 
to  forget  that  good  stock  are  like  the  cereals,  which  have 
been  brought  to  their  best  and  improved  condition  -by  ar- 
tificial means,  and  the  moment  the  efforts  to  maintain  them 
in  this  highly  improved  state  are  suspended,  they  begin  to 
deteriorate.  Cattle  can  no  more  be  kept  in  a  good  and  pros- 
perous state  than  the  cereals,  which  if  the  condition  of  the 
soil  is  neglected,  fail  to  produce  remunerating  crops.  But 
furnish  them  with  food  and  place  them  in  comfortable  cir- 
cumstances, and  profits  are  sure  to  be  returned. 

Soiling  is  adapted  to  the  circumstances  attending  the  culti* 
vation  of  a  few  or  many  acres.  The  system  consists  in  culti- 
vating those  grasses  which  come  to  maturity  in  succession, 
and  it  is  desirable  to  be  able  to  vary  the  kinds  of  green  food 
every  few  days,  though  it  is  not  necessary  to  the  success  of 
the  system. 

In  connexion  with  summer  feed,  it  is  important  also  to  have 
an  eye  to  the  winter  support  of  the  same  herd.  For  this  pur- 
pose root  crops  become  an  important  part  of  the  system  of 
soiling.  When,  for  example,  the  patches  of  corn,  oats  or  rye 
are  cut  up,  the  sugar  beet  or  turnip  may  be  sown  for  winter 
feed.  To  these,  then,  should  be  added  carrots  and  sugar  parsnips, 
The  object  of  root  culture  for  stock  is  to  supply  a  variety  of 
nutriment  for  horses  and  cattle,  which,  if  fed  with  them  once 
a  day,  may  become  much  more  thrifty  and  healthy  than  if  fed 
only  upon  dry  fodder.  For  a  Southern  grass,  the  orchard  grass 
should  take  the  place  of  Timothy.  This,  with  the  June  grass, 
red  top,  and  herds  grass,  and  a  few  others  already  described, 
will  give  all  the  winter  hay  which  may  be  required.  The 
practice  of  pulling  fodder  from  the  Indian  corn  is  much  more 
laborious  and  attended  with  more  trouble  than  that  of  mow- 
ing grass  for  hay.  An  acre  of  sugar  beet  will  produce  a 
thousand  bushels,  and  an  acre  of  carrots  over  six  hundred,  and 


NORTH-CAROLINA   GEOLOGICAL   SURVEY.  193 

the  sugar  parsnips  yields  about  eight  hundred  bushels  to  the 
acre. 

One  of  the  incidental  advantages  of  soiling  is  the  production 
of  a  large  amount  of  valuable  manure  which  may  be  saved 
under  cover,  and  to  which  may  be'  added  the  refuse  of  the 
kitchen  and  garden,  whereby  its  quantity  may  be  indefinitely 
increased. 

In  the  foregoing  observations  upon  soiling,  I  have  been  dis- 
posed merely  to  allude  to  the  subject,  believing  that  those 
planters  who  wish  to  keep  good  stock,  either  of  horses  or  cat- 
tle, will  be  inclined  to  try  this  as  a  part  of  their  system  of 
husbandry ;  a  system,  wrhich,  if  carried  out,  will  not  fail  to 
give  them  a  good  stock  of  cattle  and  cows  as  well  as  horses, 
all  of  which  may  be  kept  cheaper  and  better  than  in  the  mode 
now  pursued  in  this  State. 


CHAPTER  XY. 

i 

PALAEONTOLOGY. 

Fossils  of  the  Green  Sand  and  Tertiary — Mammals — Horsfe — Hog — Masto- 
don and  Elephant — Deer — Whales,  or  Cetaceans, 

The  distinguishing  features  or  characteristics  of  any  age  or 
epoch,  can  be  known  only  from  the  history  of  the  men 
who  were  then  living.  The  characteristics  of  the  age  when 
the  Romans  were  gaining  an  ascendancy  in  the  world,  can 
only  be  known  from  the  individual  or  collective  memories  of 
Roman  citizens.  A  history  competent  to  give  us  a  knowledge 
of  those  times,  would  blend  together  the  personal  appearance 
of  men,  their  habits,  dress,  food,  etc.,  from  which  we  should 
also  obtain  facts  or  inferences  respecting  the  country,  its  ani- 
mals and  plants,  its  climate,  topography  and  grand  divisions. 
So  of  Greece,  Egypt  and  Palestine.  The  memoiies  of  the  ac- 
tions of  these  nations  in  their  generations,  would  furnish  us  the 
U 


NOKTH -CAROLINA   GEOLOGICAL    SURVEY. 

leading  facts  respecting  the  characteristics  of  the  period  in 
which  the  respective  nations  lived. 

So,  also,  the  characteristics  of  the  fossils  furnish  at  least  a 
clue  to  the  features  of  the  epoch  during  which  they  lived. 
To  determine  these  features,  demands  an  intimate  knowledge 
of  the  present ;  for,  we  are  under  the  necessity  of  comparing 
the  past  with  the  present.  The  present  is  the  standard,  and 
no  comparison  can  be  made  of  any  value  which  neglects  the 
present.  We  find  in  the  present  certain  structures  and  forms 
which  we  know  have  certain  relations  to  climate,  or  to  the 
conditions  in  which  they  exist.  If,  then,  similar  structures  or 
forms  are  found  attached  to  an  extinct  being  of  any  epoch,  it 
is  a  fair  inference  that  that  structure  or  form  bore  a  similar 
relation  to  the  external  conditions  which  surrounded  it.  Its 
full  description,  then,  would  be  a  memoir  of  the  animal,  its 
habits  would  be  indicated,  its  relation  to  surrounding  circum- 
stances would  be  known  ;  many  inferences  would  follow  from 
each, — some  would  bear  only  upon  its  instincts,  its  food,  its 
means  of  defence  from  the  medium  in  which  it  lived,  etc. 

If,  for  example,  an  oval  shaped  bag  filled  with  coloring 
matter,  in  connection  with  a  fossil  known  as  the  Belemnite,  it 
would  be  inferred  that  this  bag  contained  a  fluid  designed  to 
conceal  it  from  its  enemies ;  that  it  would  deeply  discolor  the 
water  into  which  it  was  cast,  and  thereby,  under  its  cloud  of 
dye-stuff,  make  its  escape.  Such  a  phenomenon  is  familiar 
now  to  the  sailor.  The  cuttle-fish  is  thus  supplied  with  dye- 
stuff,  and  he  employs  it  for  escaping  from  a  pursuing  enemy ; 
and  as  this  is  so,  so  it  is  inferred,  the  animal  did  which  was 
supplied  wTith  a  similar  apparatus  in  the  period  of  the  Lias  and 
Chalk. 

We  might  go  on  and  note  hundreds  of  analogous  examples, 
but  one  must  suffice.  This  view  is  borne  out  by  one  great 
and  leading  fact,  that  all  extinct  animals  are  constructed  upon 
one  of  the  four  leading  types  which  now  prevail.  Of  the  mil- 
lions of  individual  fossils  which  have  been  seen,  not  one  is 
known  which  does  not  belong  to,  and  may  be  referred  with 
certainty,  to  one  of  the  great  leading  types  of  the  present.  It 
is*  the  plan  then,  which  really  tells  all  this,  or  makes  it  possi- 


NORTH-CAROLINA   GEOLOGICAL   SURVEY.  195 

ble  to  compare  and  infer  with  certainty.  Observation  is  the 
way,  but  the  plan  of  creation  makes  it  possible  to  deduce  a 
connected  history  of  the  past  from  the  dead  races,  and  thereby 
see  at  a  glance  how  any  former  epoch  differed  from  the  pres- 
ent, or  from  those  ancient  ones  with  which  it  was  more  inti- 
mately connected. 

My  object,  however,  is  not  so  much  to  direct  the  student  in 
this  chain  of  reasoning,  or  so  to  apply  knowledge  as  to  make 
him  acquainted  with  the  external  forms  of  the  fossils  of 
the  marl  beds.  The  figures  and  descriptions  will  enable  him  to 
know  the  objects  from  their  forms,  and  thereby  to  distinguish 
the  marl  beds  which  contain  them  from  each  other.  It  is, 
therefore,  a  practical  subject,  and  may  be  studied  as  such. 
But  the  knowledge  thus  acquired  prepares  the  way  for  further 
advances  in  science. 

The  fossils  described  in  this  part  of  the  Report,  belong  to 
four  or  five  periods,  inasmuch  as  some  of  them  are  found  in 
two  or  more  successive  ones.  These  periods  have  been  dis- 
tinguished by  the  following  names  which  are  expressive  of 
certain  ideas.  Thus,  the  oldest  is  the  cretaceous  or  chalk  for- 
mation. It  is,  however,  only  a  small  part  of  it,  and  that  part 
is  the  inferior  or  oldest  part  of  the  cretaceous  system.  This 
part  is  widely  known  as  the  Green  Sand,  and  has  been  em- 
ployed extensively  as  a  fertilizer.  The  2d,  in  the  ascending 
order,  is  the  Eocene,  which  means  the  dawn  of  the  present,  as 
a  few  species  survive,  which  were  created  in  this  epoch  or  pe- 
riod. Only  about  four  per  cent,  however,  have  lived  on 
through  all  the  vicissitudes  of  the  times.  The  third,  is  the 
Miocene.  Of  the  animals  created  during  this  period,  more 
than  half  have  perished,  and  we  know  them  only  through 
their  remains.  The  fourth  is  the  Pliocene,  the  animals  of 
which  less  than  half  have  perished.  The  fifth,  the  post-Plio- 
cene, is  known  by  its  fossils  being  similar  to  those  which  now 
live,  excepting  five  or  six  per  cent.  Hence,  it  may  happen 
that  one  of  the  four  species  of  animals  which  survive,  and 
which  was  created  in  the  Eocene  period,  may  be  found  in  all 
the  succeeding  beds,  but  it  is  evident  it  will  be  associated  in 


196  NORTH-CAROLINA   GEOLOGICAL   SURVEY. 

each  case  with  races  or  species  quite  different  from  those 
among  whom  it  was  first  connected  or  who  were  its  cotern- 
poraries. 

The  cause  of  the  extinction  of  so  many  species,  is  a  mystery. 
The  fact  is  well  established,  but  it  is  only  in  certain  cases  that 
we  can  account  for  their  disappearance.  It  appears  to  have 
been  sometimes  due  to  a  sudden  catastrophe,  the  ejection  of 
mud,  or  poisonous  matter  into  the  medium  in  which  they  live. 
This  happens  now,  and  probably  has  happened  before,  but  in 
a  majority  of  instances,  it  is  impossible  to  perceive  any  exter- 
nal cause  which  destroyed  them ;  and  hence,  we  are  left  to 
speculate  on  probabilities,  without  being  able  to  arrive  at  sat- 
isfactory conclusions. 

MAMMALIA. — EQUTJS    CABALLUS. 

There  is  scarcely  a  question  so  interesting  to  the  naturalist 
and  historian  as  that  which  relates  to  fossil  remains  of  the 
horse.  The  testimony  of  historians  is,  that  the  horse  was  not 
living  upon  this  continent  at  the  time  of  its  discovery  by 
Columbus.  The  testimony  of  the  naturalist  is,  that  the  horse 
lived  upon  this  continent  at  a  period  prior  to  its  discovery, 
its  remains  having  been  found  first  in  the  miocene,  and  lastly 
in  the  pliocene,  in  which  period  it  may  have  become  extinct. 
Its  earliest  appearance  is  in  the  former;  and  it  appears  from 
the  discovery  of  Prof.  Holmes,  of  Charleston,  S.  C.,  that  its 
remains  are  not  uncommon  in  the  latter. 

FIG.  is.  Figure  18  represents  the  crown  of  the 

third  or  fourth  molar  of  the  left  side  of  the 
upper  jaw.  It  has  complicated  enamel 
plates,  or  columns,  and  is  somewhat  worn, 
but  by  no  means  an  old  tooth,  as  its  roots 
are  undeveloped.  It  is  two  inches  long 
and  an  inch  thick.  It  is  undistinguishable 
from  the  corresponding  tooth  of  the  recent 
domestic  horse.  It  is  a  deep  brown  color,  and  looks  like  a 
fossil. 

Figure  19  represents  the  crown  of  a  tooth  of  the  third  or 
fourth  molar,  probably  the  third,  of  the  left  upper  side.  It 


NORTH-CAROLINA   GEOLOGICAL   SURVEY. 


197 


has  not  been  worn.     It  resembles  a  recent  tooth,  as  it  is 
FIG.  19.  whitish,  and  only  stained   brown  on  one 

side.  The  enamel  plates,  it  will  be  per- 
ceived differ  from  the  preceding,  and  they 
differ  also  from  those  of  the  correspond- 
ing tooth  of  the  domestic  horse.  This 
difference,  however,  may  arise  from  its 
unworn  condition,  as  the  enamel  plates 
differ  somewhat  in  configuration  as  they  wear  down.  This 
tooth  is  three  inches  long  and  one  thick. 

FIG.  20.  This  figure  (20)  represents  the  back  molar  of  the 
left  side  of  the  lower  jaw  of  the  horse.  It  differs  only 
slightly  from  the  corresponding  tooth  of  the  do- 
mestic horse.  It  is  worn,  but  belonged  to  a  young 
individual,  and  its  roots  are  undeveloped.  It  is 
three  inches  long,  one-half  an  inch  thick,  and  one 
and  a  quarter  wide. 

Figure  21  represents  one  of  the  incisors  of  the 

FIG.  21. 


horse ;  a,  front  side ;  b,  inner  side ;  c,  lateral  view.  This  scarcely 
differs  from  the  corresponding  incisors  of  the  domestic  horse. 
The  foregoing  teeth  are  from  the  miocene  of  North-Carolina, 
and  were  discovered  at  an  early  period  of  the  survey.  No.  18 
was  found  in  a  bed  at  Elizabethtown,  Bladen  county,  and  was 
accompanied  with  a  tooth  from  the  lower  jaw.  No.  19  and 
20  are  teeth  washed  up  on  the  beach  at  Plymouth,  N.  0.,  and 


198  NORTH-CAROLINA   GEOLOGICAL   SURVEY. 

No.  21  from  the  miocene  of  Pitt  county.  I  found,  also, 
molars,  in  Pitt  county.  They  occur  in  a  sandy  bed,  which 
may  be  ten  or  twelve  feet  above  the  shell  marl.  Although 
there  is  a  close  correspondence  between  the  fossil  teeth  above 
described  and  those  of  the  domestic  horse,  which  was  intro- 
duced into  this  country  since  its  discovery,  still,  it  is  probable 
that  it  is  a  different  species.  If  it  is  maintained  that  the 
fossil  and  introduced  species  are  identical  and  the  same,  it 
follows  that  the  same  species  was  created  about  the  same 
epoch,  in  two  very  different  quarters  of  the  globe,  viz :  Asia 
and  America,  and  in  climates  which  differed  materially  from 
each  other.  Farther  discoveries  must  be  made  before  this 
interesting  question  can  be  satisfactorily  settled. 

SITS   SCROFA. HOG. (Fig.   22.) 

The  only  relic  of  the  hog  which  has  been  ob- 
tained during  the  survey,  is  the  last  inferior 
molar,  scarcely  differing  from  its  fellow  in  the 
domestic  hog.  I  obtained  it  at  Washington, 
Beaufort  county,  from  the  miocene.  It  is  brown, 
and  is  partially  mineralized  by  sulphuret  of  iron. 
It  has  the  same  claim  to  genuineness  as  a  fossil, 
as  the  teeth  of  the  horse  already  described. 
22.)  The  hog  was  introduced  into  this  country  at 
the  time  of  its  settlement,  but  as  in  the  case  of  the  horse,  it 
was  peopled  by  this  interesting  animal  a  long  time  prior  to 
its  discovery.  It  also  became  extinct,  and  at  its  settlement 
was  supplied  again  from  a  foreign  country. 

PROBOSCEDIANS. MASTODON   GIGANTEUS. (Fig.    23.) 

The  bones  of  this  large  pachyderm  are  not  uncommon  in 
the  miocene  marl  of  North-Carolina. 

Fragments  of  ribs  and  bones  of  the  extremities  are  the 
most  common.  The  figure  of  the  superior  part  of  the  crown 
in  the  margin  was  taken  from  a  tooth  found  in  Halifax  coun- 
ty. Its  enamel  is  jet  black  and  highly  polished.  It  is  the 
first  or  small  molar  of  the  right  side  of  the  under  jaw.  It  is 
an  old  tooth  with  the  lubercles  worn  down,  and  was  probably 


N 

NORTH-CAROLINA   GEOLOGICAL   SURVEY.  199 

FIG.  23. 


lost  or  shed  while  the  animal  was  living.  The  figure  is  de- 
signed to  show  the  arrangement  of  the  enamel  plates. 

Bones  of  this  immense  quadruped  have  been  found  at 
numerous  places.  A  large  number  were  found  in  a  marl  pit 
near  Goldsboro',  and  a  large  back  molar  in  another  marl  pit 
in  Nash.  These  bones  are  usually  broken,  and  the  pieces 
are  rarely  more  than  from  three  to  six  inches  long.  A  cunei- 
form bone  of  the  foot  was  found  in  a  marl  bed  upon  the 
Cape  Fear.  From  the  number  of  bones  which  have  been 
found  it  is  evident  this  large  species  of  land  quadruped,  the 
largest  known,  must  have  been  very  numerous  .at  one  time. 
Its  bones  are  associated  with  fossils,  many  of  which  are  now 
extinct,  and  some  or  even  many  still  survive.  The  oldest  de- 
posit in  which  the  bones  of  the  mastodon  are  known  to  occur 
is  probably  the  miocene.  They  continued  to  occur  in  the 
subsequent  formations  until  the  latest,  which  just  precede 
the  advent  of  man ;  and,  indeed,  it  is  not  at  all  improbable 
that  man  witnessed  the  final  extinction  of  the  race.  The 
long  bones  which  I  have  examined  always  contain  animal 
matter,  an  evidence  of  their  recent  death. 

The  elephant  was  also  a  cotemporary  with  the  mastodon. 
No  teeth,  however,  have  yet  been  found  in  North-Carolina 
which  may  have  enabled  me  to  identify  its  remains.  But  to 
those  who  have  marl  beds  to  identify  its  remains,  a  tooth  (Fig. 
24)  of  this  interesting  animal  is  given  in  the  margin.  It  is  a  re- 
duced figure  of  one  found  in  the  superficial  deposits  of  New 


200 


NORTH-CAROLINA   GEOLOGICAL   SURVEY. 


1  ork.  A  tooth  belonging  to  the  elephant  was  taken  from  the 
beach  upon  Seneca  lake  New  York,  and  portions  of  a  skele- 
ton were  found  near  the  surface  in  Monroe  county.  All 
these  bones  contain  also  animal  matter,  and  they  are  usually 
associated  with  moluscous  animals  which  are  living  at  the 
present  time. 

FIG.  24. 


It  is  probable  the  mastodon  lived  in  a  period  prior  to  that 
of  the  elephant,  but  it  appears  that  both  became  extinct  at 
or  about  the  same  time. 

That  the  mastodon  and  elephant  roamed  in  herds  over  a 
large  part  of  this  continent,  seems  to  be  indicated  by  the 
fact  that  their  bones  are  found  from  the  Atlantic  to  the  base 
of  the  Rocky  mountains.  The  bones  of  the  mastodon,  how- 
ever, are  more  numerous  and  more  widely  extended  than 
those  of  the  elephant. 

RUMINANTIA. — CERVTJS    VIRGINIAN  A 

The  discovery  of  the  remains  of  the  0.  Yirginiana  deer,  is 
an  interesting  fact.  It  appears  to  have  been  cotemporary 
with  the  Mastodon  and  Elephant,  which  have  become  extinct. 
So,  also,  it  is  cotemporary  with  the  great  Irish  Elk,  which  has 
become  extinct  in  Europe. 

The  base  of  the  horn  which  I  found  in  the  Miocene  bed 
about  10  miles  above  Elizabeth,  on  the  Cape  Fear,  is  about 
six  inches  long.  In  this  horn,  the  first  branch  goes  off  from 


NORTH-CAROLINA   GEOLOGICAL    SURVEY.  201 

the  axis  nearer  the  head  than  usual,  but  this  occurs  occasion- 
ally in  individuals  of  this  species. 

It  appears  from  this  discovery  that  the  common  red  deer  of 
America  began  its  existence  at  or  about  the  same  period  as 
the  American  horse ;  but  while  the  horse  became  extinct,  the 
deer  has  survived.  In  a  fresh  water  marl  bed,  in  Orange 
county,  in  New  York,  I  found  a  horn  of  an  extinct  deer  which 
was  associated  with  the  remains  of  the  mastodon.  The  deer 
of  the  miocene  marl  survives,  while  a  more  recent  species  has 
become  extinct,  or  such  is  the  evidence  of  facts  as  they  now 
stand. 

REMAINS  .  OF  THE  PORPOISE. 

Several  vertebrse  which  appear  to  have  belonged  to  the 
porpoise,  have  been  obtained  from  the  marl  beds  near  Rocky 
Mount.  They  appear  to  belong  to  a  species  which  differs 
from  the  common  one  of  the  coast.  The  figure  shows  the  end 
of  the  vertebrse  to  which  the  intervertebral  substance  is 
strongly  attached  ;  the  other  extremity  is  smooth.  The  body 
is  encircled  in  part  with  a  deep  channel  or  groove,  which  is 
connected  with  the  holes  which  transmit  the  vessels,  and 
nerves  at  the  base  of  the  spinal  arch. 

In  addition  to  the  foregoing  remains  of  the  order,  cetacea, 
I  may  mention  the  occurrence  of  the  Zeuglodon  cetoides. 
(OwEN,)  a  fossil  of  the  eocene,  which  was  first  found  in  Ala- 
bama, and  described  by  the  late  Dr:  Harlan,  of  Philadelphia. 
The  teeth  are  entirely  unlike  those  of  the  common  cetaceans, 
and  belong  to  a  type  not  very  unlike  those  of  the  seal.  No 
teeth,  however,  have  as  yet  been  discovered  in  this  State. — 
The  remains  of  this  cetacean  consist  of  vertebra  which  were 
obtained  from  Washington,  near  the  line  of  the  Wilmington 
Eail  Eoad. 

One  of  the  largest  candal  vertebrse  of  a  whale,  (fig.  25,)  has 
broad  flat  transverse  processes,  standing  at  right  angles  to  the 
body  of  the  bone,  the  articular  ends  are  unequal,  the  anterior 
being  5-J  and  the  posterior  4f  inches  in  diameter,  and  circular, 
with  a  length  of  6  inches.  Of  this  length  the  base  of  the  trans- 
verse processes  occupies  4  inches,  and  terminate  behind  in  a 
rounded  notch ;  their  length  is  2-J-  inches. 


202 


NORTH-CAROLINA   GEOLOGICAL   SURVEY. 
FIG.  25. 


LOWER   JAW    OF   A  BALAENA   OR   WHALE. 

On  the  Meherrin,  near  Murfreesborough,  I  found  portions 
of  three  lower  jaw-bones  belonging  to  the  genus  Balaena,  to- 
gether with  many  vertebrae,  all  of  which  appear  to  belong  to 
one  species. 

These  jaws  are  imperfect, — the  anterior  part  the  left  lower 
jaw  is  smooth,  gently  covex,  and  curved  on  the  outside,  but 
rather  flat  inside.  The  wide  upper  margin  is  perforated  with 
three  holes  penetrating  the  jaw  in  a  slightly  descending 
course,  and  terminating  anteriorly  in  an  edge  produced  by  a 
champering  of  the  inside  extremity,  and  rounded  from  the 
base  up  to  the  upper  edge,  which  is  grooved  for  six  inches. 
They  are  3J  inches  wide  and  2  inches  thick,  and  nearly 
straight.  All  the  posterior  parts  of  the  jaw  had  been  lost,  and 
only  two  feet  obtained.  It  is  impossible  to  refer  these  frag- 
ments of  jaws  to  either  species  which  furnished  the  ear  bones, 
as  neither  of  these  specimens  were  obtained  at  this  locality. 
But  the  vertebrae  and  jaws  belonged  to  one  species,  and  it  is 


NORTH-CAROLINA   GEOLOGICAL   SURVEY.  203.. 

possible  hereafter  to  determine  to  which  ear-bone  belonged 
to  the  Murfreesborough  species.  It  is  evident  that  neither  of 
these  belonged  to  Prof.  Leidy's  Orycterocetus,  because  this 
belonged  to  a  different  family  of  the  cetaceans. 

OTOLITES,    OR   THE   EAR   BONES    OF   WHALES. 

The  remains  of  the  cetacea  may  be  said  to  be  numerous  in 
the  miocene  of  North-Carolina.  Vertebra  and  ribs  are  more 
commonly  found  than  other  parts  for  the  reason  that  the  in- 
dividual parts  exceed  in  number  the  other  parts  of  the 
skeleton.  The  ear  bones  are  the  least  common.  Of  this  part 
I  have  those  which  I  regard  as  having  belonged  to  at  least 
three  different  species.  I  base  this  conclusion  on  the  estab- 
lished fact  that  these  bones  possess  for  each  species  a  peculiar 
configuration  ;  that  though  the  bone  in  question  has  a  general 
resemblance  in  all  the  species  of  which  the  family  is  com- 
posed, yet  in  the  minute  details  of  construction  and  form, 
each  species  has  its  own,  which  may  be  determined  by  close 
and  careful  comparisons.  Thus,  in  the  true  whales,  the  thick 
posterior  part  is  simple,  while  in  the  cachalot  it  is  bilobed, 
and  that  this  thickened  and  convex  part  in  the  simple  kinds, 
while  it  is  variable  in  form  and  extent  in  the  different  species 
of  the  true  whales,  and  which  is  also  joined  to  certain  other 
differences,  which  may  be  observed  in  the  thin  overarching 
and  expanded  part. 

For  convenience  of  description,  these  bones  may  be  divid- 
ed, longitudinally,  into  two  principal  parts:  1.  The  thick 
involuted  convex  part  which  occupies  the  posterior  segment 
of  the  bone,  and  which  extends  back  to  a  rough  longitudinal 
surface  ;  and,  2d.  The  thinner  and  expanded  part  which  begins 
where  the  former  ends,  and  arches  over  the  first  in  different 
degrees,  forming,  posteriority,  a  convex  surface,  and  interi- 
orly towards  the  first  part  a  concavity  differing  both  in  de- 
gree and  extent  in  different  species.  The  anterior  or  eustach- 
ian  portion  is  formed  wholly  of  the  thinner  expanded  part. 
There  is  in  the  form  of  the  expanded  part  some  resemblance 
to  the  rim  of  the  human  ear. 

The  ear  bones,  in  consequence  of  the  thick  convex  part 


"204:  NORTH-CAKOLINA   GEOLOGICAL   SURVEY. 

being  simple,  are  all  referred  to  the  genus  balaena.  Other 
parts  of  the  skeleton  of  this  genus  have  been  formed,  as  the 
vertebrae,  ribs,  lower  jaw,  &c. 

The  first  of  the  bones  (Fig.  26)  which  I  propose  to  describe 
FIG  26  is  the  largest,  and  resembles  in 

form  the  same  bone  belonging 
to  the  right  whale,  (the  balaena 
mysticetus.) 

In  this  specimen  the  thick  in- 
voluted part  is  thickest  at  its 
extreme  posterior  end,  and  gra- 
dually diminishes  to  within  three 
fourths  of  an  inch  of  the  flatish, 
expanded  or  eustachian  part  of  the  tube. 

Its  surface,  as  it  passes  backward,  and  corresponding  to  the 
span  between  the  lobes  in  the  cachalot,  becomes  slightly  con- 
cave, and  the  whole  surface  to  the  boundary  backwards  and 
forwards  to  the  channel,  which  separates  it  from  the  concave 
expanded  portion,  is  irregularly  wrinkled ;  these  wrinkles  in- 
crease in  strength  to  its  junction,  with  the  latter  part,  where 
the  line  of  division  is  distinctly  defined.  At  the  posterior  part, 
there  is  a  strong  indentation,  somewhat  in  the  form  of  the  letter 
U,  surrounding  the  part  where  the  expanded  part  springs.  The 
thinner  expanded  part  forms  an  arch,  concave  within,  and  quite 
regularly  convex  without;  at  the  extremities  it  forms  expanded 
hooks.  The  concave  surface  widens  from  the  posterior  to  the 
anterior  end,  and  is  widest  just  within  the  margin.  This  bone 
differs  from  the  same  in  the  right  whale,  in  its  convex  portion 
being  lower  and  not  above  the  level  of  the  concave  cavity 
beneath  the  arch ;  and  being,  also,  perfectly  separated  by  a 
change  in  the  appearance  of  the  part,  and  also  by  the  perfect 
smoothness  of  the  concave  surface  of  the  overarching  wall, 
which,  in  this  B  mysticetus,  is  very  rugged. 

Its  length  is  3|  inches,  and  width  2J,  and  belonged  to  a 
large  whale,  though  probably  not  the  largest.  It  is,  however, 
very  bulky.  Cuvier  remarks,  that  the  ear  bones  of  the 
Balaeonoptera  are  very  small  in  proportion  to  the  size  of  the 


NORTH-CAROLINA   GEOLOGICAL   SURVEY.  205 

species ;  so  that  it  does  not  follow  that  where  the  bone  is 
small  the  spieces  must  be  small  also. 

I  propose  the  name  Boblaena  Mysticetoides  for  their  species. 

The  thick,  the  posterior  end,  is  nearly  equally  bisected  by 

the  thin  expanded  part,  and  around  it  there  is  a  deep  sinuous 

indentation   which,  on   the  inside,  is   continuous  with   the 

channel  between  the  thick  and  thin  parts. 

FIG.  27.  The  otololite,  next  in  size  to 

the  B  misticetoides,  differs  much 
from  it  in  form  and  proportion 
of  parts.  The  thick  convex  part 
is  well  defined,  but  rough,  short 
and  prominent.  It  rises  higher 
than  the  base  of  the  thin  invo- 
luted part  to  which  it  slopes  all  round.  It  is  marked  with 
two  or  three  strong  folds,  one  of  which  is  at  or  near  its  termi- 
nation forward,  and  another  beneath,  which  gives  a  slight 
emargination  to  the  bone.  It  is  separated  from  the  anterior  end 
by  a  flattened  plane  about  half  an  inch  wide,  where  their 
expanded  part  turns  and  forms  a  rather  open  hook,  unlike  that 
of  the  former,  which  is  bent  much  more  inwards.  The  pos- 
terior end  is  somewhat  obliquely  truncate,  and  at  the  root  of 
the  thin  part  there  is  a  rough  indentation  disconnected  with 
the  wide  channel  within.  The  anterior  border  of  the  thin 
part  forms  an  arch  much  less  extended  than  the  former,  and 
the  posterior  and  basal  part  is  flattened  and  angular.  Length 
3J  inches ;  widest  part  1-J-. 

Another  specimen  measuring  four  inches  long  preserves 
the  essential  characters  of  the  foregoing.     It  is  very  rugose 
around  the  thick  convex  part,  and  the  middle  fold  creates  a 
slight  twolobed  character  to  the  interior  part  and  its  base. 
The  smallest  (Fig.  28)  has  a  well-defined  convex  part,  which 
FIG.  28.  ig  smooth  though  somewhat  wrinkled 

but  rough  within,  and  the  border  rises 
almost  immediately  from  it,  especial- 
ly posteriorly.  The  space  between 
the  border  and  convex  part  widens 
anteriorly  where  it  is  only  gently 


206  NORTH-CAROLINA   GEOLOGICAL   SURVEY. 

curved,  scarcely  forming  a  hook.  Behind  the  convex  part  it  is 
very  regular,  but  the  beginning  of  the  thinner  expanded  part 
is  formed  by  a  rounded  ridge,  which  may  be  traced  from  one 
extremity  to  the  other.  It  is  far  less  angular,  and  more  regular 
than  the  preceding.  It  is  2f  inches  long ;  greatest  width  1-J- 
inches. 

This  ototite  is  one  of  the  most  common  in  the  miocene 
beds.  Unfortunately,  in  all  these  specimens,  the  thin  ex- 
panded over-arching  part  is  broken  off,  but  it  is  evident  that 
in  this  case  this  part  was  very  limited. 

The  two  smallest  are  perforated  by  boring  moluscks,  a  fact 
which  shows  that  instinct  is  sometimes  at  fault. 

It  is  probably  impossible  in  the  present  state  of  our  knowl- 
edge of  the  anatomy  of  those  extinct  whales,  to  refer  them 
to  the  species  to  which  they  belonged.  That  the  foregoing 
ear-bones  I  have  described  belonged  to  different  species  of 
the  whale,  there  can  be  no  doubt. 

Few  extinct  species  of  balaena  are  known  to  belong  to  the 
miocene  period  besides  the  orycterocetus  of  Leidy. 

SUMMARY 

Of  the  characteristics  of  the  three  foregoing  species,  derived 
from  a  comparison  with  each  other,  and  with  the  three 
which  have  been  described,  l)y  PROF.  OWEN. 

The  B.  mysticetoides  differs  from  B.  affinis  Owen,  in  the 
much  greater  extent  of  the  overarching  wall  and  the  well  de- 
lined  limits,  and  greater  prominence  of  the  involuted  part ; — 
this  part  also  bears  a  much  greater  proportion  to  the  whole  of 
the  organ  than  it  does  in  the  affinis. 

The  B.  deh'nita  Owen  is  very  strikingly  truncated  at  its 
posterior  end,  and  has  also  its  thick  involuted  part  much  less 
in  proportion  than  in  the  B.  mysticetoides,  and  its  thin  over- 
arching border  is  also  much  less  in  extent. 

It  differs  from  the  B.  gibbosa,  Owen,  in  most  of  the  charac- 
ters just  stated ;  particularly  the  extent  of  the  overarching 
wall,  its  thick  convex  part  is  much  less' prominent ;  but  it  re- 


NORTH-CAROLINA   GEOLOGICAL   SURVEY.  207 

sembles  the  B.  gibbosa  somewhat  in  its  configuration  at  the 
posterior  end,  where  the  riin  is  continued  around  it,  as  it  were, 
but  in  the  gibbosa,  it  rises  from  near  the  base,  while  in  the 
mysticetvides  it  rises  higher  and  is  surrounded  by  deep  sinu- 
ous indentations.  It  resembles  also  the  B.  emarginata  in  the 
existence  of  a  concavity  on  the  inferior  border  of  the  thick 
convex  part,  but  is  much  less ;  the  overarching  wall  exceeds 
very  much  in  extent  that  of  the  emarginata. 

The  figure  27  differs  from  the  affinis  in  its  prominent  and 
distinctly  defined  convex  involution.  It  resembles  the  B.  de- 
finita  somewhat,  in  its  posterior  truncation  ;  but  the  involuted 
part  is  more  prominent,  and  has  a  strong  ridge  or  prominence 
on  the  border  near  its  slope  to  the  concavity  ;  but  it  resem- 
bles still  more  closely  the  B.  gibbosa,  in  the  form  of  the  con- 
vex part,  but  the  thinner  overarching  wall  is  more  extensive 
and  broader  at  the  eustachian  termination,  and  the  shape  of 
the  posterior  end  differs  from  it  materially,  particularly  in  the 
strong  angle  of  the  extreme  of  the  overarching  wall. 

It  differs  from  the  B.  emarginata,  in  having  a  prominence 
at  the  base  of  the  involuted  thick  part  instead  of  an  emargi- 
nation. 

The  figure  28  differs  from  the  B.  affinis  in  its  prominent  in- 
voluted part,  and  distinct  form  or  separation  from  the  concave 
overarching  part ;  from  the  B.  definita  by  its  prolonged  pos- 
terior part,  in  which  respect  it  differs  also  from  the  gibbosa 
and  from  emarginata  by  its  absence  of  this  particular  char- 
acter, and  by  the  presence  of  strong  sugar  upon  the  part  next 
the  concavity. 

CHARACTERISTICS    OF     THE    EAR-BONE    OF     THE    COMMON    WHALE    OF 

THE     COAST. 

The  ear-bone  of  the  Balena  Mystictus,  the  common  whale 
of  the  coast,  in  my  possession,  measured,  rather  diagonally 
over  the  thick  convoluted  part,  is  5J  inches  long;-  the  great- 
est thickness  is  3  inches  and  3  tenths  ;  the  depth  or  height  of 
the  convoluted  part  is  3  inches ;  greatest  height  measured  to 
the  top  of  the  thin  convolution  4  inches  and  4  tenths.  The 


208  NORTH-CAROLINA   GEOLOGICAL   SURVEY. 

thin  involuted  expansive  is  arched  so  as  to  have  a  distance  of 
only  half  an  inch  from  the  thick  involuted  part.  This  may  be 
divided  into  three  principal  lobes;  two  of  them  make  up  two- 
thirds  of  thin  part,  and  these  are  divided  externally  by  a  deep 
sulcus,  and  internally  by  a  thick  rounded  ridge  which  extends 
nearly  to  the  base  ;  the  lobe  of  the  thickest  end  is  short.  A 
deep  sulcated  cavity  is  formed  by  the  thick  and  thin  involu- 
ted parts  of  the  bone.  This  cavity  is  3  inches  and  six-tenths 
long  and  2  inches  and  'one-tenth,  and  the  height  nearly  3 
inches. 

An  ear  bone  having  the  form  and  proportions  of  the  Balaena 
Mysticetus,  in  many  particulars,  I  have  obtained  from  Craven 
county.  The  most  important  difference  is  in  the  height  of 
the  thick  involuted  part,  the  thin  expanded  part  is  broken  off 
but  there  are  so  many  points  of  resemblance,  that  it  is  highly 
probable  it  belonged  to  this  species  of  whale.  The  fossil  ear- 
bone  is  smaller.  Its  greatest  length  is  only  4  inches  and  2 
tenths,  and  the  height  of  the  thick  involuted  part  is  only  2 
inches  and  2  tenths.  Still,  it  is  not  at  all  improbable  that  we 
may  regard  it  as  having  belonged  to  the  young  of  the  B.  mys- 
ticetus,  and  if  so  this  species  commenced  its  existence  in  the 
Miocene  period.  This  conclusion  is  founded  upon  the  proba- 
bility, that  this  ear-bone  and  certain  thick  heavy  ribs  of  a  whale, 
often  found  in  the  miocene  deposits,  belonged  to  this  species. 
It  is  probable,  too,  that  ear-bones  vary  somewhat  in  form  and 
thickness  in  the  same  species ;  this  is  certainly  true  in  the 
cose  of  the  ear-bone  of  fishes,  of  which  I  have  many  speci- 
mens, among  which  there  are  several  varieties  of  form  and  size. 
Other  forms  of  cetacean  ear-bones  occur  abundantly  in  the 
miocene  of  Tar  River.  Figure  28  belongs  to  one  of  the  rarer 
forms  of  ear-bones.  It  has  a  distinct  in- 
voluted portion.  It  is  figured  of  the  natu- 
ral size. 

Figure  29  is  another  form  of  ear  bone 
which  is  the  most  common  of  all,  except 
the  following.  It  has  no  distinct  invo- 


NORTH-CAROLINA    GEOLOGICAL    SURVEY. 


209 


the  other. 


29.  voluted  part,  though  it  is  thickened  at 

one  end  of  it.  It  is  more  or  less  wrink- 
led transversely.  In  other  respects  it 
is  rather  discoidal. 

Figure  30,  it  differs  in  form  from  all 
the   proceeding.      It  is   conical,   and 
acute  at  one  extremity  and  obtuse  at 
From  the  obtuse  extremity,  it  sends  off  a  short 
process  at  right  angles,  and  is  probably  the  point 
by  which  it  is  attached  to  the  interior  of  tne  tym- 
panic cavity. 

But  one  of  the  most  extraordinary  of  the  ear- 
bones  of  this  formation,  is  represented  by  figure 
3l.  It  consists  of  two  parts,  a  short  obtuse  conical 
portion,  and  a  long  process  extending  at  right  an- 
gles from  it.  It  is  over  seven  inches  long.  The 
process  referred  to  is  four,  measured  from  the 
base  of  the  heavy  conical  part,  and  it  extends  half  way  across 

FIG.  31. 


FIG.  30. 


it,  so  that  its  whole  length  is  about  5-J-  inches.  The  height  of 
the  conical  part  is  3^  inches.  This  is  flattened,  and  its  greater 
circumference  is  8  inches.  The  arm  or  process  is  irregularly 
triangular,  being  hollowed  out  on  two  sides  and  flattened  on 
the  other.  The  whole  form,  however,  is  difficult  to  represent 
by  means  of  a  single  figure.  The  figure  is  one-half  the  size 
of  the  original. 
15 


210 


HORTH-CAROLINA   GEOLOGICAL   SURVEY, 


ORYCTKROCETUS    QUADRATIDEN6. LEIDY.       PROC.    ACAD.     NAT.     SCL 

vii,  378. 

FIG.  32. 

A  single  tooth  belong- 
ing to  this  cetacean  was 
found  in  Pitt  county  by 
Thos.  Sparrow, -Esq.,  to 
whom  I  am  indebted  for 
an  opportunity  lor  de- 
scribing this  interesting 

o  C? 

relic. 

The  tooth  is  remark- 
ably curved  for  a  ceta- 
cean. It  is  rather  rough, 
and  is  somewhat  quad- 
rate or  angular.  This 
character,  according  to 
Prof.  Leidy,  is  not  con- 
stant. Its  transverse 
section  is  rather  ovate, 
with  the  anterior  part 
flattened.  It  was  point^ 
ed,  but  by  exposure  the 
apex  is  injured.  Its 
base  has  a  short  conical 
pulp  cavity,  less  than 
one  inch  in  depth.  Its 
surface  is  marked  by 
longitudinal  cracks. — 
The  tooth  belongs  to  the 
right  lower  jaw,  and  is 
drawn  the  natural  size. 
It  is  supposed  to  be- 
long to  the  miocene,  but  the  locality  contains  a  few  small  fos- 
sils, derived  from  the  eocene,  and  hence  this  may  be  of  that 
age. 


NORTH-CABOLINA   GEOLOGICAL   SURVEY. 


211 


ORYCTEROCETU8    CORNUTIDENS. LEIDY. 


FIG.  33. 


The  genus  Orycterocetus  was 
originally  proposed  on  the  frag- 
ment of  a  jaw,  and  several  teeth 
from  the  miocene  deposit  of  Vir- 
ginia. In  my  collection  I  have 
a  tooth  like  those  just  mentioned, 
except  that  it  is  not  quadrate, 
which  it  is  suspected,  however,  to 
be  an  unimportant  character. — 
The  specimen  was  discovered  in 
the  miocene  deposit  of  North-Car- 
olina. It  is  remarkable  for  its  re- 
semblance in  form  to  a  small  ox- 
horn,  being  elongated,  conical 
and  curved.  The  base  is  excava- 
ted as  in  the  teeth  of  the  sperma- 
ceti whale,  to  which  the  extinct 
cetacean  was  probably  allied.  In 
structure,  the  tooth  appears  to  be 
wholly  composed  of  dentine.  The 
length  of  the  specimen  in  the 
curve  is  4J  inches,  but  it  appears 
when  entire,  to  have  been  half  an 
inch  longer.  The  section  of  the 
base  is  oval,  and  is  14  lines  in  one 
diameter  and  12  lines  in  the 
other. 


212 


NORTH-CAROLINA   GEOLOGICAL   SURVEY. 


FIG  34. 


The  oldest  specimen  of  fossil  be- 
longing to  the  whale  or  cetacean 
family,  belongs  to  the  genus  Phy- 
seter,  and  is  regarded  as  the  P.  an- 
tiquus,  (fig.  34.)  It  occurs  in  the 
eocene  of  Craven  county.  The 
size  of  the  teeth  pro  re  that  they 
belonged  to  the  largest  of  the  class. 
The  largest  tooth  measures  six  inch- 
es in  circumference,  and  is  five  and 
a  half  inches  long,  though  a  por- 
tion has  been  broken  from  the  base. 
Its  form  is  quadrangular,  and  pre- 
sents a  curve  in  front,  but  is  rather 
straight  behind.  It  shows  no  con- 
ical cavity,  but  is  solid  throughout. 
It  shows'  a  tendency  to  exfoliate 
concentrically.  Many  fragments 
more  or  less  rolled  and  otherwise 
defaced,  have  been  seen  in  the  mi- 
ocene  beds  upon  the  Tar  River. — 
It  is  probable  they  may  have  been 
removed  from  a  lower  to  an  upper 
formation. 


irORTH-CABOLINA   GEOLOGICAL   SUEVEY.  213 


CHAPTEK  XYI. 

REPTILIA. 

Description  of  Reptilian  remains  of  the  marl  beds  of  North-Carolina, — 
Reptiles  of  the  Green  sand. 

I  was  fortunate  in  discovering  a  vertebra  of  a  large  size  on 
the  lower  Cape  Fear,  which,  at  the  time,  I  supposed  to  be 
new.  As  the  discovery  was  confined  to  this  single  piece  of 
the  skeleton,  I  deemed  it  insufficient  to  draw  from  it  special 
conclusions  respecting  the  family  of  saurians  to  which  it  be- 
longed. 

Since  this  discovery,  Prof.  II.  D.  Rodgers  has  presented  to 
Prof.  Owen,  of  London,  a  collection  of  vertebrae  from  the 
green  sand  of  New  Jersey,  among  which  I  find  the  saurian 
described,  to  which  my  North-Carolina  fossil  must  belong. 

Figure  34  (A.)  represents  the  vertebra  from  the  upper  part  of 
the  green  sand  of  North-Carolina.  It  belongs  to  the  lumber 
region.  Its  type  is  procelian,  that  is,  it  is  concave  before 
and  convex  -behind,  like  the  crocodiles  of  the  present  day. 
The  body  is  long,  and  from  the  anterior  half  it  sends  off' 
strong  processes  at  nearly  right  angles,  which  are  thin  and 
strong.  The  articulating  extremities  are  less  concave  and 
convex  than  those  of  the  alligators  of  the  Southern  States. 
In  this  character  I  find  it  agrees  essentially  with  those  of 
New  Jersey. 

The  abdominal  face  is  smooth,  and  marked  by  two,  or  a 
pair  of  elongated  holes,  situated  rather  nearer  the  concave 
than  the  convex  end.  The  body  is  cylindrical,  especially  pos- 
teriorly. Prof.  Owen  refers  the  New  Jersey  saurian  to  the 
lizards  and  to  the  mososaurian  type.  The  name  which  has 
been  conferred  upon  this  remarkable  saurian  is  Macrosaurus. 
If  my  determination  is  right  with  respect  to  the  identity  of 
the  New  Jersey  and  North-Carolina  specimens,  it  will  be 
known  by  the  same  name.  This  vertebra  is  three  and  three 


214 


NORTH-CAROLINA   GEOLOGICAL   SURVEY. 


.       .  .       ,     ,.  .  ,  .   ,  ,      ,     i~ 

quarter  inches  long,  including  convexity,  which  equals  half 
an  inch,  and  six  inches  from  the  end  of  one  parapophysis  to 
the  other ;  across  the  concave  articulation  nearly  two  and  a 
half  inches ;  across  the  convex,  two  inches ;  length  of  the 
lateral  process,  nearly  two  inches. 

FIG.  34  (A.) 


The  entire  length  of  this  saurian  cannot  have  been  less 
than  twenty-five  feet,  and  it  is  a  fact  worthy  of  notice,  that 


NORTH-CAROLINA    GEOLOGICAL    SURVEY. 


gaurians  of  this  description  inhabited  a  region  as  far  north  as 
New- York,  while  at  the  present  day  their  limits  are  confined 
to  the  central  parts  of  North-Carolina.  This  fact,  no  doubt, 
indicates  a  milder  climate  in  New- York  and  New-Jersey  than 
is  known  at  the  present  day.  All  the  large  land  reptiles  are 
confined  to  the  warmer  regions  of  the  globe. 


CROCODILUS     ANTIQUUS.— LEIDY. 

Another  extinct  saurian  (fig.  35,  A.,)  is  indicated  in  the  dis- 
covery of  vertebras,  which  belong  to,  or  are  found  in,  the  mio- 
cene  marls.  The  most  perfect  one  which  I  have  obtained,  is 

the  2d  caudal,  which 

FIG.  35.  (A.)  as    Jt    is    possible     to 

identify  it,  may  be 
compared  with  the 
Alligator  luscius,  the 
common  large  rep- 
tile of  the  Southern 
States,  inasmuch,  too 
as  it  belongs  to  the 
same  type  of  verte- 
brae. 

This  vertebra  dif- 
fers from  the  corres- 
ponding one  to  which 
I  have  referred  it ;  it 
is  rather  larger  and 

thicker,  and  the  proportion  of  its  parts  differ  also.  Its  length 
is  one  quarter  of  an  inch  greater,  but  its  diameter  at  the  con- 
cave end  is  three-eighths  greater,  and  the  size  or  diameter  of 
the  body  is  still  greater.  The  fossil  is  thick  through  its  whole 
length,  and  varies  but  little  at  the  ends  ;  or  it  is  much  less 
compressed  laterally  than  the  vertebra  of  the  living  Alliga- 
tor, and  what  is  equally  worthy  of  note,  is,  that  the  transverse 
processes  come  out  more  immediately  from  the  body  of  the 
vertebra  than  the  other.  One  more  point  may  be  made;  a 
ridge  of  bone  begins  near  the  middle  at  the  concave  end,  and 
runs  a  little  downwards,  until  it  reaches  a  slightly  constricted 


216 


NORTH-CAROLINA   GEOLOGICAL   SURTET. 


part  just  before  the  border  which  surrounds  the  convex  ex- 
tremity; this  gives  the  appearance  of  breadth  to  the  bone 
when  we  look  upon  the  abdominal  face.  There  is  a  slender 
sharp  ridge  occupying  the  same  relative  position  in  the  Alli- 
gator, but  it  seems  to  originate  at  the  convex  extremity.  A 
slight  groove  runs  longitudinally  upon  this  face.  Length,  one 
and  eight-tenths;  width,  over  the  concave  end,  one  and  five- 
tenths  inches. 

From  all  that  I  have  been  able  to  glean  from  the  discover- 
ies of  others  in  this  country,  these  vertebra  appear  to  belong 
to  a  species  which  has  been  discovered  in  the  miocene  marls 
of  New  Jersey  and  Virginia.  The  species  is  now  extinct,* 

The  cranial  plates,  one  of  which  is  illustrated  by  figure  30, 
belongs  to  a  large  unknown  saurian.  These  were  taken  from 

FIG.  36. 


the  miocene  upon  the  Neuse,  fifteen  miles  below  Goldsboro'. 
They  are  over  half  an  inch  thick,  and  ornamented  with  deep 
sculpturings,  and  from  their  massiven ess  might  be  referred  to 
the  Macrosaurus.  But  this  reptile  belongs  to  an  older  formation. 
I. have,  however  a  laniary  tooth  of  the  proper  dimensions  to 
correspond  in  size  with  the  saurian,  which  may  have  been 
provided  with  this  impenetrable  armour,  and  also  the  middle 


*  Proceeding  of  the  Academy  of  Nat.  Sciences,  Phil.,  Vol.  V,  p.  307. 


NORTH-CAROLINA    GEOLOGICAL    SURVEY. 


217 


part  of  a  femur  to  match  both  the  plates  and  tooth,  and  all 
from  the  miocene  or  shell  marl.  The  materials,  however,  for 
drawing  np  a  proper  description  of  this  saurian,  do  not  exist 
at  present. 

MOSSOSAURUS. 

Tooth  sharp  pointed,  pyramidal  and  curved  backwards  ; 
enamel  moderately  and  finely  wrinkled  ;  surface  divided  into 
two  unequal  parts  by  well  defined  and  finely  serrate  carinae, 
the  anterior  of  which  is  considerably  curved  on  the  last  half 
inch,  which  forms  the  apex.  Base  of  the  outer  surface  smooth, 
and  forming  the  segment  of  a  large  circle  ;  this  smooth 
band  is  usually  covered  with  a  thin  enamel,  and  is  a  little  over 
a  line  wide.  The  rest  of  the  outer  surface  is  divided  by  three 
ridges,  the  middle  is  strong,  and  extends  to  the  point  ;  the  an- 
terior dies  out  about  half  an  inch  from  the  apex  ;  the  posterior 
is  inconsiderable,  and  extends  a  little  more  than  half  way  to 

the  apex  ;  these  ridges  divide  the  surface 

towards    the    base 

into    three    slight- 

ly concave  surfaces. 

The     inferior     has 

eight  distinct  ridges 

none  of  which  reach 

the  apex  ;  these  di- 

vide this  strongly  convex  face  into  nine 

slightly  concave  facets,  of  which  those  ad- 

jacent to  the  carinae  are  the  widest,  (Fig. 

36,  A.)  side  view,  natural  size,  (Fig.  37,) 

viewed  from  the  point,  showing  the  di- 

vision into  parts  and  its  polygonal  form. 
Ifc-is  possible  this  tooth  may  differ  from 

others  which  have  been  described.     It 

differs  from  the  one  described  by  Dr. 
DeKay*  in  being  finely  rugose,  and  distinctly  serrate,  neither 
does  he  speak  of  angularities,  though  they  are  faintly  indi- 


.  36.  (A.) 


FIG.  37. 


*  Annals  of  the  Lyceum  of  N.  Y.,  vol.  3,  p.  186. 


218 


NORTH-CAROLINA    GEOLOGICAL    SURVEY. 


cated  as  existing  upon  the  outer  face  in  his  transverse  section, 
but  that  those  faces  are  concave  has  not  been  stated  by  any 
writer. 

The  transverse  section  of  the  tooth,  Mossosaurus  Hoffman  i, 
given  by  Prof.  Owen,  has  no  angularities  at  all  on  either 
face — the  figure  of  the  M.  Maximiliani  exhibits  them  upon 
the  anterior  face,  but  none  upon  the  inner. 

The  tooth  which  I  have  just  described  is  perfect,  and  not 
worn  ;  the  figures  are  good  illustrations  of  its  characters,  and 
it  appears,  therefore,  that  the  characters  are  either  not  uniform 
or  else  there  are  two  species  belonging  to  the  green  sand.  It 
is  evident  that  the  tooth  in  question  belongs  to  the  species, 
Maximiliani,  rather  than  the  lioffrnani  or  gracilis. 


Fio.  37.  (A.) 


POLYGONODON    RECTUS, LEIDY.        MOSSOSAURUS    RECTUS. 

Tooth  long,  pointed,  compressed;  near- 
ly equally  divided  on  the  outer  and  in- 
ner faces ;  the  faces  are  formed  by  five 
equal  and  similar  planes,  bounded  by 
angular  ridges,  only  two  of  which,  on 
each  face,  can  be  said  to  approach  the 
apex ;  these  are  the  two  anterior  and  twro 
posterior  ridges  curved  backwards  ;  bi- 
carinate  ;  but  the  posterior  edge  is  near- 
ly straight,  while  it  has  a  convexity  be- 
fore which  gives  an  apparent  curvature 
which  does  not  exist ;  edges  smooth ; 
enamel  is  probably  thin  or  removed, 
leaving  a  dense  dentine,  with  fine  longitudinal  cracks  which 
appear  at  first  like  fine  striae.  The  tooth  is  broken  at  the 
base  of  the  crown,  showing  a  shallow  pulp  cavity. 

This  tooth  differs  from  any  of  the  preceding  in  its  form  and 
surface.  It  is  particularly  noticeable,  that  the  part  near  the 
base  is  distinctly  angular,  and  is  divided  into  ten  nearly  equal 
planes,  and  is  bounded  by  well  defined  angles.  All  these 
angles  extend  a  little  above  the  middle  of  the  tooth.  It  dif- 
fers from  either  of  the  three  species  of  Mossosausus  in  its  pro- 
portions. It  also  differs  from  the  teeth  of  the  Leiodon,  by  be- 


NORTH-CAROLINA   GEOLOGICAL   SURVEY. 


219 


ing  much  more  compressed.  The  teeth  of  the  Polyptichodon 
are  circular,  and  the  teeth  also  of  the  Pliogonodon,  which  I 
found  upon  the  Cape  Fear,  are  also  quite  circular  and  conical. 
It  is  possible  it  may  be  a  palatine  tooth  of  the  M.  Maximilian!. 
It  differs,  however,  in  form  from  those  teeth.  It  appears  to 
have  had  that  kind  of  attachment  to  the  jaw,  which  has  been 
called  acrodont.  Length,  one  and  three-quarter  inches ; 
width,  at  base,  seven-sixteenths. 


FIG 


POLYPTYCHODON — OWEN.       POLYPTOCHODON    KUGOSUS. E. 

The  teeth  (Figs.  38  and  39)  which  are  represented  in  the 
margin  were  discovered  in  a  bed  of  miocene  marl  at  Elizabeth- 
town,  Bladen  county,  in  1852-'3.  They  were  regarded  at 
the  time  as  having  belonged  to  an  extinct  undescribed  species. 
I  have  had  hopes  that  other  parts  of  this  saurian  would  be 
discovered  which  would  throw  some 
light  upon  its  organization  and  form, 
but  as  yet  no  bones  which  can  be  re- 
ferred to  the  genus,  or  species  to  which 
the  teeth  belonged,  have  come  to 
light.  Saurian  bones  of  a  large  size 
are  not  wanting  which  may  have  be- 
longed to  the  teeth  under  considera- 
tion, but  more  than  one  species  have 
been  discovered.  In  one  instance  the 
middle  of  a  large  femur;  in  others 
cranial  plates,  the  sculpturing  of  which 
are  quite  different,  are  among  the 
bones  which  have  been  discovered. 
These,  however,  are  disconnected  frag- 
ments, and  hence  are  insufficient  to 
settle  the  question  of  ownership.  The 
epoch  to  which  the  bones  referred  to 
belong  is  not  at  all  established.  Large 
saurian  vertebra  have  been  found  in 
the  green  sand,  and  teeth  resembling 
those  found  at  Elizabethtown  in  the 
eocene  marl  upon  the  Neuse.  Hence  it  is  probable  that  the 


220 


NORTH-CAROLINA   GEOLOGICAL   SURVEY. 


FIG.  39. 


epoch  of  these  reptiles  is  earlier  than  that  of  the  miocene 

beds.  They  are  found  in  those 
beds  for  the  same  reason  that 
the  exogyra  costata  of  the  green 
sand  is  also  found  in  the  mio- 
eene.  While  it  is  clear  enough 
that  fossils  have  been  washed 
out  of  the  green  sand  into  the 
miocene.  I  have  no  evidence 
that  they  have  been  transport- 
ed into  the  eocene,  the  next 
series  above.  The  deposits  seem 
to  have  quietly  succeeded  the 
green  sand  ;  but  when  the  mio- 
cene period  arrived,  there  was 
a  breaking  up  of  the  older 
series,  and  their  contents  carri- 
ed immediately  up  to  this  pe- 
riod, and  under  favorable  cir- 
cumstances fossils  of  both  periods  were  intermingled  together, 
and  hence  I  regard  the  animals  under  consideration  to  have 
lived  before  the  miocene  beds  were  deposited. 

The  teeth  which  I  have  figured  I  have  referred  to  a  genus 
of  crocodilian  reptiles  established  by  Prof.  Owen,  and  which, 
in  England,  belonged  to  the  chalk  or  cretaceous  system. 

The  following  description  is  drawn  from  the  teeth  before 
me :  Teeth  thick  and  conical,  and  slightly  curved ;  trans- 
verse section  circular  or  round ;  enamel  traversed  longitudi- 
nally by  numerous  transversely  rugose  cracks,  the  strongest 
of  which  reach  the  apex ;  no  trenchant  edges  or  carinae 
proper. 

The  teeth  are  only  gently  curved ;  they  are  very  strong 
and  robust,  and  the  enamel  is  traversed  by  rather  irregular 
rugose  ridges,  which  appear  like  cracks.  The  inside  ridge  is 
stronger  than  the  others,  and  are  formed  of  two  confluent 
ones,  and  takes  the  place  of  a  carina,  and  extends  to  the 
point  in  the  young  tooth ;  but  in  old  and  worn  teeth  most  of 
the  ridges  terminate  considerably  below  the  apex.  The  sur- 


NORTH-CAEOLINA    GEOLOGICAL    SURVEY. 


221 


face  of  the  young  tootli  (Fig.  39)  is  very  rough,  and  the  edge? 
of  the  rugosities  are  realy,  irregularly  serrate,  and  run  into 
each  other.  The  section  is  round  at  all  places,  from  the  base 
to  the  apex.  Its  crown  is  hollow,  and  its  pulp  cavity  presents 
a  conical  hollow  which  extends  about  one-third  of  the  length 
of  the  crown.  On  exposure  to  the  weather,  the  crown  ex- 
foliates in  conical  layers.  Below  the  crown,  that  part  known 
as  the  root  is  hollow,  but  has  a  thick  strong  shell,  which  on 
the  concave  side  has  three  wide  shallow  furrows ;  the  middle 
one  is  exactly  in  the  concavity ;  they  occupy  about  one-third 
of  the  cylinder;  the  remainder  is  perfectly  circular. 

Prof.  Owen's  description  of  the  potyptychodon*  is  as  fol- 
lows :  "  Teeth  thick  and  conical ;  transverse  section  of  the 
crown  circular,  without  larger  or  trenchant  ridges ;  enamel 
ridged  longitudinally,  but  only  a  few  reaching  the  apex. 
The  crowns,  when  weathered,  exfoliate  in  a  conical'  manner 
by  detached  layers,  like  a  cone  in  conej  base  having  a  con- 
ical pulp  cavity  which  opens  into  the  crown  in  distinct 
sockets." 

The  foregoing  description  of  Prof.  Owen,  of  the  genus  Po- 
lyptychodon,  applies  so  well  to  our  teeth,  that  there  can  re- 
main scarcely  a  dtmbt  as  to  their  generic  identity.  It  is, 
however,  only  a  generic  similarity  ;  the  species  appears  to  be 
quite  different  from  both  of  the  species  described  by  Prof. 
Owen,  and  from  its  remarkable  rugose  enamel,  I  propose  as 
its  specific  name,  rugosus. 

It  differs  from  the  Alligator  in  the  absence  of  a  deep  con- 
striction at  the  base  of  the  crown,  from  the  Pliogonodon  of 
Leidy,  by  its  robustness  and  rugosities,  and  from  the  Ellipton- 
odon,  by  its  circular  section,  this  having  a  circular  section  only 
at  the  base  of  the  crown,  while  in  the  former  the  crown  haa 
a  circular  section  from  base  to  apex. 

Sculptured  Cranial  Plate,  (Fig.  40.) — These  plates  are  sep- 
arated from  each  in  the  line  of  suture,  and  are  generally  bro- 
ken. They  are  thick  and  strong,  and  were  no  doubt  sufficiently 


*  Palaeontographical  Society's  translation,  p.  46,  rol.  for  1851.     (Description  of  ths 
P.  interuptus  and  continuus.) 


222 


NORTH-CAROLINA   GEOLOGICAL    SURVEY. 


BO  to  resist  the  entrance  of  a  musket  ball.     The  same  remarks 

it   regards    ownership 


as 


FiG.  40. 


have  already  been  made, 
respecting  other  bones  of 
this  class,  so  common  in 
these  deposits.  That  there 
were  two,  at  least,  power- 
ful reptiles,  is  evident  from 
their  bones  and  teeth,  but 
in  no  instance  have  two 
been  found  attached,  and 
in  such  relations  that  it 
|would  be  safe  to  affirm  that 
they  belonged  to  the  same  individual. 


FIGS.  41   &  42. 


EL  LIPTONODON     COMPRESS  US . EMMON  8 . 

Tootli  curved,  robust,  sub-conical  and 
pointed  ;  crown  circular  at  base,  becom- 
ing elliptical,  and  finally  sub-elliptical,  or 
with  the  inside  more  flattened  or  less 
convex  than  the  other ;  bicarinate ;  the 
anterior  ridge  becoming  obsolete  near 
the  base  of  the  crown,  and  without  ser- 
ratures  or  rugosities ;  enamel  rather  tine- 

O  ' 

ly  wrinkled  longitudinally,  or  faintly  ru- 
gose, and  none  of  the  rugosities  extend  to 
the  apex;  dentine  is  concentric;  pulp 
cavity  open,  conical,  carinate.  Figures 
natural  size.  Figure  42,  transverse  sec- 
tion. 

This  tooth  is  broken  at  the  base  of  the 
crown,  and  has  lost  a  small  part  of  its 
apex. 

It  differs  very  clearly  from  the  Polyp- 
tychodon,  Pliogonodon,   Mossosaurus  or 
Fleiosaurus.     The  clear  and  distinct  marks  of  difference  aiv 
shown  in  the  figures  of  each  referred  to  except  the  Pleiosau- 
rus.     This  tooth  was  found    in   the  miocene  near  the  Cape 


NORTH -CAROLINA  GEOLOGICAL  SURVEY, 


223 


Fear  River,  in  Bladen  county.  As  the  bones  which  have 
been  found  in  these  beds  indicate  the  existence  at  a  prior  pe- 
riod of  two  large  and  formidable  saurians,  so  the  teeth  con- 
lirm  this  view,  and  I  have  placed  in  this  connexion  a  sculp- 
tured cranial  plate,  (tig.  40,)  which  may  have  belonged  to  this 
species. 

Additional  discoveries,  however,  are  required  before  it  is 
possible  to  determine  to  which  of  these  plates  the  teeth  re- 
spectively belong.  All  the  bones  which  are  found  in  the  mi- 
ocene  beds,  are  broken,  though  they  are  mixed  with  perfect 
delicate  shells.  This  fact  proves  that  the  bones  were  subjec- 
ted to  violence  before  they  were  imbedded  in  the  rniocene, 
and  hence  belong,  probably,  either  to  the  eocene  or  green  sand. 

PLIOGONODON    NOBILI3.       LEIDY. (FigS.    43    &    44.*) 

In  the  collection  of  Prof.  Ernmons  there  are  two,  much 
mutilated  teeth  of  a  saurian  discovered  in  a  miocene  deposit 
of  Gape  Fear,  North-Carolina.  These  teeth,  which  have  lost 

their  fang  and  summit,  are  clou, 
gated  conical,  nearly  straight  or 
only  slightly  curved  inwardly. 
Their  section  is  circular  with  an 
inner  pair  of  opposed  carinne  ; 
and  their  surface  is  subdivided 
into  numerous  narrow  planes  and 
provided  with  a  few  vertical  in- 
terrupted plicae,  which  are  more 
numerous  internally.  The  base 
of  the  crown  is  conically  hollow  : 
the  dentine  is  concentric ;  and 
the  enamel  is  finely  wrinkled. 

The  specimens  measure  three-fourths  of  an  inch  in  diameter 
at  base,  and  are  about  one  and  a  half  inches  long,  but  when 
perfect  their  crown  has  been  a  half  inch  longer. 


FIGS.  43  &  44, 


*  These  teeth  appear  to  differ,  one  has  a  coarser  aspect,  and  the  striar  are  coarser, 
and  it  is  more  curved,  and  proportion  differs.    Description  by  Prof.  Leidj. 


224 


NORTH-CAROLINA    GEOLOGICAL    SURVEY. 


FIGS.  45  &  46 


From  the  teeth  of  Mososaurus  those  of  Pliogonodon  differ 
in  their  narrower  proportion,  their  straightness,  their  circular 
transverse  section,  their  relatively  narrower  planes,  and  in 
their  possession  of  plicae.  From  the  teeth  of  Polyptychodon 
they  differ  in  the  possession  of  dissimilar  planes  and  carinae, 
and  in  their  less  degree  of  robustness ;  and  from  those  of 
Pleiosaurus  in  the  existence  of  divisional  planes  and  the  cir- 
cular section. 

PREPANODON    IMPAR.       LEIDY. (FigS.  45  &  46.*) 

This  genus  and  species  are  proposed  on  the  crown  of  a 
tooth  resembling  the  corresponding  portion  of  the  inferior 

canine  of  a  bear,  except  that  it  has 
but  a  single  carina,  and  that  on  the 
concave  border  internally.  The  spe- 
cimen was  discovered  by  Prof.  Em- 
mons,  at  Elizabethtown,  Cape  Fear, 
North-Carolina.  It  is  black  in  color, 
curved,  conical,  most  convex  exter- 
nally, and  is  oval  in  transverse  section.  The  base  is  hollowed 
conically,  and  the  enamel  is  smooth.  The  length  of  the  spe- 
cimen is  three-quarters  of  an  inch ;  the  antero  posterior  di- 
ameter of  its  base  is  seven  lines,  and  its  transverse  diameter 
five  lines. 

The  tooth  I  suspect  to  have  belonged  to  a  crocodilian  rep- 
tile, though  it  may  possibly  even  prove  to  be  a  mammalian 
relic. 


*  Described  by  Prof.  Leidy. 


NORTH-CAROLINA   GEOLOGICAL   SURVEY. 


225 


CHAPTER  XVII. 


PISCES. 


Description  of  the  remains  of  Fish  in  the>  North-Carolina  Marl  beds. 


FIGS.  47  &  48. 


ISCHYRHIZA   ANTIQUA. LEIDY. 

The  curious  genus  Ischyrhiza, 
was  first  brought  to  my  notice 
by  the  discovery  of  a  tooth  in 
the  Green  Sand  of  ISTew  Jersey,  by 
Prof.  Leidy.  My  collection  con- 
tains several  teeth  discovered  on 
the  Neuse  River.  In  most  speci- 
mens the  crown  has  lost  its  apex, 
but  the  fang  is  entire.  In  the  per- 
fect condition,  the  crown  has  been 
laterally  compressed,  conical  and 
inverted  with  smooth,  shining  en- 
amel. The  fang  expands  from  the  crown  in  a  pyramidal  man- 
ner; is  quadrilateral,  curved  backward,  and  divided  at  base 
antero-posteriorly  ;  the  division  becoming  deeper  posteriorly. 
The  larger  specimen,  in  the  figure,  which  is  of  a  red  color, 
when  perfect,  was  nearly,  or  perhaps  quite  two  inches  in 
length.  Its  fang  is  an  inch  long,  eight  lines  antero-perterior- 
ly  at  base,  and  six  lines  transversely.  The  base  of  the  crown 
is  oval  in  section,  and  measures  six  lines  antero-perteriorly, 
and  five  lines  transversely. 

The  smaller  specimen  is  black  in  color,  and  was  about  half 
an  inch  shorter  than  the  other.  Its  fang  is  about  ten  lines 
long,  and  at  base  is  about  six  lines  square.  It  belongs  to 
miocene  of  Korth-Carolina. 


IX)SSIL    SQUALIDAE    OF    THE   TERTIARY    OF  THE   EASTERN    COUNTIES. 

The  fossil  squalid ae,  or  sharks,  are  known  only  by  their 
teeth,  as  these  are  the  only  parts  which  are  usually  preserved. 
16 


226  NORTH-CAROLINA   GEOLOGICAL   SURVEY. 

Their  vertebrae  are  sometimes  preserved,  but  they  must  be 
exceedingly  rare  in  beds  which  are  as  loose  as  the  clays  and 
sands  of  the  tertiary  deposits.  But  the  teeth,  being  protect- 
ed by  a  very  dense  enamel,  and  having  a  firm  strong  core,  re- 
sist change  for  ages ;  it  is  in  these  organs,  therefore,  that 
memorials  of  this  highly  interesting  order  of  fish  have  been 
preserved. 

The  teeth  being  attached  loosely  to  a  cartilaginous  jaw. 
are  almost  always  separated  and  detached ;  and  hence,  they 
occur  singly.  Of  the  mode  in  which  they  are  connected,  we 
are  informed  by  the  living  species  which  inhabit  the  adjacent 
seas.  From  this  source  of  information,  we  may  be  assured 
that  these  single  teeth  were  arranged  in  several  rows  in  both 
jaws ;  that  only  a  single  one,  those  of  the  front,  stood  up- 
right, while  the  remainder  lay  flat  with  the  points  directed 
backwards,  or  obliquely  so.  When  the  front  teeth  drop  out. 
its  place  is  supplied  at  once  by  the  uprising  of  that  one  which 
is  opposite  the  vacant  space.  The  teeth,  though  very  numer- 
ous, differ  but  little  in  form,  though  they  differ  more  in  size. 
The  most  remarkable  difference  may  be  observed  on  compar- 
ing the  symphysal  teeth,  or  middle  row  with  those  on  each 
side.  Thus,  Fig,  49,  shows  a  front  section  of  the  lower  jaw 
of  the  galeocerdo  arctimis ;  the  outer  row  standing  upright, 

FIG.  49. 


those  behind  lying  flat,  and  the  middle  teeth  consisting  of  a 
series  of  small  ones.  This  figure,  therefore,  is  a  type  by 
which  the  reader  may  compare  the  prevailing  arrangements 
in  the  existing,  as  well  as  in  this  extinct  family  of  fishes. 


NORTH-CAROLINA   GEOLOGICAL   SURVEY. 


227 


GENUS    CARCHARODON. — SMITH. 

Teeth  very  large,  broad,  triangular  and  rather  uniformly 
dentated  in  both  jaws.  The  enamel  is  usually  cracked  longi- 
tudinally ;  roots  massive  and  divergent ;  inside  nearly  flat ; 
surfaces  smooth,  and  scarcely  ever  striated. 

CARCHARODON   MEGALODON.— AGASS.       (Fig.    50.) 

This  species  has  the  form  of  an  equilateral  triangle,  though 

FIG.  50. 


228 


NORTH-CAROLINA   GEOLOGICAL   SURVEY. 


it  admits  of  slight  variations ;  teeth  somewhat  oblique,  or  in- 
clined to  the  posterior  end ;  upper,  or  outer  side,  nearly  flat ; 
imder  side  prominently  convex  in  the  middle ;  enamel  cracked 
longitudinally  on  both  sides,  particularly  along  the  middle ; 
serratures  rather  indistinct  from  the  use  of  the  tooth ;  core 
coarsely  striated.  It  is  usually  found  in  the  miocene  beds, 
and  is  the  most  common  upon  the  Cape  Fear. 

If  the  size  of  the  teeth  furnish  an  indication  of  the  strength, 
size  and  ferocity  of  this  species  of  shark,  then  it  must  have 
been  one  of  the  largest  and  most  formidable  animals  of  the 
ocean,  combining,  as  Prof.  Owen  remarks,  with  the  organiza- 
tion of  the  shark,  its  bold  and  insatiable  character,  they  must 
have  constituted  the  most  terriffic  and  irresistable  of  the  pre- 
daceous  monsters  of  the  ancient  deep.  The  largest  of  the 
teeth  measure  sometimes  six  inches  in  length,  and  from  four 
to  live  wide  at  base. 

The  jaws  of  the  largest  species  of  shark  known  in  modern 
times  measure  about  four  feet  around  the  upper,  and  three 

feet  eight  inches  around 


the  lower  jaw.  The 
length  of  the  largest 
tooth  is  two  inches,  and 
the  total  length  of  the 
shark,  when  living,  was 
thirty-seven  feet.  If  the 
proportions  of  the  ex- 
tinct shark  bore  the 
same  as  those  of  the 
living,  their  length  must 
Iiave  been  over  one 
hundred  feet,  equaling 
in  this  respect,  the  larg- 
est of  the  whales. 

Figure  51  shows  a  smaller  tooth  of  the  carcharodon  mega- 
lodon. 


NORTH-CAROLINA   GEOLOGICAL   SURVEY. 


229 


CHARCHARODON    FEROX. N.    S.      (FigS.  52,  53,  54.) 

Form  nearly  an  equilateral  triangle,  thick ;  inner  face  very 


FIGS.  52  &  53. 


230 


NORTH-CAROLINA   GEOLOGICAL   SURVEY. 


convex,  outer  nearly  flat,  and  slightly   champered  towards 


FlG-  54-  the  edges,  and  also  slightly 

convex  near  the  middle ;  ser- 
atures  small,  root  thick,  stout 
and  straight  across  the  base, 
and  sloping  on  the  inner  face. 
The  form  of  this  tooth  diifers 
materially  from  the  megalod- 
on,  especially  in  the  relations 
of  its  height  and  breadth ; 
height,  four  inches  and  a  half, 
breadth  at  base,  five  inches ; 
thickness  of  the  root,  one  inch 
and  a  half,  measured  over  the 
slope  ;  length  from  the  apex  to 
the  base  of  the  root,  five  inch- 
es, measured  along  the  edge  ; 
thickness  at  the  middle,  one 
inch.  Found  in  the  eocene  of 
Craven  county,  JST.  C.  The 
dimension  of  this  species  of 
shark  equals  that  of  the  car- 
charodon  megalodon.  The 
tooth  is  thicker  and  stouter 
than  this  species,  and  more 
convex  posteriorly,  straighter 
across  the  base,  and  propor- 
tionally wider.  Fig.  52  shows 
the  outline  of  the  tooth,  fig.  54 
is  an  edge  view,  and  figure  53 
a  transverse  section,  showing 
convexity  of  the  inferior  face,  and  the  flatness  of  the  superior. 

CARCHARODON    STJLCIDENS. AGASS.       (FigS.  55  &  56.) 

Teeth  large,  thin  and  pointed ;  their  forms  correspond  very 
closely  to  that  of  an  isosceles  triangle.  They  are  flat  on  one 
side ;  the  enamel  extends  to  the  root  on  both  sides ;  it  is  more 
regularly  sulcated  upon  the  convex  than  upon  the  other  side  ; 
fig.  55  young  of  the  sulcidens. 


NORTH-CAROLINA   GEOLOGICAL   SURVEY. 
FIG.  56. 


231 


CARCHARODON   ANGUSTIDEN8.      (FigS.  57  &  58. 
FIG.  57. 


232 


NORTH-CAKOLINA   GEOLOGICAL   SURVEY. 


Crown  only  slightly  oblique,  rather  thick,  but  comparative- 
ly narrow,  but  wide  at  base,  and  armed  with  serrated  wing- 
lets,  pointing  upwards  and  outwards ;  the  serratures  are  strong- 
er than  those  upon  the  crown  ;  roots  massive,  and  separated 
by  a  distinct  arch.  Figure  58,  a  tooth  which  should  probably 
referred  to  this  species,  though  the  arch  of  the  root  is  flatter. 

Prof.  Gibbs,  on  the  authority  of  Prof.  Agassiz,  has  merged 
in  the  carcharodon  angustidens,  the  following  species :  C.  lan- 
ceolatus,  0.  heterodon,  C.  megalotis,  C.  semi-serratus,  C.  au- 
riculatus,  C.  turgidus,  C.  semi-serratus,  and  C.  toliapicus,  on 
the  ground  that  they  are  insufficiently  characterized  and  not 
clearly  distinguishable  from  each  other. 

CAECHAEODON   TRTANGTJLARIS,    N.    S.       (Fig.    59.) 

Crown  of  the  tooth  rather  thin  ;  the  posterior  faces  of  the 
crown  meeting  in  the  central  line  at  an  obtuse  angle,  but  upon 

FIG.  59. 


**•••—• 


NORTH-CAROLINA   GEOLOGICAL   SURVEY. 


238 


FIG.  59,  a. 


each  side  of  this  line  they  are  quite  flat ;  enamel  thin,  serra- 
tures  small,  root  thick,  striated  and  heavy,  with  a  very  low 
arch. 

This  tooth  scarcely  exhibits  the  usual  convexities  of  either 
face ;  the  faces  being  bounded  by  plane  surfaces,  the  meeting 
of  which  give  an  obtuse  angle  when  obtained  by  a  central 
section  through  the  crown.  It  belongs  to  the  eocene,  and  was 
obtained  from  a  bed  near  JNTewbern. 

C.    CRASIDENS,    N.    S.       (Fig.  59,  a.) 

Tooth  sub-conical,  thick,  slightly  oblique ;  inner  face  very 
convex,  outer  flat  at  base,  evenly  but  flatly  convex  near  the 

apex,  with  an  inconsiderable 
ridge  extending  from  the 
base  to  a  point  near  the  apex, 
and  somewhat  ridged  across 
the  whole  of  the  base  of  the 
outer  face ;  serrae,  sub-equal, 
and  armed  with  serrate  wings 
at  base ;  root  thick  and  pro- 
minent on  the  inside ;  en- 
amel extends  on  the  outer 
face  to  the  root,  and  is  ex- 
tended continuously  over  the 
wings.  This  tooth  belongs 
to  the  eocene  at  Wilmington. 
It  is  distinguishable  from  oth- 
er teeth  belonging  to  this 
order  of  fishes,  by  its  very 
uniform  degree  of  thickness 
from  the  base  of  the  root, 
near  its  termination,  at  the  apex. 

CARCHARODON   CONTORTIDENS. N.    S. (Fig.  60.) 

Tooth  an  irregular  cone,  with  the  crown  twisted  near  the 
summit;  base  of  the  root  nearly  plane,  with  the  branches 
projecting  upwards,  rather  than  downwards,  so  much  so  as  to 
stand  upright  when  placed  upon  its  base;  inside  the  base 


234 


NORTH-CAROLINA   GEOLOGICAL   SURVEY. 


projects  enormously  inward ;  enamel  thin  ;  serratures  small, 
subequal ;  inner  face  very  convex  ;  outer  nearly  flat  at  base, 
but  traversed  longitudinally  by  an  inconsiderable  prominence. 


FIG.  58. 


FIG.  60. 


Only  one  tooth  of  this  description  has  been  obtained  from 
the  eocene  at  Wilmington.  The  form  of  the  tooth  is  very 
peculiar,  and  may  be  readily  distinguished  by  the  great  thick- 
ness of  its  root  and  projection  inward.  This  projection  is 
on  a  level  with  the  branches  of  the  root.  The  twist  also,  at 
the  extremity,  is  also,  a  prominent  feature  in  this  tooth.  It 
is  probable,  this  tooth  indicates  the  existence  of  a  genus, 
which  should  be  separated  from  the  carcharodon,  but  the  ex- 
istence of  a  single  tooth  does  not  furnish  all  the  characteris- 
tics which  probably  belong  to  it. 


NORTH-CAROLINA   GEOLOGICAL   SURVEY.  235 

SPHENODUS    RECTIDENS. N.    S. (FigS.  61  &  62.) 

Tooth  very  long ;  comparatively  slender ;  both 
62*  faces  convex  ;  internal  more  so  than  the  external ; 
becoming  narrower  towards  the  edges ;  the  base 
in  some  of  the  teeth  trenchant,  then  nearly  par- 
allel two- thirds  the  length ;  enamel  rather  thick 
grooved  on  the  inside,  and  cracked  longitudinally 
on  both,  with  a  texture  coarser  than  in  the 
lamna ;  root  unknown.  Figure  62,  transverse 
section.  Green  sand  of  l^orth-Carolina. 

GENUS    HEMIPRISTIS. 

Apex  simple  and  smooth;  margins  of  the  tooth 
denticulated  to  a  point  near  the  apex. 


HEMIPRISTIS   SERRA. — AGASS. — (Fig.    63.) 

The  H.  serra  is  characterized  by 
teeth  which  are  serrated  to  a  point  near 
the  apex,  where  the  serratures  cease, 
and  the  margins  are  left  smooth. 


HEMIPRISTIS    CRENULATUS. N.    S. 

Form  similar  to  the  H.  serra ;  sides 
convex,  long  at  base,  and  rather  thick ; 
enamel  smooth,  and  marked  with  only 

a  few  cracks ;  edges  at  base  faintly  crenate ;  entire  towards 

the  apex. 

GENUS    OXYRHINA. 

Tooth  flat,  broad,  oblique,  lanceolate  and  smooth,  widening 
at  base  rapidly ;  root  thin  and  nearly  straight,  and  destitute 
of  spreading  branches  or  forks. 


236  NORTH-CAROLINA   GEOLOGICAL   SURVEY. 

OXYRHINA  XYPHODON. AGASS. (Fig.  64:.) 

Lanceolate ;  base  of  the  flat  side 
marked  with  shallow  furrows;  en- 
amel extends  a  little  lower  on  the 
inner  than  outer  side. 


OXYRHINA   HASTILIS. AGASS. (FigS.  65  &  66.) 

Tooth  rather  elongated  ;   lanceolate  ;    nearly  equilateral ; 
bone  of  the  enamel  more  arched  than  that  of  the  oxyrhina 


FIG.  65. 


FIG.  66. 


xyphodon,  and  the  root  seems  to  be  less  developed, 
ly  resembles  the  xyphodon. 


It  close- 


OXYRHINA   DESORH. GIBBS. — (Fig.  67.) 

Tooth  thick  and  strong  ;  roots  well  developed  and  forked ; 
enamel  similar  in  texture  to  the  carcharodon,  and  also  cracked 
longitudinally. 


NORTH-CAROLINA   GEOLOGICAL   SURVEY. 


237 


It  differs  from  the  former  in  the  character  of  the  enamel, 
curvatures,  the  absence  of  serratures.. 
and  the  form  and  development  of  its 
root. 


PIG.  67. 


GENUS   GALEOCERDO. — AGASS. 

This  genus  is  an  inhabitant  of  our 
present  seas,  and  the  species  arcticus 
(Fig.  49)  very  closely  resembles  the 
galeocerdo  aduncus,  whose  teeth  are 
abundant  in  the  miocene  marl  beds 
of  North-Carolina.  In  both  jaws  the 
teeth  are  similar  and  equal.  They  form 
five  rows,  which  contain  twenty-three 
teeth  each,  an  odd  small  tooth  occu- 
pying a  middle  position  over  the  sym- 
physis.  The  back  teeth  become  small  and  are  relatively 
shorter  than  the  side  teeth,  presenting  in  this  respect 
an  approach  to  the  form  of  the  teeth  described  as  the 
galeocerdo  latidens.  In  two  species  of  galeocerdo  which  dif- 
fer in  size,  the  serratures  are  constant  and  preserve  a  great 
uniformity;  and  the  common  character  of  the  serratures 
seems  to  be,  that  which  might  be  called  compound,  by  which 
I  mean,  that  each  notch  is  itself  notched,  and  it  is  possible 
that  many  of  the  species  possessed  this  character  more  or 
less,  but  have  lost  it  by  wear  in  their  usage. 

Figure  49  shows  the  arrangement  of  the  front  teeth  of  the 
lower  jaw  in  the  galeocedo  arcticus,  and  the  position  of  a  small 
series  of  teeth  immediately  above  the  simphysis. 


GALEOCERDO    ADUNCUS. AGASS. 

Tooth  oblique  angulated,  and  winged  on  one  side,  or  with 
the  sides  unequal.  Anterior  face  convex,  posterior  rather 
flat.  Serrate,  serratures  unequal,  the  first  upon  the  wing  the 
largest ;  upon  the  arched  edge  the  serratures  are  largest  upon 
the  lower  half  of  the  crown. 


NORTH-CAROLINA   GEOLOGICAL   SURVEY. 


GLLEOCERDO  EGARTONI. 

Tooth  small,  rather  flat,  lanceolate,  slightly  oblique,  convex 
on  both  faces  of  the  crown,  but  concave  at  the  base  on  the 
outer  face  ;  root  spreading  widely,  and  obscurely  wrinkled  ; 
serratures  sub-equal,  serrate  or  finely  lobed  ;  the  enamel  ex- 
tends lower  on  the  outer  than  the  inner  side.  The  latter 
character  I  am  disposed  to  regard  as  its  most  distinguishable, 
for  though  the  size  of  the  teeth  of  this  species  may  vary  con- 
siderably, the  character  of  the  serratures  will  be  preserved. 

GALEOCERDO    STTB-CRENATUS,    N.    S. 

Tooth  nearly  upright,  or  with  only  a  slight  obliquity  poste- 
riorly ;  anterior  edge  formed  by  an  arch  belonging  to  the 
lower  half,  while  the   apical  extremity  or  half  the  edge  is 
straight,  posterior  edge  is  also  straight  for  two-thirds  the  dis- 
tance from  the  apex  to  the  base,  below  which,  the  edge  is 
drawn  inwards ;  there  is  a  constriction  also  on  the  opposite 
edge  at  the  base  of  the  crown  ;  edges  rather  obsoletely  cre- 
nate  than   serrate,  smooth  near  the  apex,  and  the  smoother 
wing  of  the  posterior  edge  stands  at  right  angles  to  the  axis 
of  the  crown ;  upper  face  rather  flat,  and  marked  by  a  faint 
rounded  ridge  extending  from  the  base  to  the  apex,  and  the 
surface  slopes  only  from  this  ridge  to  the  margins.     The  char- 
acteristics of  this  species  will  be  gathered  from  the  preceding 
description.     The  absence  of  distinct  serratures,  the  form  of 
the  crown,  its  constriction  at  base,  are  the  most  important 
points,  in  which  respects  it  differs  from  any  which  I  have 
seen. 

GALEOCERDO    PRISTODONTUS. AGASS. (Fig.  68.) 

Crown  large,  oblique ;  anterior  edge  irregularly  arched, 
and  extending  much  farther  upon  the  base 
than  the  opposite  edge ;  upon  the  flat,  or 
nearly  flat  face,  or  outer  one,  the  enamel 
extends  below  that  on  the  convex  side : 
seratures  unuequal.  Hare  in  Xorth-Caro- 
lina,  but  I  have  several  specimens,  and 
from  Dr.  Gibbs's  account  of  it,  it  seems 
to  be  still  more  rare  in  South-Carolina. 


NOKTH-CAROLINA   GEOLOGICAL   SURVEY. 


239 


FIG.  69. 


G.    LATIDENS. (Fig.    69.) 

Differs  from  the  preceding  in  its  pro- 
portional length  of  base,  being  considera- 
bly greater. 

The  crown  is  low,  and  the  enamel  ex- 
tends lower  upon  the  outer  face ;  the  sen- 
atures  subequal ;  apex  pointed. 

It  is  much  more  common  than  the  G. 


pristodontus. 


GENUS   LAMNA. 

Teeth   rather   flat,   narrow   and   elongated ;   smooth,   and 
usually  furnished  with  appendages  at  base. 

LAMNA    ELEGANS. AGASS. (FigS.  70,  71  &  71  A.) 

Tooth  narrow,  lanceolate ;  inner  face  quite  convex,  outer 
rather  flat  and  smooth ;  the  former  regularly  striate  at  base. 


FIG.  70. 


FIG.  71  A. 


FIG.  71. 


but  towards  the  middle  the  striae  degenerate  into  wrinkles ; 
the  outer  ones  are  short,  and  but  reach  the  edge  of  the  tooth  at 
base.  The  L.  elegans  is  very  common  in  the  miocene  beds  of 
North-Garoltna.  Fig.  71  A,  side  view. 

L.    (ODONTASPIS)    CONTOETIDENS. 

Specimens  which  answer  to  the  figures  of  this  species,  given 
by  Prof.  Gibbs,  especially  in  the  irregular  form  and  absence 


240 


NORTH-CAROLINA   GEOLOGICAL   SURVEY. 


of  denticulatious  at  base.  In  other  characters  there  is  only  a 
slight  difference  between  this  and  the  L.  elegans.  They  are 
found  in  the  same  beds. 


FIG.  72. 


FIG.  73. 


L.    COMPRESSA. 

Compressed  or  flat,  both  faces  convex  and 
sub-equal,  base  irregularly  denticulated;  root 
wide  and  spreading.  It  differs  widely  from 
L.  elegans  and  contortedens,  but  resemble* 
the  otodus;  but  Prof.  Gibbs  remarks  that  they 
are  more  lanciform,  and  the  core  more  slen- 
der than  the  otodus. 

Figures  73  and  74  appear  to  belong  to  the 
lamna.  They  are  rather  thick 
and  stout,  and  resembles  verv 

FIG.  74.  * 

^  .closely     an     oxyrhma.      Mio- 

JrA  cene. 

J|4\  Figures  75,  76,  77,  78,  79, 

4H     ^  8°  an(*  81   kelon     to  the  eo- 


FIGS. 75   &  76. 


cene. 


FIGS.  77  &  78. 


FIGS.  79  &  80. 


FIG.  81. 


L.    CRASIDENS. 

Tooth  thick  and  comparatively  short ;  not  very  thick  and 
projecting  inwardly ;  inner  face  striate  as  in  the  preceding 
species. 

GENUS    OTODUS. 

Tooth  rather  broad  and  flat,  and  armed  with  equal  sharp 
denticles  at  base ;  root  rather  thick,  projecting  inward. 

OTODUS    APPENDICULATUS. AGASS. 

Tooth  oblique,  sharp  or  pointed,  faces  unequally  convex : 


I 


NORTH- CAROLINA   GEOLOGICAL   SURVEY. 


241 


denticles  rather  prominent  and  strong ;  line  of  base  nearly 
horizontal ;  roots  spreading  widely,  forming  a  very  obtuse 
angle  with  each  other. 

I  have  referred  also  the  following  figures  of  teeth  to  the 
genus  otodus :  82,  83,  84,  85,  86,  87,  88.  They  all  belong  to 
the  eocene  formation,  and  occur  in  a  layer  near  the  top.  They 
are  from  the  plantation  of  Mr.  Wadsworth,  of  Craven  county. 


FIGS.  82  &  83. 


FIGS.  84  &  85. 


FIGS.  86  &  87. 


FJG.  bfe. 


GENUS    CORAX. 

The  following  figures  of  teeth  found  in  the  eocene  of  Craven 
county.  I  am  unable  to  refer  them  to  species  already  describ- 
ed, viz :  82,  83,  84,  85. 


FIG.  82a. 


FIG.  83a. 


FIG.  84a. 


FIG.  85a. 


FIGS.  86a  &  87a. 


GENUS  ODONTASPIS. — (Figs.  86a,  87a,  88a,  89a.) 

This  genus  should  have  followed  larnna: 
I  now  introduce  it  for  the  purpose  of  re- 
ferring to  odontaspis,  (figs.  86  and  87,) 
which  appear  to  belong  to  this  genus  ra- 
ther than  lamna.  So,  also,  figs.  88  and 
89,  which  are  destitute  of  basal  denticles : 
but  the  cutting  edge  of  the  crown  extends 
over  the  fangs  and  is  slightly  expanded 
on  this  part  of  the  tooth;  it  preserves  also 
its  cutting  edge.  Eocene  of  Craven  county. 
I  have  no  facilities  at  hand  which  en- 
able me  to  make  a  correct  reference  of  the 
eocene  odontolites,  and  have  to  trust  to 


242  NORTH-CAROLINA   GEOLOGICAL   SURVEY. 

my  memory  in  making  the  references  to  the  genera  to  which 
they  belong. 

CARCHARODON. — (Fig.  90.) 

NOTE. — The  annexed  figure  of  a  tooth,  which  may 
probably  be  referred  to  this  genus,  is  confined  to 
the  eocene  of  Craven  county.  I  have  been  unable 
to  refer  it  to  a  species  already  made  known. 


SUB    ORDER. THE   RAYS. 

The  rays  are  distinguished  from  sharks  proper,  by  the  flat- 
ness of  their  bodies.  There  are  several  species  in  the  sea 
bordering  the  coast  of  ^Torth- Carolina,  one  of  which  is  known 
by  the  name  of  sting  ray.  The  rays  form  three  families:  1, 
the  pristides,  familiarly  known  as  the  saw  fishes,  whose  muz- 
zles are  elongated  into  a  flat  long  extension,  armed  on  each 
margin  by  pointed  teeth;  2,  rajides,  or  rays,  whose  muzzle  is 
simple,  but  whose  tails  are  not  armed  with  a  sting;  3,  the 
mylliobatides,  comprehending  those  rays  whose  tails  are 
armed  with  a  sting.  The  remains  of  the  latter  family  are 
known  in  the  tertiary  and  cretaceous  of  North-Carolina. 
Their  teeth  differ  in  form  from  those  of  the  sharks,  and  would 
scarcely  be  regarded  as  teeth  at  all,  were  it  not  for  their  oc- 
currence in  the  living  species  upon  the  coast.  They  are 
placed  in  the  mouth  in  the  form  of  a  pavement,  and  occupy 
the  areas  within  the  mouth  of  both  jaws.  They  differ  in  form 
from  the  pycnodonts  in  being  angular.  They  are  employed 
in  crushing  hard  bodies,  as  the  shells  of  the  inolusca.  Their 
mouths  are  placed  below,  and  well  situated  for  seizing  the 
animals  upon  which  they  feed. 
' 

FAMILY    PRISTIDES. 

Fish  which  have  a  prolonged,  bony  muzzle,  armed  with  a 
plain  horizontal  series  of  teeth  upon  each  margin. 

GENUS  PRISTIS. — (Fig.  93.) 

Single   teeth  broken   from  the  flat  plate  near  its  junction 
have  been  found  in  the  superior  layer  of  the  eocene  in  Cra- 


NORTH-CAROLINA   GEOLOGICAL   SURVEY. 


243 


ven  county.     One  margin  is  grooved  the  whole  length,  and 
straight,  the  other  is  curved  and  grooved  only  at 
FIG.  93.        base.     Figure  the  natural  size.     I  have  also  found 
smaller  ones,  which  belong  apparently  to  the  same 
species. 

FAMILY   MILIOBATIDES. 

Rays  whose  tails  are  provided  with  serratine 

stings. 

GENUS    MILIOBATIS. 

Sting  dentated  upon  one  margin.     No  stings  of 
this  kind  have  as  yet  been  met  with. 

GENUS    TRYGON. 

Sting  with  both  margins  dentated. 


TRYGON    CAROLINENSIS. N.    S.       (FigS.    91    &    92.) 

Teeth  in  mosaic,  the  ends  angular,  they 
being  bounded  by  six  lateral  planes. 

Sting  serrate,  (Figs.  94  &  95,)  grooved 
longitudinally,  rounded  on  one  side.  Fig. 


FIQ.  91. 


FIG.  92. 


FIG.  94. 


FIG.  95. 


95  shows  the  form  of  a  tranverse  section. 
These  specimens  were  found  in  the  upper  part  of  the  eocene 
marl  in  Craven  county,  and  as  the  teeth  and  stings  were  found 
in  proximity,  it  is  inferred  that  they  belonged  to  one  specie. 

CLASS    GANOIDEA. FAMILY    PYCNODON1TDAE. 

This  family  possess  teeth  of  a  cylindrical  form,  and  which 
are  arranged  upon  both  planes  of  the  jaws  in  the  form  of  a 
pavement.  The  longer  axis  lies  across  the  mouth  from  side 
to  side,  but  set  in  rows  arranged  from  before  backwards.  The 
middle  rows  contain  the  longest  teeth,  and  they  diminish  in 
length  towards  the  sides  of  the  mouth.  An  idea  of  this  ar- 


244  NORTH-CAROLINA   GEOLOGICAL   SURVEY. 

rangement  may  be  obtained  by  an  inspection  of  the  mouth 

of  the  mylliobates,  the   common  sting  ray  of  the 

FIG.  96.    coask     jn  this  fish  the  teeth  are  set  also  in  pavement, 

tbut  they  are  not  angular.  But  the  teeth  in  the 
Pycnodonts  are  not  placed  with  so  much  regularity 
as  in  the  Myliobatides. 
Fig.  96  is  figure  of  a  tooth  belonging  to  the  back 
part  of  one  of  the  middle  rows  of  the  pavement,  or 
mosaic.  It  may  be  called  Pycnodus  Carolinensis. 

The  teeth  of  this  species  of  fish  occur  in  the  miocene 
associated  with  those  belonging  to  the  genera  galeocerdo  and 
lamna.     The  family  of  pycnodonts  began  their  career 
97'      in  the  Permian,  but  were  the  most  numerous  in  the 
Jurassic  period. 

Another  species  of  pycnodont  is  represented  by  its 
tooth  in  fig.  97,  which  appears  to  be  much  less  com- 
mon than  the  preceding. 

SCALE   OF   A    GANOID. (Fig.    98.) 

A  single  scale  (fig.  98,)  was  found  in  the  miocene  upon  the 
Cape  Fear.     The  fish  was  closely  related  to  the  gar-pike,  (le- 

pidosteus,)  of  most  of  the  Ame- 
rican rivers.  The  scale  occupied 
a  position  in  the  first  row  of 
scales  back  of  the  head.  The 
fish  of  this  class  had  already  be- 
come rare  at  the  commencement 
of  this  epoch.  The  gar-pike  is  the  only  surviving  one  of  this 
family  in  the  American  waters. 

CLASS  CYCLOIDEA. — (Figs.  99,  100.) 

The  annexed  figures  represent  a  pe- 

,»/•.-,          ,       i«  , 
culiar  form  ot  nsn  teeth,  which  are  quite 

common  in  some  of  the  marl  beds  in 
Edgecombe  county.  They  were  attach- 
ed by  ligament,  and  probably  occupied 
a  position  in  the  throat. 


SORTH-CABOLINA   GEOLOGICAL   SURVEY, 


24:5 


CHAPTER  XVIII. 


• 

MOLLUSCA.       CLASS — CEPHALOPODA. 

Tills  class  embraces  those  mollusca,  whose  locomotive  or- 
gans are  attached  to  the  head.  They  have  the  form  of  mus- 
cular arms  or  tentacles.  Besides  the  arms  surrounding  the 
head,  they  have  fins  and  an  apparatus  by  which  they  can  pro- 
pel themselves  through  the  water  by  its  ejection  in  a  stream. 

Some  are  covered  by  a  shell,  coiled  in  a  vertical  plane,  'as 
the  nautilus ;  others  are  naked  or  destitute  of  an  external 
shell,  but  have  an  internal  one,  which  varies  much  in  form  in 
the  different  families. 

Their  eyes  are  well  developed  and  their  mouths  are  provi- 
ded with  jaws  somewhat  similar  to  the  mandibles  of  a  bird. 
They  are  predatory  and  live  on  fish,  crabs  and  shell  fish. 

The  most  remarkable  part  of  the  apparatus  by  which  they 
seize  their  prey,  are  the  circular  discs  arranged  on  the  under 
side  of  their  arms,  by  which  they  are  enabled  to  produce  in- 
stantaneously a  vacuum  when  applied  to  the  surface  of  a  fish 
or  a  slightly  yielding  body.  By  this  arrangement  they  are 
able  to  seize  and  hold  most  securely  their  captives,  and  de- 
vour them  at  leisure.  As  a  means  of  escape  from  enemies 
more  powerful  than  themselves,  they  are  provided  with  a  bag 
or  sac  filled  with  a  dark  fluid  which  they  can  eject  at  will,  and 
thereby  discolor  the  surrounding  water  and  escape  unseen. 

This  sac  is  called  the  ink-bag,  and  the  liquid  is  employed 
for  the  manufacture  of  the  India  ink.  Even  the  consolidated 
fluid  in  the  fossil  ink-bags  is  used  for  this  purpose. 

This  class  is  a  large  one,  and  the  species  which  compose  it 
are  found  in  all  seas.  They  were  also  extremely  numerous 
in  ancient  times,  and  their  hard  parts  as  external  and  internal 
shells  are  preserved  as  relics  of  extinct  races.  One  of  the 
most  common  fossils  of  the  green  sand  is  the  Belemnite,* 
which  is  an  internal  shell,  though  its  form  is  quite  unlike  one. 

*  From  belemnon,  a  dart. 


246 


NORTH-CAROLINA   GEOLOGICAL   SURVEY. 


FIG.  101. 


BELEMNITELLA   AMERICANA. (Fig.    101.) 

The  belemnitella  is  sub-cylindrical  and  tapering  to  a  point 
from  its  base.  The  sides  are  marked  by  numerous  ramose 
furrows,  though  they  are  arranged  without  much  order,  and 
being  crowded  they  give  the  surface  a  granulated  appearance. 
The  base  has  a  fissure  which  extends  through  the  wall  to 
a  conical  chamber.  On  the  back,  there  is  an  ele- 
vated convex  surface,  narrow  toward  the-  base, 
but  widens  towards  the  apex,  where  it  is  lost. 

This  genus  presents  a  great  variety  in  form  and 
size ;  but  the  foregoing  characters  are  its  constant 
characteristics.     It  occurs    at  Black  Rock   and 
Rocky  Point,  and  is  one  of  the  characteristic  fos- 
sils of   the  green  sand.      It  is 
found  also  in  the  miocene  beds, 
but  is  there  by  accident. 


FIG.  103.      Fia.  102. 


Fio.  105.    FIG.  104. 


BELEMNITELLA   COMHRESSA. N.    8. 

(Fig.  102.) 

Shell  slender,  transverse  sec- 
tion elliptical  at  base,  and  it  be- 
comes gradually  more  flattened 
to  its  apex ;  the  fissure  of  the 
base  is  short ;  surface  uneven 
and  somewhat  irregular.  This 
species  is  entirely  destitute  of 
the  granulations,  or  the  convex 
surface  of  the  preceding  species. 
The  green  sand  of  North-Caro- 
lina is  poor  in  cephalopods.  I 
have  not  yet  observed  either  an  ammonite 
or  nautilus,  though  they  occur  sparingly  in 
the  eocene. 

In  the  eocene  of  Craven  county  I  found 
numerous  specimens  of  the  bony  or  horny 
cores  of  the  jaws  of  cephalopods.  I  have 
not  been  able  to  determine  the  family  to 
which  they  belong.  Fig.  104  represents  their 
form  and  size.  They  occur  only  in  the  up- 


NORTH-CAROLINA    GEOLOGICAL    SURVEY.  247 

per  part  of  the  formation  associated  with  sharks'  teeth,  and 
teeth  and  stings  of  one  or  two  species  of  ray. 

CLASS    GASTEROPODA. FAMILY    MURICIDAE. 

The  muricidae  are  generally  readily  distinguished  by  their 
roughness  occasioned  by  the  periodical  expansion  of  its  lip. 
These  being  permanent,  the  shell  is  strongly  marked  by  the 
rough  shelly  expansions  along  the  lines  of  growth,  as  in  the 
murex.  The  shell  preserves  its  spiral  form ;  the  outer  lip  is 
entire  behind,  and  the  front  prolonged  in  a  straight  canal. 
The  eyes  of  this  family  are  sessile  and  seated  on  tentacles ; 
the  animal  has  a  broad  foot. 


GENUS   MUREX. LINN.  ROCK    SHELL. 

The  shell  is  roughened,  or  winged  with  the  periodical  ex- 
pansions of  its  lip,  which  are  permanent  after  it  has  advanced 
to  a  mature  state. 

MUREX     UMBRIFER — CON. CERASTOMA     UMBRIFER TOUMEY      AND 

HOLMES FOSSILS    SOUTH-CAROLINA   FROM  CON.    MSS. (Fig.  104a. ) 

FIG.  io4a.  Shell  fusiform  ;  whirl? 

subcari nated,  or  angulat- 
ed  and  provided  with 
six  foliated  and  rather 
broad  renexed  lamina, 
spirally  arranged.  Mi- 
ocene Cape  Fear  River. 


MUREX   GLOBOSA. (Fig.  105  A.) 

Shell  rather  globose,  or  obtusely  fusiform,  and  with  four 
principal  varices ;  intermediate  ones  irregular  and  spirally, 
traversed  by  many  angular  ridges,  body  whirl  inflated,  aper- 
ture oval,  peristome  continuous,  and  extended  posteriorly  on 
the  body  whirl,  forming  an  angulated  canal ;  outer  lip  ridged 
within  and  crenulated  on  the  margin ;  collumela  lip  ridged. 


248 


NORTH-CAROLINA   GEOLOGICAL   SURVEY. 


and  one  ridge  at  the  posterior  angle ;  beak  reflexed.     Mio- 
cene of  the  Cape  Fear  River ;  half  the  natural  size. 


FIG.  105  A. 


FIG.  106. 


MUREX   SEXCOSTATA. — (Fig.  106.) 

Shell  fusiform,  with  three  spinons  varices,  and  traversed 
spirally  by  angular  ridges.  Canal  closed 
beak  slightly  reflexed.  The  body  whirl  has 
six  ridges  or  ribs,  with  an  intermediate 
lesser  ridge.  Shell  imperfect, 

BUSICON  CARICA,  CON — PYRULA  CARICA,  GOUJLD, 
FULGUR  CARICA,  CON. 

This  shell    is    pyriform,  swollen,    thick 
and  heavy.     The  outside  is  ornamented  by 
transverse  striae,  and  also  with  compressed 
tubercles,  which  stand  upon  the  most  prom- 
inent part  of  the  body  whirl.     The  outer 
lip  is  simple  and  sharp,  pillar  lip  flexuous 
and  concave  above. 
The  suture  of  this  species  is  not  channelled,  neither  has  it 
a  turrited  spire.     It  is  one  of  the  most  common  fossils  of  cer- 


NORTH-CAROLINA   GEOLOGICAL    SURVEY. 


249 


tain  marl  beds  upon  the  Cape  Fear  river,  but  is  less  common 
upon  the  Neuse.  It  is  one  of  the  common  living  species  upon 
the  Atlantic  coast. 


FIG.  107. 


BUSICON  PERVERSUM,  CCN PYRULA  PERVERSA,  REEVE. (Fig.  107.) 

This  shell  is  also  pear-shaped  and 
swollen.  The  prominent  part  of  the 
whirl  is  ornamented  with  tubercles,  and 
is  also  coronated  ;  the  whirls  are  turned 
to  the  left. 

It  is  common  upon  the  coast.  It  is 
very  abundant  in  a  post  pliocene  de- 
posit at  Beaufort,  but  is  also  often  met 
with  upon  the  Cape  Fear. 


BUSICON  CANICULATUM,  CON. — PYRULA  CAN- 
ICULATA,      GOULD. FULGUR      CANICULA- 
TUM, CON. — (Fig.  108.) 
Shell  somewhat  pear-shaped,  spire  de- 
pressed, and  ornamented  with  revolving 
lines ;  body  whirl  swollen  ; 
canal   long   and    straight; 
suture    channelled.     Com- 
mon on  the  coast,  and  ra- 
ther common,  also,  in  the 
miocene. 


FIQ.  108. 


PYRULA  CAROLINEUSIS — TTK>- 
MEY  AND  HOLMEL, — H.  TER- 
TIARY FOSSILS  OF  SOUTH- 


CAROLINA. 

Description  :  "  Shell,  pear-shaped ;  spire  short,  depressed ; 
suture  profoundly  canaliculated,  margined  by  the  obtuse  ca- 
rina  at  the  angle  of  the  whirl ;  body  whirl  truncated  above ; 
angular  whirls  of  the  spire  angulated  in  the  middle,  and  in- 


Fossils  of  South-Carolina, — Tuomey  and  Holmes,  p.  145-'6.  , 


250 


NORTH-CAROLINA    GEOLOGICAL    SURVEY. 


clined  slightly  to  the  summit,  having  fine  revolving  lines  in- 
distinct, but  prominent  and  waved  on  the  base  of  the  body- 
whirl  ;  canal  long  and  tapering."  Miocene  marl,  Cape  Fear. 

PYRULA     SPIRATA,    LAM. FULGUR   PYRTJLOIDES.     SAY. FTTLGUR 

PYRUM,    CON. 

Shell  pyriform ;  spire  depressed  obtuse ;  whirls  flattened, 
and  traversed  by  numerous  revolving  lines  ;  suture  canicula- 
ted.  It  still  lives  upon  the  coast,  and  is  common  in  the  post 
pleiocene  of  North-Carolina. 


PIG.  109. 


FIG.  110. 


PYRULA    RETICTJLATA LAM SYCOTYPUS    RETICU- 

LATUS.    (Fig.    109.) 

Shell  thin,  cancellated ;  spire  very  short  > 
surface  marked  by  revolving  lines,  which  are 
intersected  by  longitudinal  ones,  giving  the 
shell  its  reticulated  appearance  or  character. 
Occurs  both  in  the  miocene  and  post  pleio- 
cene beds,  particularly  at  Beaufort.  It  is  of- 
ten much  larger  than  the  figure. 

FUSTJS  LAM. 

The  genus  Fusus  is  distinguished  by 
its  straight  open  canal  and  the  ab- 
sence of  plaits  upon  the  columella. 


FUSUS    QUADRICOSTATUS. (Fig.  110.) 

Shell  thick,  spire  depressed,  body 
whirl,  inflated  and  ornamented  by 
four  elevated  equidistant  spiral  belts, 
umbilicus  large. — Newborn. 


FUSUS    EQUALIS. N.  S. (Fig.    111.) 

Shell  thick,  spire  rather  short,  conical ;  whirls  eight  round- 
ed and  somewhat  ventricose,  and  ornamented  by  numerous 


••: 


NORTH-CAROLINA   GEOLOGICAL   SURVEY. 


251 


FIG.  111. 


FIG.  Ill  A. 


spiral  subequal  lines,  coarser  and 
more  distant  upon  the  back  and  ros- 
trum ;  aperture  and  rostrum  rather 
less  than  twice  the  length  of  the 
spire ;  outer  lip  ridged  internally ; 
pillar  lip  spirally  ridged.  Miocene  of 
Cape  Fear  River. 

FUSUS    EXILIS. (Fig.  Ill  A.) 

Shell  fusiform,  spire 
elongated,  composed 
of  seven  whirls,  orna- 
mented by  revolving 
striae  and  longitudi- 
nal ribs;  aperture  one 
half  the  length  of  the 
shell. 


FUSU8   LAMELLOSUS. N.  S. (Fig  112.) 

Shell  small)  fusiform ;  spire  composed  of  five 
or  six  whirls,  ornamented  with  ten  strong  scalari- 
form  ribs,  each  rib  on  the  body  is  composed 
of  three  sharp  crenulated  plates,  the  one  in  the 
middle  being  the  largest. 


FUSUS   MONILIFORMIS. N.  S. (Fig.  123.) 

Shell  small ;  whirls  four,  ornamented  with  raised  beaded 
spiral  lines,  between  which  there  are  lines  nearly  sim- 
ple ;  spire  rather  shorter  than  the  aperture ;  aperture  oval ; 
canal  short ;  the  two  upper  whirls  are  smooth.  Miocene  of 
Cape  Fear.  Eare. 

FASCIOLARIA. 

This  genus  is  characterized  by  its  elongated  fusiform  shape, 
its  round  or  angular  whirls,  open  canal,  and  its  folds  upon 
columellar  lip,  which  is  more  or  less  tortuous.  The  folds  upon 
the  lip  are  quite  oblique,  and  two  or  three  in  number. 


FIG.  112. 


252 


NORTH-CAROLINA    GEOLOGICAL    SURVEY. 


FA8CIOLARIA    DISTANS. — LAM.       (Fig.  113.) 


FIG.  113. 


PIG.  114. 


This  shell  at  first  sight  appears  smooth, 
but  a  careful  inspection  shows  that  it  is 
finely  striated  longitudinally  ;  its  spire  is 
composed  of  six  or  seven  convex  or  pro- 
minent whirls,  and  its  pillar  has  but  one 
plait. 

It-is  a  common  shell  upon  the-  coast, 
and  in  the  post  pleiocene  at  Beaufort. 
but  not  uncommon  in  the  miocene  of 
Cape  Fear. 


FASCIOLARIA   ELEGANS.  —  N.  S.       (Fig. 

Shell  elongated,  acute;  whirls  eight 
rounded,  ornamented  with  wide,  and 
finely  striated  ribs  ;  striae  transverse  to 
the  ribs,  or  longitudinal  ;  ribs  of  the  body 
whirl,  about  fifteen,  the  middle  of  the 
body-whirl  upon  the  outer  lip,  the  four 
widest  ribs  alternate  with  three  narrow 
ones  ;  plaits  three,  concealed  within  the 
pillar  lip;  spire  longer  than  the  aper- 
ture. 

This  shell  is  rare  in  the  miocene  of 
North-Carolina.  It  would  pass  for  fusus 
if  the  pillar  lip  was  not  examined  just 
within  the  aperture,  the  plaits  reaching 
only  to  its  edge,  but  they  are  strong  and 
well  developed  through  its  entire  length. 

It  is  possible  this  shell  may  have  been 
previously  described,  but  its  broad,  flat 
and  very  prominent  ribs  are  so  peculiar, 
that  if  observed  and  described,  it  could 
scarcely  escape  detection.  Figure  half 
the  natural  size. 


NORTH-CAROLINA    GEOLOGICAL    SUKVEY. 


253 


FIG.  115. 


FASCIOLARIA    SPARROWI. N.    S.       (Fig.    115.) 

Shell  rather  thick,  turbinate  ; 
whirls  six  or  seven  rounded,  or- 
namented with  spiral,  and  rather 
rounded  ribs;  ribs  of  the  body- 
whirl,  about  ten,  striated  longitu- 
dinally, but  obliquely  striated  on 
the  upper  part  of  the  whirl ;  plaits, 
three  upon  the  pillar  lip ;  the  ribs 
alternate,  being  coarser  and  finer 
for  the  ribs  which  belong  strictly 
to  the  aperture ;  aperture  larger 
than  the  spire. 

This  species  is  quite  distinct 
from  the  former,  the  ribs  are  less 
numerous,  flatter,  and  the  striae 
are  partly  oblique  and  partly  lon- 
gitudinal, or  in  the  direction  of 
the  axis  of  the  shell.  The  five 
upper  whirls  have  varices  in  both 
species.  Rare  in  the  miocene 
marl  bed  of  Mrs.  Purdys,  Bladen 
county.  One-half  the  size. 
This  fine  fossil  is  dedicated  to  Thos.  Sparrow,  Esq.,  of  Beau- 
fort county. 


FASCIOLARIA   ALTERNATA. — N.    S. 

Shell  rather  small,  but  thick  turbinate  ;  whirls  six  or  seven 
slightly  inflated,  body  whirl  elongated  and  ornamented  with 
strong  spiral  lines,  and  with  fine  ones  between,  but  which 
are  frequently  obsolete.  All  the  whirls  are  tuberculated. 
Spire  shorter  than  the  aperture  Plaits  two. 

FASCIOLARIA    NODULOSA. N.     S.        (Fig.    116.) 

Shell  rather  thick,  whirls  about  seven,    all   nodulose  or 


254 


NORTH-CAROLINA   GEOLOGICAL   SURVEY. 


ornamented   with   varices    and    spiral  subequal  striae.     Mi- 
ocene of  the  Cape  Fear 
river. 

FASCIOLAKIA  ACUTA.— N. 

s.  (Fig.  117.) 
Shell  elongated,  a- 
ctite,  whirls  about  sev- 
en, ornamented  by 
spiral  subequal  ribs, 
with  obsolete  ones  be- 
tween them,  six  upper 
whirls  have  also  equal 
varices ;  longitudinal 
striae  very  fine,  aper- 
ture shorter  than  the 
spire.  Miocene  of  the 
Gape  Fear  river. 

CANCELLARIA    CAROLINENSIS. N.    S.       (Fig.    118.) 

Shell  thick,  angulated,  whirls  few,  oblique,  carinated  and 
ornamented  by  two'subspinous  spiral  bands,  body  whirl  trans- 
versely, rugose  towards  the  aperture, 
rugae  subcrenulated,  aperture  trian- 
gular, and  acute  in  front,  umbillicus 
large,  open,  and  funnel  shaped. 

I  should  have  hesitated  to  have 
placed  this  interesting  fossil  in  the 
genus  cancellaria  were  it  not  that  a 
closely  allied  species,  the  C.  acutan- 
gulata,  Faujas,  is  thus  referred  by 
high  authority.  The  C.  acutangulata 
is  one  of  the  characteristic  fossils  of 
the  miocene  beds  of  Dax,  south  of  France.  Its  surface  is 
is  ornamented  like  a  cancellaria,  but  the  aperture  in  both  the 
Dax  and  North-Carolina  specimens  is  triangular,  but  both 
have  rather  obsolete  folds  upon  the  pillar  lip  ;  they  are  rather 
more  obscure  in  our  specimen  than  in  that  from  Dax.  The 


FIG.  118. 


NORTH-CAROLINA   GEOLOGICAL   SURVEY. 


255 


existence  of  this  interesting  fossil  in  North-Carolina  proves 
the  close  analogy  between  the  miocene  of  France  and  that  of 
the  southern  States,  and  it  seems  that  the  new  species  really 
replaces  the  C.  acutangulata  in  our  miocene  beds. 

I  am  indebted  to  I.  Lea,  Esq.,  of  Philadelphia,  for  speci- 
mens for  comparison. 

It  occurs  at  Mr.  Flowers'  marl  bed  on  the  Cape  Fear. 
Bladen  county. 

CANCELLARIA   EETICULATA. (Fig.    119.) 

Shell  thick,  ovate,  spire  acute,  whirls  about  six,  and  angulat- 
ed  and  ornamented  by  prominent,  longitudinal  and  revolving 
ridges,  which  produce  a  cancellated  surface.     Columulla  with 
FIG.  119.  several  strong  oblique    sharp  folds  ;  outer 

lip  traversely  ridged  within. 


FIG.  120. 


RANELLA   CAUDATA. (Fig    120.) 

Shell  turbinate,  winged ; 
whirls  four  or  five,  angulat- 
ed  and  strongly  ridged  longi- 
tudinally ;  surface  traversed 
by  lesser  revolving  ridges. 
Two  opposite  ridges  are  pro- 
duced more  than  the  others, 
one  of  which  forms  the  margin  of  the  outer  lip  ;  canal  long 
and  straight.  Common  on  the  coast,  and  rather  rare  in  the 
miocene  of  North-Carolina. 


FIG.  121. 


FAMILY  BUCCTNIDAE. BUCCINTJM   MULTIRUGA- 

TUM. CON.       (Fig.    121.) 

Shell  thick,  ovate;  spire  composed  of 
five  whirls,  marked  with  deep  impressed 
revolving  lines ;  apex  rather  obtuse ;  col- 
umella,  with  a  strong  fold  at,  base  and  a 
slight  prominence  at  the  base  of  the  body 
whirl ;  bicarinate^  aperture  greater  than 
half  the  length  of  the  shell.  Miocene  of 
Cape  Fear  River. 


256 


NORTH-CAROLINA    GEOLOGICAL    SURVEY. 


FIG.  122 


BUOCINUM  PORCINTJM. — SAY. (Fig.  122.) 

Shell  thick,  fusiform ;  spire  composed  of  five  or  six  whirls* 
ribbed  longitudinally,  and  marked  with  num- 
erous raised  revolving  lines ;  beak  short  and 
only  slightly  reflexed ;  outer  lip  marked  with- 
in by  numerous  ridges.  Buccinum  vibex, 
buccinum  trivittatum  and  obsoletum  are  as- 
sociated with  the  preceding  species.  - 


BUCCINUM   MULTILINEATUM. N.    S. — (Fig.    124.) 

Shell  small  and  thick,  turreted  ;  whirls  six. 
and  marked  by  many  impressed  spiral  lines, 
between  which  there  are  also  many  narrow  flat  spiral  bands :. 


FIG.  123. 


FIG.  127. 


FIG.  170. 


FIG.  125. 


FIG.  124. 


whirls  furnished  with  strong  longitudinal  ribs,  interrupted  at 
the  suture,  aperture,  ovate  and  less  than  half  the  length  of  the 
shell ;  canal  short  and  directed  backwards ;  the  body  whirl1 
has  about  thirteen  ribs.  Rare  in  the  miocene  of  Cape  Fear. 

BUCCINUM   MONILIFORMIS. N.  S. (Fig.    125.) 

Shell  small,  thick  and  robust,  rugose  ;  whirls  about  six,  and 
ornamented  with  moniliform  ribs.  This  shell,  though  small, 
has  all  the  marks  of  being  mature.  The  flat  spiral  bands, 
which  as  they  cross  the  ribs  and  give  them  a  beaded  appear- 
ance, are  strongly  marked  on  all  the  whirls.  Rare  in  the 
miocene  of  Cape  Fear. 


NORTH-CAKOLINA   GEOLOGICAL   SURVEY. 


257 


FIG.  126. 


FIG.  128. 


BUCCINUM   BIDENTATUM. — (Fig.  126.) 

Shell  quite  small,  thick,  robust;  whirls  about  five,  two 
upper  smooth,  the  others  are  ornamented  with 
ribs  and  spiral  bands;  aperture  oval,  acute 
behind,  outer  lip  furnished  with  two  rather 
prominent  teeth,  or  short  ridges ;  canal  wide  and 
very  short. 

BUCCINUM   OBSOLETUM. — (Fig  127.) 

Surface  granulated;    spire  shorter  than   the   body.     The 
common  species  of  the  coast ;  is  rare  in  the  miocene  of  North- 
Carolina.      The   specimen  figured 
was  a  young  shell,  and  broken. 


GALEODIA  HODGII. — CON.      (Fig.  128.) 

Shell  rather  thick  ;  elliptical,  ob- 
tuse ;  whirls  about  five,  inflated, 
and  ornamented  with  numerous 
fine  spiral  lines,  which  are  quite 
prominent  at  base ;  these,  with  the 
fine  lines  of  growth,  give  the  sur- 
face a  cancellated  appearance ; 
collumellar  lip  marked  with  many 
irregular  plicae  ;  aperture  nearly 
twice  the  length  of  the  spire.  Mi- 
ocene of  Cape  Fear. 


TEEEBEA   DISLOCATUM  J    SAY. — ACUS    DISLOCATUM. 

Shell  thick,  elongated,  acute  ;  whirls  many,  slightly  convex, 
upper  portion  constricted,  forming  a  revolving  band,  parallel 
to  which,  there  are  numerous  spiral  raised  lines;  lines  of 
growth  longitudinal  and  conspicuous,  which  give  to  the  sur- 
face a  reticulated  appearance. 

Common  in  the  miocene  marls  of  North  Carolina. 
18 


258 


NORTH-CAROLINA   GEOLOGICAL   SURVEY. 


TEEEBRA   TJNILINEATA  ;    TUOMEY    AND   HOLMES FOSSILS  OF    SOUTH- 
CAROLINA. — (Fig.   129.) 


FIG.  129. 


Shell  thick,  elongate  bands  alternate,  acute, 
tapering  gradually  to  a  point ;  whirls  many, 
seventeen  or  eighteen,  and  ornamented  by 
revolving  impressed  lines,  and  passing  just 
above  the  middle  of  the  whirl ;  the  upper 
part  of  the  spire  is  also  marked  by  short 
longitudinal  ribs,  which  are  interrupted  by 
spiral  lines.  Oblique  lines  of  growth  are 
usually  conspicuous.  In  old  specimens, 
the  ribs  are  obsolete. 

Common  in  the  miocene  of  North-Caro- 
lina. 


TEREBRA   NEGLECTA. N.    S. 

Shell  terete ;  spire  composed  of  many 
whirls,  traversed  spirally  by  a  deeply 
impressed  line,  dividing  it  into  two  un- 
equal parts ;  the  lower  has  three  or  four  interrupted  spiral 
lines,  the  upper,  none.  The  ribs  of  the  upper  part  are  more 
obtuse  than  the  lower,  and  die  out  before  they  reach  the  di- 
viding impressed  line  ;  in  the  lower,  they  cross  it  from  line  to 
line. 

In  this  species,  the  revolving  lines  are  lewer  than  in  the 
T.  dislocatum,  and  in  the  latter,  they  are  common  to  both 
parts  of  the  whirl.  In  the  unilineata,  there  is  but  one  dis- 
tinct revolving  line. 


Fir,.  120  a. 


FIG.  181. 


DOLIUM    OCTOCOSTATUM. N.    S.       (Fig.  129  a.) 

Shell  small,  thin  ;  whirls  three,  infla- 
ted ;  body-whirl  ornamented  with  eight 
spiral  ribs,  connected  by  short  bars, 
peristome  interrupted ;  aperture  ovate  ; 
umbilicus  small,  open  ;  outer  lip  crenu- 
latedi 


NORTH-CAROLINA  GEOLOGICAL  SURVEY.  259 

OLIVA;  LAM. — STREPHONA;   BROWN. 

The  olives  are  shells  of  great  beauty,  being  highly  polished 
and  covered  with  a  porcellanous  enamel,  the  surface  of  which 
is  marked  by  spots  and  bands  of  a  great  variety  of  colors. 
The  shell  is  cylindrical,  dense  and  heavy ;  the  spire  is  short, 
with  channelled  sutures,  and  the  aperture  long  and  narrow ; 
the  anterior  part  is  notched ;  the  columella  is  callous  and  stri- 
ated obliquely.  The  body- whirl  is  furrowed  near  the  base. 
The  olives  are  the  inhabitants  of  warm  climates,  and  are  very 
active. 

OLIVA  LITERATA. SAY.       (Fig.  130.) 

Shell  cylindrical,  thick  and  polished ;  spire  depressed ;  vo- 
lutions angular  and  channelled ;  apex  acute ;  outer  lip  sharp, 
inner  marked  with  numerous  sharp  folds;  aperture  linear,  in- 
cised above  and  notched  below. 

This  shell  is  very  common  in  many  of  the  miocene  marl 
beds  in  the  State.  It  is  also  living  and  common  on  the  coast. 
The  fossil  frequently  retains  the  polish  of  the  living  shell ;  the 
colors  have  disappeared. 

OLIVA   ANCILLARIAEFORMIS. LEA. 

Shell  small,  oval,  thick,  and  polished ;  spire  elevated,  acute ; 
suture  channelled  ;  aperture  narrow ;  inner  lip  thickened  by 
callus  and  marked  by  a  few  obscure  folds. 

The  foregoing  description  applies  to  a  small  oliva,  with  a 
large  amount  of  callus  upon  its  inner  lip ;  but  it  appears  to  be 
a  thicker  shell  than  the  one  to  which  I  have  referred  it.  It 
is  the  most  common  upon  the  Cape  Fear  river. 


OLIVA. 

An  oliva,  (fig.  131a,)  larger  than  the  preceding,  and  more 
cylindrical)  and  having  a  higher  spire,  is  occasionally  found  in 
the  miocene  beds  of  the  Cape  Fear.  It  has  six  whirls,  and 
the  folds  upon  the  inner  lip  extend  to  the  suture. 


OLIVA   CANALICULATA. LEA. 

With  this  addition  to  the  olives,  we  have  four  or  five  species 
belonging  to  the  miocene  period. 


260 


NORTH-CAROLINA   GEOLOGICAL   SURVEY. 


FAMILY    CYPREIDAE. 

The  shells  in  this  family  are  remarkable  for  their  forms, 
polish  and  beauty.  They  are  rolled  as  a  scroll,  and  are  cov- 
ered with  a  porcellanous  enamel.  The  spire  is  concealed,  the 
aperture  is  long  and  narrow,  and  the  outer  lip  is  inflexed  and 
thickened.  It  comprehends  the  beautiful,  spotted  and  banded 
shells  known  as  the  cowry. 

CYPRAEA   CAROLINENSIS. (Fig.    131.) 

FIG.  131.  Shell  ovate,  flattened  on  the 

side  of  the  aperture ;  outer  lip 
prominent  at  the  apex;  margins 
of  the  lips  ornamented  with  num. 
erous  plaits,  and  receding  from 
each  other,  beginning  at  the 
most  prominent  part  of  the  whirl. 
In  some  of  the  miocene  beds  it 
is  quite  common. 

CYPEEA   PEDICTJLUS. 

It  is  a  small  ovate  shell,  and 
transversely  ribbed,  and  with  a 
narrow  groove  along  the  back. 
I  have  not  yet  met  with  it  in  the 
marl  beds  of  this  State,  though 
it  appears  to  be  common  in  South 
Carolina. 


FIG.  132. 


MITKA   CAROLINENSIS. — (Fig.    132.) 

Shell  fusiform,  thick,  or  elongate,  and  tapering 
towards  each  extremity ;  whirls  slightly  convex, 
channeled  above,  and  traversed  by  numerous 
spiral  raised  lines  ;  columella  lip,  furnished  with 
numerous  oblique  plaits,  of  which  the  upper 
one 'is  the  strongest;  canal  wide  and  straight. 
Miocence  marl  of  North-Carolina.  The  shell 
is  often  found  much  larger  than  the  figure. 


NORTH-CAROLINA   GEOLOGICAL  SURVEY. 

MARGINELLA  OLIVAEFORMIS. PORCELLANA  OLIVAEFORMIS  I  TUOMEY 

&   HOLMES,    FOSSILS    OF   SOUTH- CAROLINA,    p.    131. — (Fig.  133.) 

"  Shell  elongated,  oval ;  spire  profoundly  obtuse ;  aperture 
linear ;  labrum,  (or  outer  lip)  tumid,  reflexed,  profusely  cre- 
nulated  within ;  columella  with  three  raised  plaits." 

With  this  description,  several  specimens  agree,  which  I 
have  found  in  the  marl  beds.  It  is,  however,  rare. 

FIG.  138.      FIG.  136.  FIG.  133. 


FIG.  139. 


FIG.  135.  FIG.  134. 

MARGINELLA   LIMATULA. — (Fig.  134.) 

Shell  ovate ;  spire  short ;  outer  lip  unequally  crenulated ; 
columella  lip  four  plaited ;  aperture  contracted  above  by  de- 
position of  callus. 

MARGINELLA    CONSTRICTA. — N.    S.      (Fig.    135.) 

Shell  polished,  cylindrical ;  spire  short ;  aperture  constricted 
in  the  middle  by  the  imbending  of  the  outer  lip ;  plaits  four 
crowded  at  the  base ;  margin  of  the  outer  lip  smooth. 

MARGINELLA   OVATA. — N.    S.       (Fig.  136.) 

Shell  ovate  ;  spire  much  depressed ;  aperture  uniform ; 
outer  lip  marked  with  numerous  crenulations  within ;  colu- 
mella with  six  or  seven  plaits,  the  upper  becoming  obsolete. 

MARGINELLA   INFLEXA. — N.    S.      (Fig.    13T.) 

Shell  oval ;  spire  somewhat  elevated  ;  obtuse  at  base ;  mar- 
gin of  the  outer  lip  inflexed  above  the  middle  ;  smooth  inside ; 
plaits  four,  and  very  prominent  upon  the  inner  lip.  Diners 
from  the  constricta  in  the  height  of  the  spire. 


262 


NORTH-CAROLINA   GEOLOGICAL   SURVEY. 


MARGINELLA   ELEVATA. N.    S.       (Fig.  238.) 

The  thickened  outer  lip  and  the  plaits  of  the  inner,  show 
this  to  belong  to  the  genus  marginella,  though  it  has  a  close 
resemblance  to  an  oliva  in  the  elevation  of  the  spire ;  whirls 
four. 

ERATO   LAEVIS? — (Fig.  139.) 

Shell  obtusely  ovate;  wide  at  the  base  of  the  spire; "spire 
depressed ;  both  lips  crenulate,  but  most  distinct  upon  the 
outer  lip ;  resembles  very  closely  a  marginella.  Miocene 
marl  of  Cape  Fear  river.  (Kare.) 

It  is  difficult  to  distinguish  this  from  the  English  species 
with  the  aid  only  of  figures.  It  may  be  indentical,  and  I  have- 
therefore  referred  it  to  the  English  species. 

FAMILY    YOLUTIDAE. 

The  volutes  have  a  thick,  short  ornamented  shell.  The 
spire  is  particularly  so,  and  it  is  also  provided  with  a  mamil- 
lated  apex.  Aperture  is  large  and  elongated,  and  the  coin- 
mella  has  several  plaits. 

VOLUTA   MUTABILIS. — CON. 

The  shell  is  fusiform  and  thick,  and  has  a  conical  spire  and 
a  papillated  apex ;  whirls,  convex  and  contracted  near  the 
sutures,  arid  the  two  principal  whirls  are  ornamented  with 
short  ribs;  lines  of  growth  distinct,  and  crossed  by  faint  re- 
volving lines ;  plaits,  two  and  rather  distant,  and  faint  indica- 
tions of  an  intermediate  one.  Found  in  the  miocene  of  the 
Cape  Fear  river. 

VOLUTA  TRENHOLMII:  TUOMEY  &  HOLMES,  FOSSILS  OF  SOUTH-CAR- 
OLINA, p.  128.— (Fig.  140.) 

"  Shell  fusiform,  ventricose  ;  whirls  compressed  above,  spi- 
rally and  transversely  striated  ;  striae  wrinkled  and  coarse  at 
base;  spire  short  and  sub-cancellated,  papillated;  aper- 
ture semi-lunar ;  outer  lip  acute,  smooth  within ;  columella 
lip  very  thin,  decumbent,  almost  obsolete,  semi-callous,  not 
distinguishable  from  the  body-whirl,  but  by  outline  and  color. 

* 


NORTH-CAROLINA   GEOLOGICAL   SURVEY. 


263 


Oolumellar   tumid,  tortuous;    obliquely  plaited  with   three 

folds." 


FIG.  140 


FIG.  141. 


VOLUTA   OBTUSA. — N.    S.       (Fig.  141.) 

Shell   fusiform,    contracted    above 
the  body-whirl,  and  forming  thereby 
a  sub-cylindrical  spire  ;  spire  obtuse 
apex  papillated  and  hook- 
ed ;    body-whirl    plaited 
longitudinal^  at  its  top  ; 
columellar  lip  furnished 
with  only  two  plaits. 

Mr.  Flowers   miocene 
marl,  Bladen  county. 


FAMILY    CONIDAE. 

As  the  name  implies,  the  shells  are  conical  from  the  great 
preponc^erance  of  the  body  whirl  over  the  short  depressed 
spire.  The  aperture  is  long  and  narrow,  and  the  outer  lip  is 
notched  near  its  suture. 

CONTJS    ADVERSARITJS — CON. (Fig.    142.) 

Shell  conical  and  turned  to  the  left ;  the  surface  is  marked 
by  revolving  lines ;  towards  the  face  of  the  pillar  lip  the  lines 
are  strong ;  whirls  of  the  spire  rather  concave ;  edges  eub- 
carinated  ;  labrum  sharp,  edge  convex,  and  forming  a  sinus 
near  the  suture.  Common  in  all  the  marl  beds  upon  the 
Neuse  and  Cape  Fear  rivers. 

CONUS   DILTJIVIANUS. (Fig.    143.) 

Shell  conical,  much  smaller  than  the  preceeding,  and  the 
whirls  are  turned  to  the  right ;  surface  markings  the  same ; 
the  revolving  lines  are  less  oblique  than  in  the  C.  adversarius. 


264 


NOKTH-CAROLINA   GEOLOGICAL   SUKVEY. 


They  are  associated  together  in  about  equal  numbers.  Neither 
species  are  found  in  older  beds. 


FIG.  142. 


FIG.  143. 


FIG.  130. 


FIG.  ISla. 


FIG.  144. 


PLEUKCTOMA     LUNATUM. LEA.       TURRIS      LUNATUM. FOSSILS     OF 

SOUTH-CAROLINA. (Fig.    144.) 

Shell   thick,  elongate,   acute,  subfusiform;    strongly   and 
obliquely  ribbed;  spire,  eight  whirled,  angulated  above  and 
ornamented  by  a  narrow  sutural  band. 

The  upper  part  of  the  whirls  are  construct- 
ed so  as  to  present  to  the  eye  a  narrow  spiral 
band.  Rather  common  in  the  marl  of  Cape 
Fear  river. 

PLEUROTOMA    LIMATULA. CON.       (Fig.  145.) 

Shell  rather  small,  sub-fusiform ;  spire  com- 
posed of  five  or  six  whirls  ;  whirls  constricted 
above  and  sub-angulated,  forming  a  sutural 
spiral  collar ;  ribs  oblique  and  coarse.  It  is 
about  one  inch  long. 

PLEUKOTOMA  COMMUNIS. — CON. 

Shell  small,  sub-fusiform ;  whirls  about  six,  indistinct ;  body- 
whirl  traversed  spirally  by  four  other  sharp  ridges. 


NORTH-CAROLINA   GEOLOGICAL   SURVEY.  265 

PLEUROTOMA  ELEGANS. — N.    S.      (Fig.    146.) 

Shell  small,  sub-turrited ;  whirls,  about  nine,  constricted 
above,  ornamented  by  numerous  longitudinal  ribs,  and  tra- 
versed by  many  fine  raised  spiral  lines,  which  become  very 
distinct  upon  the  pillar  lip. 

The  spiral  lines  are  very  regular  and  equi-distant.  The 
body  whirl  has  about  sixteen  ribs.  Figure  natural  size. 

FIG.  148.       FIG.  147.  FIG.  145.       FIG.  146. 


PLEUROTOMA   TUBERCULATA. N.    S.       (Fig.    147.) 

Shell  small,  thick,  sub-acute  ;  whirls,  seven  or  eight ;  apex 
sub-tuberculated,  constricted  above,  and  traversed  spirally  by 
rather  coarse  raised  lines ;  apex  papillated,  and  the  first  whirl 
is  spirally  lined,  and  without  tubercles  or  short  ribs.  It  is 
more  widely  constricted  than  the  preceding. 

PLEUROTOMA   FLEXUOSA. N.    S.       (Fig.    148.) 

Shell  small,  thick,  sub-turbinate ;  whirls,  seven  or  eight, 
and  ornamented  by  flexuose  ribs,  which  extend  across  the 
whirl;  ribs  alternating  with  those  of  the  adjacent  whirl. 
There  are  about  ten  ribs  belonging  to  the  body-whirl. 

FAMILY   NATICIDAE. 

The  genus  Natica  belongs  to  a  family  of  shells  which  is 
characterised  by  a  globular  form,  few  whirls,  or  a  low  and 
obtuse  spire,  a  semilunar  aperture,  an  acute  outer  lip,  and  an 
umbilicus  often  covered,  wholly  or  in  part,  by  a  thick  cal- 
lus. The  species  are  all  marine. 

NATICA   HEROS. — SAY.      Fig.    149. 

Shell  sub-globose,  spire  depressed,  whirls  four,  convex  ; 
lines  of  growth  obscure ;  aperture,  ovate ;  umbilicus  simple 
and  rather  large. 


266 


NORTH-CAROLINA   GEOLOGICAL   SURVEY. 


This  species  is  common  in  the  miocene  marl  of  North- 
Carolina.  It  is  also  living  upon  the  coast,  but  is  more  abun- 
dant, according  to  Dr.  Gould,  north  of  Cape  Cod  than  south 
of  it. 

NATICA   DUPLICATA. — SAY.      FlG.    150. 

Shell  thick,  ovate ;  spire  somewhat  prominent  and  pyrami- 
dal by  the  compression  of  the  whirls ;  and  surface  marked 
by  faint  revolving  lines;  the  lines  of  growth  more  distant; 
umbilicus  partially  closed  by  a  thick  dense  callus. 

FIGS.  150. 


NATICA. — (Fig.    151.) 

Shell  thick,  spire  depressed ;  umbilicus  perfectly  closed  by 
a  thick  rough  callus,  which  extends  to  the  angle  where  it  be- 
comes much  thickened ;  suture  distinct.  It  agrees  with  the 
clausa  in  part,  but  it  is  a  much  larger  shell,  being  one  inch 
and  eight-tenths  in  diameter.  Fossils  answering  in  size  to 

the  clausa  exist  in  the  miocene  marl  on  the   Cape  Fear 
. 
river. 


NORTH-CAROLINA   GEOLOGICAL   SURVEY. 


267 


NATICA   CANRENA. Fig.    152. 

Shell  rather  thick,  lines  of  growth  surrounding  the  spire, 
very  distinct,  resembling  wrinkles ;  umbilicus  partially  closed 
with  callus. 

Occurs  frequently  in  the  miocene  marl  of  North-Carolina. 


FIG.  151. 


FIG.  153. 


FIG.  149. 


NATICA   FRAGILIS. — (Fig.    153. 

Shell  small,  surface  marked  by  revolving  lines  and  lines  of 
growth,  which  give  it  a  cancellated  appearance. 


268 


NORTH-CAROLINA  GEOLOGICAL   SURVEY. 


FAMILY   PYRAMIDELLIDAE. 

This  family,  when  restricted  to  existing  species,  embraces 
shells  of  a  small  size,  and  which  are  spiral  slender,  pointed 
and  turrited ;  aperture  small,  and  the  columella  has  one  or 
more  prominent  plaits.  Shells  which,  in  form,  bear  a  very 
close  resemblance  to  this  family,  are  found  in  very  ancient 
rocks,  but  which,  in  comparison  with  those  of  the  present  day, 
were  of  a  gigantic  size. 

PYRAMIDELLA  ARENOSA. CON.      (Fig.    154.) 

Shell  smooth,  and  still  somewhat  polished,  subulate ;  suture 
angularly  channelled,  columella  with  two  folds ;  outer  lip  pro- 
vided with  three  teeth.  It  is  a  rare  shell  in  the  miocene  of 
North-Carolina. 


FIG.  154.       FIG.  161.        FIG.  158.       FIG.  160.       FIG.  157. 


PYRAMIDELLA   RETICULTA. N.    S.       (Fig.  155.) 

Shell  turrited;  whirls,  six  or  seven,  and  ornamented  by 
numerous  longitudinal  ribs,  and  less  distinct  spiral  lines  giv- 
ing the  surface  a  reticulated  appearance ;  columella  three 
plaited.  It  closely  resembles  the  P.  elaborata — H.  E.  Lea. 


FIG.  155.       FIG.  156.       FIG.  159.       FIG.  162. 


FIG.  164. 


CHEMNITZIA. (Fig.    156.) 

Shell  slender,  elongated ;  many  whirled ;  whirls  longitudin- 
ally plaited  and  marked  by  obscure  spiral  lines ;  aperture 
simple,  ovate.  Bather  rare  in  the  shell  marl  at  Magnolia. 


NOETH-CAEOLINA   GEOLOGICAL   SUEVEY.  269 

CHEMN1TZIA   EETICULATA.—  N.    S.       (Fig.  156a.) 

It  has  six  reticulated  whirls,  and  about  six  revolving  ridges 
to  each  whirl.    Miocene  of  Lenoir. 

FIG.  180.       FIG.  156a.       FIG.  166.  FIG.  165. 


GENUS   EULIMA  ;   EISSO. 

Shell  small,  white,  polished,  porcellanous,  elongated,  whirls 
numerous,  flat ;  outer  lip  sharp,  but  thickened  within ;  pillar 
lip  reflected  over  the  columella. 

EULIMA  LAEVIGATA. — PASITHEA  LAEVIGATA. — H.  E.  LEA.    (Fig  157.) 

Shell  small,  acute,  rather  conical,  polished  and  porcellan- 
ous ;  whirls,  about  nine ;  suture,  obscure  linear. 

EULIMA   SUBULATA. — N.    S.       (Fig.    158.) 

Shell  subulate,  porcellanous ;  whirls,  nine  or  ten,  slightly 
convex ;  sutural  space  rather  wide  ;  aperture  elongated.  This 
shell  is  not  uncommon  in  the  shell  or  miocene  marl  of  Lenoir 
county. 

FAMILY   CEEITHIADAE — CEKITHIUM    (TEIPHOEIS)    MONILIFEEUM  I     H. 

E.    LEA.— (Fig.    159.) 

Shell  subulate,  sinistral,  thick,  costate,  sutures  small ;  whirls, 
ten,  flat;  ribs  three,  moniliforrn;  columella  smooth;  canal 
short  and  deep. 

CEEITHIUM. — (Fig.  160.) 

Shell  small,  elongated ;  whirls,  many,  slightly  convex,  or- 
namented with  numerous  longitudinal  ribs,  which  extend 
across  the  whirl ;  canal  short  and  deep. 

CERITHIUM   ANNULATUM. — N.    S.      (Fig.    161.) 

Shell  small  but  thick ;  whirls  many,  ornamented  with  three 


270  NORTH-CAROLINA  GEOLOGICAL   SURVEY. 

sharp  spiral  ridges.    These  ridges  are  but  slightly  oblique  to 
the  axis  of  the  shell 

CERITHIUM   BICOSTATA. N.    S.       (Fig.    162.) 

Shell  small,  thick,  tapering  from  the  base ;  whirls  many, 
and  ornamented  with  two  spiral,  nodulose  ribs. 

TEREBELLTJM    ETIWANENSIS.-— TOUMEY    AND    HOLMES FOSSJLS     OF 

SOUTH-CAROLINA. 

Shell  subulate ;  whirls  many,  pointed,  flattened  and  orna- 
mented with  two  sharp  spiral  ribs ;  sutural  line  deep,  especi- 
ally below. 

This  shell  presents  considerable  variation  in  passing  from 
its  immature  to  its  mature  state.  In  the  young  the  spiral 
ridges  are  placed  near  the  suture,  and  the  space  between  is 
concave ;  the  waving  lines  of  growth  gives  it  an  obscurely 
beaded  appearance.  It  is  the  most  common  univalve  in  the 
marl  beds  of  Edgecombe  county. 

TEREBELLUM    CONSTRICTUM. N.    S. 

Shell  rather  thin  terete  ;  whirls  many  convex ;  lower  ones 
deeply  constricted  on  the  line  of  suture,  and  ornamented  by 
two  principal  raised  revolving  lines  placed  nearer  «the  lower 
margin  than  the  upper  ;  the  finer  parallel  lines  are  numerous  ; 
longitudinally,the  spire  is  frequently  marked 
FIG.  163.  kv  obsolete  ribs  ;  lines  of  growth  indistinct. 

It  differs  from  the  T.  Etiwanensis  in  the 
position  of  the  principal  revolving  lines, 
and  the  lower  rounded  whirls. 


TEREBELLUM    BURDENII. TOUMEY    &   HOLMEfl. 

FOSSILS   OF   SOUTH-CAROLINA,    P.    122. 

(Fig.  163.) 

"  Shell  subulate,  turrited  ;  whirls  flatten- 
ed, spirally  ribbed  and  transversely  striated, 
which  give  the  ribs  a  beaded  character.'7 


NORTH-CAROLINA  GEOLOGICAL  SURVEY.  271 

SCALARIA   MULTI6TRIATA. — (Fig.    164.) 

Shell,  small  whirls  numerous,  rather  convex  and  ornament- 
ed with  many  sharp  longitudinal  ribs. 

All  the  specimens  of  this  species  of  scalaria  which  fell  under 
my  observation  were  imperfect  at  the  aperture.  Shell  marl 
of  Lenoir  county. 

SCALARIA   CTJRTA. — N.    S.       (Fig.    165.) 

Shell  thin  and  delicate ;  whirls  about  four,  ornamented  with 
rather  flexuose,  sharp,  longitudinal  ribs.  Shell  marl  of  Lenoir 
county. 

SCALARIA   CLATHRTJS. (Fig.    166.) 

All  the  specimens  of  this  species,  when  found,  were  im- 
perfect. It  differs  from  the  preceding  in  having  transverse 
ribs  between  the  longitudinal  ones. 

PETALOCONCHUS . — LEA. — PETALOCONCHU8   SCULPTtJRATUS. 

(Fig.  169.) 

Shell  vermiform,  tubular,  provided  with  two  longitudinal 
plates  internally ;  externally  it  has  nodulose  ribs 

•IQ' 169>  or  costae.  The  shell  is  curiously  twisted  into  knots, 
but  sometimes  it  is  rolled  up  into  a  coil  somewhat 
conical,  as  in  the  figure,  after  which  it  is  coiled 
irregularly.  It  is  very  common  in  the  miocene 
marl  beds  of  the  State. 


FAMILY   LITORINIDAE. — LITORINA   LINEATA. — N.    S. 

(Fig.  170.) 

Shell  rather  small,  thick  conical  ;  wrhirls  five 
nearly  flat,  and  the  two  lower  are  ornamented  with  many 
spiral  ridges,  which  are  crossed  by  obscure  lines  of  growth ; 
three  upper  whirls  smooth. 


, 


272  NORTH-CAROLINA  GEOLOGICAL   SURVEY. 

FAMILY   TURBINIDAE — TEOCHUS    PHILANTROPUS. — (Fig.    167.) 

Shell  conical,  but  rather  depressed; 
whirls  slightly  angular  at  base,  and  orna- 
mented with  spiral  beaded  lines,  alternat- 
ing in  size. 

TROCHUS. — (Fig.  168.) 
It  appears  to  differ  from  T.  armillatus, 
but  I  am  unable  to  refer  it  to  any  of  the 
FIG.  lesT          species  described  in  the  miocene  beds. 

DELPHINULA   QTJADRICOSTATA. — N.    S.       (Fig.    180.) 

Shell  small,  thin ;  whirls,  few,  angulated  and  furnished  with 
four  ribs,  which  are  crossed  by  lines  of  growth ;  aperture  an- 
gular. 

Found  occupying  the  interior  of  the  large  univalve  shells 
of  the  miocene. 

\ 

ADEORJBIS. WOOD.       (Fig.  181.) 

I  have  placed  this  figure  under  this  genus,  though  it  does 
not  agree  with  it  in  every  particular. 

FAMILY   TORNATELLIDAE. 

This  family  has  a  convoluted  shell ;  it  is  cylindrical,  or  sub- 
cylindrical,  with  a  long  narrow  aperture ;  columella  plaited. 

TORNATINA   CYLINDRICA. — N.    8.       (Fig.  182.) 

FIG.  182.  Shell  small,   convoluted,  cylindrical,  porcel- 

,.  lanous,  or  polished  ;    spire  depressed ;  whirls, 

f  1 1          angulated  ;  suture  channelled  ;    aperture  long 

and  narrow ;  outer  lip  arcuate  ;  columella  with 

one  fold. 

This  small  shell  resembles  a  cyprea,  or  some  of  the  smaller 
species  of  olivas.  It  is  not  uncommon  in  the  miocene ;  it  is 
usually  found  in  the  cavities  of  the  larger  univalves. 


NORTH-CAROLINA   GEOLOGICAL   SURVEY.  273 

FAMILY   HELICTDAE LAND-SHELLS. HELIX   TRIDENTATA. 

(Fig.  183.) 

Shell  depressed,  or  flattened,  convex  ;  whirls,  four  and  ob- 
liquely wrinkled  ;  aperture  contracted  and  furnished  with  two 
teeth  on  the  outer  lip,  and  one  upon  the  inner  lip ;  the  latter 
is  curved. 

FIG.  186.          FIG.  185.          FIG.  184.          FIG.  183. 


H.    LABYRINTHICA. (Fig.    184.) 

Shell  small  and  of  a  conical  form ;  whirls,  six  and  marked 
with  oblique  lines  of  growth  ;  lip  reflexed ;  inner  lip  furnished 
with  a  single  tooth  extending  within  the  shell. 

< 

FAMILY    LIMNEIDAE. FRESH-WATER    SHELLS. PLANORBIS     BICARI- 

NATUS.— (Fig.  185.) 

Shell  deeply  concave  on  both  sides ;  whirls,  three ;  carina- 
ted  on  both  sides ;  lip  on  the  left  extending  beyond  the  plane 
of  the  preceding  whirl. 

This  fresh  water  shell  is  rare  in  the  miocene  beds  of  the 
Cape  Fear. 

FAMILY   PALUDINIDAE. 

This  family  embraces  certain  gasteropods,  most  of  which 
live  in  fresh  water,  as  lakes,  ponds  and  rivers.     The  form  of 
their  shells  is  conical  or  globose,  covered  with  a  thick  green 
epidermis.     The  aperture  is  rounded  and  the  whirls  convex  ;  ' 
peristome  continuous. 

t 

PALUDINA   SUBGLOBOSA. N.  S.       (Fig.  186.) 

Shell  rather  thin,  turbinated ;  whirls,  four,  rounded  or  con- 
vex, short ;  aperture  rounded ;  third  whirl  marked  by  four  or 
five  spiral  obsolete  lines.  It  has  a  close  resemblance  to 
Gould's  and  Halderman's  genus  Amnicola. 

Miocene  of  Cape  Fear,  but  it  is  by  no  means  a  common 
shell. 

19 


274 


NOKTH-CAEOLINA   GEOLOGICAL   8UKVEY. 

CLASS  BKACHIOPODA. 


FIG.  187. 


ORBICULA    LUGUBEIS. CON.       (Fig.    187.) 

Shell  corneous,  oblong-ovate,  depressed ; 
concentrically  lamellose  ;  apex  behind  the 
centre ;  posteriorly,  it  is  marked  by  a  few 
radiating  lines;  interiorly,  it  is  smooth, 
and  there  is  a  short  longitudinal  ridge  on 
the  median  line.  In  some  of  the  miocene 
beds  in  Wayne  county,  it  is  quite  common. 


FAMILY   DENTALLDAE. TOOTH    SHELLS. 

The  dentalidae  are  hollow,  curved  tooth-like  shells.  They 
are  usually  ornamented  by  longitudinal  ridges,  but  sometimes 
they  are  smooth  and  polished.  They  have  a  round  or  circular 
aperture. 


DENTALUM   ATTENUATUM. SAY.       (Fig.    188.) 

Shell  gently  curved,  and  ornamented 
with  twelve  rounded  ribs  ;  aperture  cir- 
cular. Common  in  the  shell  marl  of  this 
State. 


FIG.  188. 


D.    THALLUS. CON.       (Fig.  189.) 

Shell  small,  polished,  curved  and  ta- 
pering towards  both  extremities.  Com- 
mon in  the  shell  marl. 


FIG.  190. 


FIG.  189. 


CAECUM  ANNULATUM. N.  8.       (Fig.  190.) 

Shell  minute  curved  ;  ends  subequal ; 
aperture  circular ;  surface  annulated. 

This  minute  shell  is  quite  common  in  the  miocene  of  this 
State.  It  is  found  in  the  interior  of  larger  ones,  which  it 
probably  inhabited. 


FAMILY    CALYPTKAEIDAE. LIMPETS. BONNET    LIMPETS — CUP    AND 

SAUCER   LIMPETS. 

The  limpets  have  but  one  valve.     It  is  sometimes  saucer 


NORTH-CAROLINA  GEOLOGICAL   SURVEY. 


275 


shaped  or  sub-conical,  and  passing  into  a  cone,  within  which 
there  is  an  appendage  somewhat  similar  in  form  to  the  outer 
cone.  These  cones  are  frequently  sub-spiral.  They  adhere 
to  rocks  and  stones  with  their  apertures  below. 


CRUCIBULTJM   COSTATTTM. 

Shell  rather  thick,  circular  at  base,  and  furnished  with 
strong  but  rather  irregular  ribs ;  apex  sub-central ;  margin 
crenulated. 

CRTJCIBULTJM   RAMOSUN. — CON.       (Fig.  191.) 

Shell  ovate ;  apex  sub-central ;  ribs  prominent  and  orna- 
mented by  a  series  of  subordinate  diverging  ridges,  but  par- 
tially interrupted  by  the  lines  of  growth  ;  inner  cup  sub-con- 
ical, entire,  and  marked  by  circular  ridges,  or  lines  of  growth. 


FIG.  196. 


FIG. 192 


FIG.  193. 


FIG    195. 


FIG.  191. 


FIG.  194. 


276  NORTH-CAEOLINA   GEOLOGICAL   SURVEY. 

C.   DUMOSUM. 

Shell  depressed,  sub-conical,  oblong  or  oval  at  base ;  sur- 
face ornamented  with  spiral  ribs,  and  whose  spines  are  hol- 
low. 

C.  MULTILINEATUM. (Fig.  192.) 

Shell  rather  small,  depressed,  very  thin  ;  apex  elevated, 
sub-central,  disk  marked  with  radiating  lines.  Rather  com- 
mon in  the  miocene.  Usually  occupies  the  interior  of  other 
shells. 

TEOCHITA   CENTBALIS. — (Fig.    193.) 

Shell  rather  small,  very  thin,  round,  ovate ;  apex  medial 
minutely  spiral  and  acute.  Associated  with  the  foregoing 
shells  of  this  family. 

CEEPIDULA. LAM. 

Crepidula  has  the  limpet  shape,  but  a  posterior  oblique 
marginal  apex.  Interior  has  a  horizontal  plate,  forming  a 
partition  which  curves  the  posterior  half.  They  vary  in  form, 
which  is  very  much  dependent  upon  the  surface  to  which  they 
are  attached. 

CEEPIDULA   FOENICATA. — (Fig.  194.) 

Shell  obliquely  oval ;  surface  convex,  smooth  or  wrinkled  ; 
apex  turned  to  one  side :  diaphragm  concave  below,  occupy- 
ing half  the  shell.  Common  in  the  miocene  of  North-Caro- 
lina. 

CEEPIDULA   SPINOSA. 

Shell  depressed,  oval,  costate  and  spinous,  especially  to- 
wards the  margin.  Common  in  the  miocene. 

CEEPIDULA   PLANA. SAT.       (Fig.  195.) 

Shell  nearly  flat,  slightly  convex  ;  diaphragm  convex  ;  the 
form  is  very  variable,  assuming  the  shape  of  the  surface  upon 
which  it  rests. 


NORTH-CAROLINA   GEOLOGICAL   SURVEY.  277 

FAMILY   FISSURELLIDAE. KEY-HOLE    LIMPETS. 

Shell  limpet  shaped;  some  have  the  margin  notched  in 
front ;  in  others  the  apex  is  perforated.  Adhere  to  rocks  and 
stones. 

FISSURELLA   REDIMICULA. (Fig.    196.) 

Shell  ovate,  oblong,  elevated,  and  rather  thick;  surface  or- 
namented with  fine  longitudinal  ridges,  which  are  intersected 
by  circular  lines  of  growth,  which  gives  the  surface  a  reticu- 
lated appearance ;  margin  entire,  but  ridged  internally ;  apex 
truncated,  figure  inclined,  oblong. 

This  shell  is  not  an  uncommon  occupant  of  the  shell  marl 
beds  of  this  State. 

CLASS  LAMELIBRANCHIATA. 

FAMILY    OSTREIDAE. 

"  Shell  inequivalve  and  nearly  inequilateral ;  free  or  adhe- 
rent resting  on  one  valve  ;  beaks  central,  straight  ligament  in- 
ternal ;  muscular  impression  single  and  behind  the  centre ; 
hinge  usually  without  teeth." 

OSTREA   VIRGINIANA. 

Shell  thick,  strongly  and  radiately  plicated ;  concentrically 
laminated  and  imbricate ;  upper  valve  nearly  flat ;  pliated 
towards  the  margin ;  beaks  laterally  curved ;  very  variable. 
Common  in  the  miocene  beds  of  North- Carolina. 

OSTREA   CAROLINENSIS. 

Shell  ob-ovate,  thick,  compressed,  concentric  lamina  imbri- 
cated, and  transversely  plaited ;  beaks  broad  and  prominent. 
Fosset  large  and  bounded  laterally  by  strong  ridges. 

Occurs  in  the  miocene  of  North- Carolina,  but  is  less  com- 
mon than  the  preceding. 

Ostrea  radians  and  O.  sellaeformis  belong  also  to  the  mio- 
cene beds,  together  with  the  Anomia  ephippium ;  the  latter  is 
always  broken. 

, 


278 


NORTH-CAROLINA   GEOLOGICAL   SURVEY. 


EXOGYRA   COSTATA.- 


FIG.   A. 


[Fig-  A.) 

Shell  sub-oval,  very  thick, 
lower  valve  convex,  and  cov- 
ered with  strong  corrugated 
ribs;  apex  lateral,  with  about 
two  volutions;  upper  valve 
flat,  thick,  supplied  with  nu- 
merous elevated  concentric 
squamose  plates.  It  belongs 
to,  and  is,  one  of  the  charac- 
teristic fossils  of  the  green 
sand  at  Black  Rock,  on  the 
Cape  Fear,  and  at  Rocky 
Point,  twenty  miles  north  of 
Wilmington.  It  is  found  in 
the  miocene  at  several  places  on  the  Cape  Fear,  but  its  pres- 
ence is  due  to  accident. 


FIG.  B. 


CUCULLAEA   VULGARIS. — (Fig.  B.) 

This  fossil  occurs  in  the  form  of  an  in- 
side cast  of  the  shell ;  it  is  inflated,  sub- 
triangular,  flattened  before,  beaks  prom- 
inent and  in-curved;  shell  thick,  and 
marked  with  numerous  delicate  longitu- 
dinal striae. 

It  is  associated  with  the  Exogyra  and 
Belemnitella  at  Black  Rock  in  the  green 
sand. 

The  C.  vulgaris  is  placed  here  fromite 
association  with  the  E.  costata. 


FAMILY   PECTENIDAE. PECTEN,    SCALLOP. 

Shell  sub-orbicular,  regular,  resting  on  the  right  valve, 
usually  ornamented  by  fretted  or  scaly  ribs  radiating  from 
the  hinge ;  right  valve  most  convex,  with  a  notch  below  the 
front  ear ;  hinge  margin  straight,  united  by  a  narrow  liga- 
ment ;  cartilege  internal  in  a  central  pit. 


NORTH-CAROLINA   GEOLOGICAL   SURVEY.  279 

The  scallop  of  our  coast  is  regarded  as  a  delicacy.  It  lives 
in  shallow  water,  and  is  taken  in  great  numbers !  at  low  tide 
from  banks  which  are  just  submerged.  They  move  through 
the  water  by  opening  and  shutting  their  valves./  Fossil  pec- 
tens  or  scallops  are  very  abundant  in  most  of /the  miocene 
marl  beds  in  this  State.  The  large  scallops,  P.  Jeffersonius 
and  P.  JVladisonius  abound  in  beds  upon  the  Neuse  and  Tar 
rivers,  while  they  are  less  numerous  upon  the  Cape  Fear. 
Another  large  species  is  found  upon  the  Meherrin,  in  North- 
ampton county,  which  I  have  not  met  with  elsewhere.  It 
replaces  the  English  species,  the  Pecten  princeps,  which  it 
closely  resembles. 

PECTEN    COMPARILIS. 

Shell  medium  size ;  both  valves  convex  with  twenty-three 
or  twenty-four  ribs,  prominent  and  angular  inside  at  base ; 
ribs  and  spaces  between  nearly  equal ;  ears  radiately  striate. 
of  the  most  common  fossils  upon  the  Cape  Fear. 

PECTEN    EBOREUS. (Fig.  197.) 

FIG.  197. 


280 


NOBTH-CAKOLINA   GEOLOGICAL   SURVEY. 


S(hell  comparatively  thin,  and  light  and  compressed  valves; 
circular,  sometimes  oblique  and  equilateral ;  ribs  twenty-four, 
marked  on  the  outside  with  concentric  squamose  lines  of 
growth,  which  are  undulating,  the  last  of  which  are  strong ; 
lower  valve  less  convex  than  the  upper.  It  differs  from  the 
comparilis  in  being  concentrically  marked,  and  thinner,  be- 
sides it  grows  much  larger. 

PECTEN   PKINCEPOIDES. N.  S. (Fig.    198.) 

Shell  large,  rather  thick,  compresed,  sub-inequilateral,  ra- 
diating striae  coarse  and  very  numerous  ;  transversely  marked 


by  lines  of  growth,  giving  the  surface  a  wrinkled  appear- 
ance; ears  unequal ;  buccal  ear  sinuate,  radiating  striae  nu- 
merous, inside  smooth,  striae  obsolete ;  tig.  reduced. 

This  is  a  large  species  of  pecten,  is  closely  allied  to  the  P. 
princeps  of  the  English  crag.  It  is  common  in  the  miocene 
marl  on  the  Meherrin  river,  at  Murfreesboro'.  It  is  five 
inches  long,  and  five  and  a  quarter  wide.  It  is  readily  dis- 
tinguished by  the  absence  of  ribs  proper,  and  the  presence 
of  coarse  radiating  striae,  which  have  intermediate  ones, 


NORTH-CAROLINA   GEOLOGICAL   SURVEY. 


281 


which  do  not  reach  the  hinge  or  umbo ;  many  of  the  striae, 
however,  fork  or  divide. 

P.    PEEDEENSIS. 

Shell  thick  and  strong,  broadly  ovate ;  ribs,  eight,  broad 
striae,  lines  of  growth  strong  towards  the  margin  ;  beak  pro- 
jecting beyond  the  hinge  line. 

Only  one  valve  has  been  found  of  this  species,  and  being 
old  and  its  striae  obliterated  in  part,  and  its  characters  are 
less  distinct  than  is  usual  in  specimens  belonging  to  this  ge- 
nus. 

p.  MORTONI. 

Shell  large,  circular,  compressed,  thin,  pearly  ;  equivalve 
equilateral ;  concentrically  marked  by  fine  lines  of  growth ; 
on  the  outside,  ribs  are  invisible ;  inside,  ornamented  by  about 
eighteen  pairs  of  ribs,  which  are  prominent  at  the  margin, 
and  obsolete  towards  the  hinge. 

This  beautiful  shell  occurs  in  the  naiocene  at  Waccamaw 
Lake,  North-Carolina,  and  has  not  been  observed  upon  the 
JSreuse  or  farther  north. 

P.    JEFFERSONIUS. (Fig.  199.) 

FIG.  199. 


282 


NORTH-CAROLINA   GEOLOGICAL   SURVEY. 


Shell  very  large,  ribs,  ten,  and  wide,  and  longitudinally 
marked  by  fine  ridges,  which  are  not  squamose.  This  species 
is  sometimes  between  nine  and  ten  inches  wide,  and  seven  or 
eight  inches  long,  and  are  often  used  in  cooking  oysters  in 
place  of  a  frying  pan.  It  is  one  of  the  characteristic  fossils  of 
this  miocene. 

p.  MADISONIUS. — (Fig.  200.) 

In  the  P.  Madisonius,  the  ribs  number  about  fifteen,  and 
they  are  ornamented  with  three  squamose  ridges  each.  There 
is  also  an  equal  number  between  them ;  they  coalesce  towards 
the  hinge. 


FIG.  200. 


FIG.  201. 


A.  pecten,  (fig.  201,)  is  quite  com- 
mon in  North-Carolina,  which  I  have 
not  been  able  to  refer  to  its  proper 
species.  It  is  one  of  the  most  com- 
mon in  the  shell  marl  of  the  middle 
part  of  the  eastern  counties.  It  has 
ten  prominent  ribs,  but  they  are  or- 
namented in  a  different  style  from 
that  which  prevails  in  the  young  of 
the  P.  Jeffersonius. 


NORTH-CAROLINA   GEOLOGICAL   SURVEY. 


283 


FIG.  202. 


One  of  the  most  common  pectens  of  the  white  eocene  marl, 
is  represented  by  figure  202.  It  differs 
from  the  P.  membranacea  in  having  only 
about  half  the  number  of  ribs.  The  P, 
membranacea  having  upwards  of  eighty, 
while  this  has  about  forty-four. 

An  observer  cannot  fail  to  perceive  the 
striking  difference  in  the   species  of  pec- 
tens   of  the  white    eocene  marl  of  E"ew- 
Hanover  and  Onslow  counties,  and  those  of  the  miocene. 


FIG.  203. 


FIG.  203A. 


PLICATULA   MARGINATA. — (Fig.  103.) 

Shell  strong  and  thicl^  but  rather 
small;  valves  sub-equal,  ovate,  wedge- 
form,  with  three  strong  radiating  plicae. 

FAMILY  MYTILTDAE. MYTILUS  INCRASSATT76. 

(Fig.  203A.)       \ 

Shell  nacreous,  thick,  somewhat  infla- 
ted, marked  with  concentric  lines  of 
growth;  anterior  margin  arched  ac- 
cuminate;  posterior  rounded,  some- 
what dilated ;  umbones  acute.  It 
is  usually  much  injured  by  exfolia- 
tion and  rarely  perfect. 


FIG.  203B. 


it  may 
ocene. 


CRENELLA.— (Fig.  203B.) 

Shell  small,  short, 
thin,  smooth  in  the 
middle ;  hinge,  mar- 
gin crenulated  behind 
the  ligament.  It  ap- 
pears to  be  rare,though 

be  owing  to  its  frailness.    Mi- 


284: 


NORTH-CAROLINA   GEOLOGICAL   SURVEY. 


ARCADAE. 

The  valves  in  the  Arcadae  are  equal,  regular,  and  usually 
oblique ;  the  teeth  are  arranged  in  long  rows,  resembling  a 
comb ;  at  the  extremes  they  are  longer  and  frequently  curved 
or  corrugated. 

ARCA   LIENOSA. SAY. (Fig.  204.) 

Shell  large,  inflated,  oblique ;  ribs  subequal,  numerous, 
with  a  groove  or  channel  in  the  middle ;  anterior  side  angu- 

FIG.  204. 


lar ;  lines  of  growth  distinct,  giving  a  striate  appearance ;  the 
ligament  area  is  marked  by  strong  lines  diverging  from  be- 
neath the  umbo;  umbones  distant;  inside  margin  strongly 
sulcate  or  ribbed.  It  has  about  37  ribs.  A  living  shell  upon 
the  Florida  coast,  but  found  abundantly  in  the  miocene  of 
Js^orth- Carolina. 

A.  SCALARIS. 

Shell  oblong,  ovate ;  ribs  twenty-one,  strong  and  trans- 
versely rugose,  ligament  area  short,  transversely  marked  by 
lines,  and  crossing  striae  parallel  to  the  hinge  line. 


A.    INCILE. SAY. 

Shell  very  oblique,  sub-quadrangular ;  anterior  side  very 
short,  posterior  sinuate ;  ribs  unequal,  stronger  on  the  poste- 
rior margin ;  rounded  before,  angular  behind,  and  much  pro- 


NORTH-CAKOLINA   GEOLOGICAL    SUEVEY. 


285 


duced ;  umbones  incurved,  distant ;  ligament  area  crossed  by 
transverse  lines. 

This  shell  has  about  thirty-one  principal  ribs,  with  inter- 
vening raised  lines,  and  transversely  marked  by  lines  of  growth. 

A.    CENTENARIA. — (Fig.   205.) 

Shell  sub-quadrate  and  ovate,  nearly  straight  and  slightly 

FIG.  205. 


contracted  at  base ;  ribs  tine,  alternating  in  size ;  margins 
rounded;  beaks  approximate;  hinge  area  narrow;  margins 
entire. 

The-  striae  or  ribs  in  this  species  are  very  numerous  and 
line,  while  these  together  with  its  quadrangular  form  will 
serve  to  distinguish  it  from  others  of  the  same  genus.  Com- 
mon in  the  miocene  of  North-Carolina.  The  h'gure  was 
drawn  from  a  specimen  obtained  from  the  indurated  sand  be- 
neath the  miocene  bed  at  Elizabethtown,  Bladen  county,  and 
is  referred  to  the  centenaria  but  with  doubt. 

A.  IDONEA. 

Subcordate  inequivalve  ventricose ;  elongated  and  only 
slightly  oblique;  beaks  very  prominent  and  distant;  ribs 
about  twenty -five,  crenulated,  or  transversely  ridged;  hinge 
area  wide  and  marked  by  divergent  striae  or  channels.  Com- 
mon in  the  miocene  of  North-Carolina. 

A.    TKANSVERSA. 

Shell  rather  thin,  subrhomboidal,  rounded  with  about 
thirty-two  ribs ;  area  rather  narrow,  with  two  or  three  undu- 
lated grooves.  Common  in  the  miocene,  and  still  living  upon 


286 


NORTH-CAROLINA   GEOLOGICAL   SURVEY. 


the  coast.    A.  limatula  and  stillicidium  are  also  miocena 
shells,  and  common  in  the  marl  beds  of  the  Cape  Fear  river. 


FIQ.  206. 


VERTICORDIA. WOOD. (Fig.  206.) 

I  have  met  with  two  or  three  specimens  only 
of  the  fossil  which  I  have  referred  to  this  genus. 
It  is  found  in  the  interior  of  large  shells. 


GENUS    PECTUNCULUS. 

Shell  orbicular,  nearly  eqilateral,  smooth  and 
radiately  striated ;  hinge  with  a  semi-circular  row  of  trans- 
verse teeth. 


PECTUNCULTJS    SUBOVATUS. (Fig.  207.) 

Shell  orbicular,  inequilateral,  with  radiating  sulci,  becom- 

FIG.  207. 


ing  obsolete  with  age  ;  teeth  nearly  obliterated  in  the  centre  ; 
teeth  largest  on  the  shorter  side  of  the  valve ;  marginal  ones 
broad  and  separated ; — Conrad.  This  is  probably  one  of  the 
most  common  miocene  fossils  of  the  shell  marl  in  the  State. 


P.    LENTIFORMIS. 

Shell  orbicular,  sub-equilateral ;  the  radiating  striae  are  nu- 
merous ;  beaks  small  in  proportion  to  the  size  of  the  shell ; 
hinge  teeth  in  the  centre,  wanting  or  obsolete.  This  fine  spe- 
cies in  some  marl  beds  upon  the  Cape  Fear,  is  quite  common, 
and  is  very  large  and  thick ;  some  are  four  to  four  and  a  half 
inches  across. 


NORTH-CAROLINA   GEOLOGICAL   SURVEY. 


287 


FIG.  208. 


P.    ARATUS.— (Fig.  208.) 

This  is  the  smallest  species  of  this  gernis  be- 
longing to  the  shell  marl.  It  is  also  one  of  the 
most  common.  P.  passus  and  P.  quinqueruga- 
tus  are  also  common  in  certain  localities. 


LEDA   ACTJTA. — (Fig.  208A.) 

FlG-  208A-  Shell  small,  thick,  inflated  pos- 

^/sssss^i\      /(^^fc^*.         teriorl  j;  margin  acute  or  beaked, 

<?:  )  1  ^^^       slightly  open ;  anterior  margin, 

short  rounded  ;  surface  concen- 
trically striated.     This  fossil  re- 
sembles nucula,  but  it  is  not  pearly  in  the  interior,  and  its  ab- 
dominal margin  is  smooth. 

cute, 

NUCULA   PROXn18.   about  *208B.) 

Shell  small,  6^,ra^(mooth,  interior  pearly  ; 
anterior  margin  short ;  posterior  side  elongat- 
ed, obtuse ;  margin  crenate.  ET.  limatula  is 
more  common  in  the  marl  beds  of  this  State 
than  the  N.  proxima ;  miocene. 


FIG.  20SB. 


FAMILY    CHAMACEDAE. 

The  shell  is  thick,  inequavalve,  with  sub-spiral  beaks,  hinge 
teeth  1 — 2,  muscular  impression  one,  and  large ;  reticulated 
palleal  line  simple. 

CHAMA. 

The  shell  is  attached  to  other  bodies  by  its  left  umbo : 
hinge-tooth  of  the  free  valve  thick,  curved,  and  received  be- 
tween the  teeth  of  the  other  valve. 


CHAMA   ARCINELLA. (Fig.  209.) 

Shell  thick,  or  orbicular-cordate  squamose ;  the  radiating 
ribs  spinose,  strong,  tubular  or  folded ;  intervening  space 
coarsely  punctate  and  rugose.  Common  in  the  marl  bed  at 
Elizabethtown,  Bladen  county. 


NOKTH-CAROLINA   GEOLOGICAL   SURVEY. 
FIG.  209. 


C1U 


FIG.  210. 


Pig.  210.) 

o'he'll    thick, 


squamose,    or 


concentrically  laminated  and 
imbricate  ;  lamina  striated,  sin- 
istral,  crenulated  interiorly ; 
upper  valve  flat.  Figure  low- 
er valve  natural  size.  Abun- 
dant in  the  miocene  of  North- 
Carolina,  especially  on  the 
Cape  Fear. 

OHAMA  CONGEEGATA. 

Shell  thick,  orbicular,  with 
its  surface  composed  of  plates 
or  lamina ;  in  the  flat  valve  the  places  are  crenulated  or  plai- 
ted. 

CHAMA  STRIATA. N.  S.       (Fig.  211.) 

Fig.  211.  Shell  small,  ovate,  rather  thick  for  its  size? 

lower  valve  distinctly  striate.     Usually  found 
in  the  hollow  or  inside  of  the  univalves. 


FOR* 


NORTH-CAROLINA   GEOLOGICAL    SURVEY. 


289 


FAMILY   CYPRINIDAE. 

Shell  regular,  equivalve  oval  or  elongated ;  valve  close, 
solid ;  epidermis  thick  and  dark  ;  ligament  external,  conspic- 
uous cardinal  teeth  1 — 3  in  each  valve  ;  pedal  scars  close  to 
or  confluent  with  the  adductors  ;  pallia!  line  simple. — Wood- 
ward. 

ASTARTE 

Shell  small,  thick,  compressed,  smooth  or  concentrically 
furrowed ;  Innule  impressed  ;  ligament  external ;  hinge  teeth 
2 — 2  ;  anterior  tooth  in  the  right  valve  large  and  thick. 


FIG.  212. 


ASTARTE   CONCENTRICA. (Fig.  212.) 

Shell  small,  thick,  triangular,  compress- 
ed, concentric ;  furrows  close  and  regular  • 
umbones  acute,  recurved ;  margin  cre- 
nate.  It  is  about  one  inch  long,  and  cr_e 
broad.  It  is  rather  common  in  the  mio- 
cene  of  North-Carolina. 


ASTARTE   UNDULATA. (Fig.  213.) 

The  broad,  variable  and  concentric  furrows  will  serve  to 
distinguish  it  from  the  foregoing.  It  is  comparatively  a 
broader  shell.  The  Undulata  seems,  however,  to  be  quite 
variable,  and  the  figure  shows  one  of  the  extremes  of  this 
species. 

CRASSATELLA   IJNDULATA. (Fig.  214.) 

Shell  oblong,  ovate,  compressed,  mark- 
ed upon  the  outside  with  coarse  concen- 

FIG.  214. 


FIG.  213. 


290 


NORTH-CAROLINA   GEOLOGICAL   SURVEY. 


FIG.  215. 


trie  striae ;  umbo  flattened ;  apex  sub-acute ;  inner  margin 
entire.  One  of  the  most  common  fossils  of  the  shell  marl. 

O.  GIBBESII  :    TUOMEY  &  HOLMES,  FOSSILS  OF  SOUTH-CAROLINA  J  p.  74. 

(Fig.  215.) 

"  Shell  somewhat  triangular,  thick,  con- 
centrically furrowed ;  buccal  side  rounded; 
anal  side  somewhat  beaked,  angular,  with 
a  longitudinal  ridge  ;  umbones  incurved; 
lunule  somewhat  excavated." 

•  In  addition  to  the  foregoing,  I  may  add  the  following  as 
common  in  the  ]^orth-Carolina  shell  marl  beds :  Crassatella 
alta,  C.  Marylandica,  C.  Protexta,  C.  Melina. 


FAMILY 


FIG.  21  SB. 


CYCLASIDAE. CORBICULA  DENSATA. CYRENA  DENSATA. 

CON.— (Fig.  215A.) 

FIG.  215A.  Shell  orbicular  striated  concen- 

trically,   polished,   lateral    teeth 
elongated. 

This  shell  is  very  abundant  at 
the  miocene  marl  bed  of  Mr. 
Flower,  on  the  Cape  Fear. 


FA^SLY   CORBULIDAE. — CORBTTLA   CU~ 

NEAT  A. (Fig.  215B.) 

Shell  small,  thick,  ovate,  con- 
centrically striate ;  anterior  margin  rounded  ; 
posterior  elongated,  or  somewhat  rostrate. 
Common  in  the  shell  marl. 


FAMILY  LUCENIDAE. 

This  family  have  orbicular  shells,  both  free 
and  closed  with  hinge  teeth,  somewhat  varia- 
ble as  one  or  two  laterals,  or  one  and  one, 
and  the  other  obsolete  ;  pallial  line  simple,  muscular  im- 
pressions two,  elongated  and  rugose.  The  family  is  princi- 
pally composed  of  tropical  and  temperate  species,  and  live 


NORTH-CABOLINA   GEOLOGICAL   SURVEY. 


291 


upon  sandy  or  muddy  bottoms,  and  exist  from  the  sea  shore 
or  shallow  water  to  the  greatest  habitable  depths. 


LUCINA  BEUGIEEE. 

The  shell  is  orbicular,  white,  with  depressed  umbones,  and 
the  margins  are  either  smooth  or  only  finely  crenulated; 
hinge  teeth  2 — 2,  laterals  1—1,  muscular  impressions  rugose ; 
anterior,  elongated  and  within  the  pallial  line ;  umbanal  area 
with  an  oblique  furrow. 

LUCINA   PENNSYLVANIA. LINN.       (Fig.    216.) 

Shell  orbicular,  thick,  solid,  and  concentrically  ribbed,  or 

posteriorly  it  has  a  strong 
fold  or  groove.  The  fold  ex- 
tends across  the  shell,  and 
produces  a  notch  in  the  pal- 
lial margin.  Common  in  the 
miocene  upon  K"euse  and 
Cape  Fear  rivers. 


FIG.  216. 


LUCINA  CONTRACTA. 

Shell  orbicular,  somewhat 
inflated ;  ribs  concentric,  un- 
equal, marked  in  the  intevals 
with  striae ;  posteriorly  the 
margin  is  channeled. 

It  is  larger  than  the  preceding,  and  has  no  fold,  and  its 
ribs  are  unequal. 


FIQ.  217. 


L.    CEENULATA. (Fig.    217.) 

Shell  small,  thin,  orbicular,  somewhat  inflated, 
jfif\^       concentrically    lamellated,    lunule    excavated- 

1  iggjjy          In  addition  to  the  foregoing,  the  following 
species  have  been  observed  in   the  miocene : 
Lucina  anadonta,  L.  radians,  L.  divaricata,  L.  multihineata, 
and  L.  squamosa. 


292  NORTH-CAROLINA   GEOLOGICAL   SURVEY. 

FAMILY   VENERIDA. 

This  important  family  is  represented  by  many  existing 
species  in  our  seas  at  the  present  time.  It  is  too  well  known 
to  require  a  minute  description.  It  is,  however,  known  from 
other  forms  by  its  regular  oblong  thick  shell,  though  it  is 
sometimes  nearly  round  ;  by  its  strong  external  ligament,  and 
its  three  diverging  prominent  teeth  in  each  valve.  Its  pallial 
line  is  sinuated. 

The  venerida  are  elegant  and  beautiful  shells,  often  highly 
colored,  though  some  of  the  best  known  are  externally  dull. 
This  family  appeared  first  in  the  Oolitie  period,  and  they  have 
increased  in  number  and  importance  down  to  the  present 
time,  when  they  have  acquired  their  maximum  develope- 
raent. 

VENTJ8   MERCENARIA. 

Shell  solid,  surface  marked  by  numerous  concentric  lines 
of  growth,  obliquely  cordate  •  posterior  margin  produced ; 
anterior  short ;  umbones  recurved,  lunule  cordate ;  pallial 
line  sinuated  ;  margin  crenulated. 

VENUS   TRIDAENOIDES. — CON.       VENUS    DIFFORMIS. SAY. 

Shell  very  thiqk  and  heavy ;  globose,  wrinkles  upon  the 
surface  undulating ;  plaits  wide,  extending  fronl  the  umbo  to 
the  margin. 

This  species  may  be  distinguished  by  its  thickness  and  wide 
external  plaits,  which  are  usually  strongly  marked,  though 
sometimes  they  are  feebly  developed.  It  is  one  of  the  most 
common  fossils  of  the  miocene  beds  of  North- Carolina. 

VENUS   RILEYI. 

Shell  large,  thick,  oblong,  posterior  margin  prolonged, 
anterior  one  short ;  surface  concentrically  striate,  and  marked 
by  fine,  longitudinal  lines,  which  are  distinct  after  the  dermal 
covering  exfoliates.  This  is  one  of  the  largest  species,  being 
sometims  6 — 7  inches  wide.  Common  in  the  miocene  of 
Cape  Fear  river. 


NOETH-CAEOLINA    GEOLOGICAL   SUEVEY. 


293 


V.    CEIBEAEI — CON. 
FIG.  218. 


North-Carolina. 


(Fig.  218.) 

Shell  thick,  medium  size, 
slightly  ventricose,  furnished 
upon  the  outside  by  about 
twenty-five  sharp  lamelliform 
concentric  and  recurved  ribs, 
crenulated  upon  the  umbbnal 
side  ;  ribbed  or  ridged  trans- 
versely on  the  ventral  side, 
the  ridges  extending  across 
to  the  adjacent  rib ;'  lunule 
crenulated. 

Kecent  upon  the  coast  of 


v. 


FIG.  219. 


LATILIEATA   CON. VENUS   PAPHIA. — LAM.       (Fig.   219.) 

Shell  sub-trigonal,  thick  and  pon- 
derous for  its  size ;  ribs  fine,  con- 
centric, and  very  thick ;  irregularly 
stirate,  crenulate  upon  the  lower 
margin  ;  umbo  slightly  flattened. 

This  shell  is  readily  known  by  its 
thick  ribs,  and  deep  subci  between 
them.  Common  in  the  miocene  of 
North-Carolina. 


Fro.  220. 


VENUS    MELTASTEIATA. — (Fig.    220.) 

Shell  small,  sub-orbicular,  striated  concentri- 
cally, rather  irregular,  interruptedly  radiated. 

Yenus  pramagna,  cancellata  and  subnasuta  are 
also  rather  common  fossils  of  the  miocene. 


294 


NOKTH-CAROLINA   GEOLOGICAL   SURVEY. 


CYTHERCA   SAYANA. — (Fig.    221.) 

Shell  inflated,  concentrically  striate,  anterior  side  angulat- 
ed ;  umbones  prominent,  incurved ;  margin  smooth ;  himule 
cordate. 


FIG.  221. 


c.   REPOSTA. — (Fig.  222.) 

Shell  smooth,  moderately  inflated,  thick,  beaks  prominent, 
dorsal  margin  depressed;  anterior  margin  rounded,  lunule 
lanceolate. 


FIG.  222. 


C.    REPORTA. — (Fig.    223.) 

This  fossil,  which  the  annexed  figures  represent,  is  very 
common  in  a  sandy  marl  bed  in  Brunswick  county.  It  pre- 
serves its  original  polish,  and  closely  resembles  the  foregoing. 
It  is,  however,  proportionally  wider  than  the  repostia.  It  is 
highly  polished  and  smooth,  but  has  concentric  striae.  Urn- 


NOBTH-CAROLINA   GEOLOGICAL   SURVEY.  295 

bones  flattened,  the  flattened  part  extending  across  the  shell, 
being  bounded  anteriority  with  an  obscure  rounded  ridge. 


FIG.  223. 


ARTEMES  TRANSVERSUS. — N.  s.     (Figs.  223a  and  224.) 

FIG.  223a. 


296  NOKTH-CAKOLINA   GEOLOGICAL   6UKVEY. 

Shell  sub-orbicular,  depressed,  sub-equilateral,  concentrical- 
ly striate  ;  broader  than  long ;  lumule  small,  lines  of  growth 
or  concentric  striae  regular,  simple,  and  somewhat  coarse  and 
distant.  Fig.  224  shows  the  hinge. 

FIG.  224. 


This  fossil  appears  to  differ  from  the  Artimus  concentrica 
of  the  coast ;  its  linus  of  growth  are  about  half  as  numerous 
and  are  also  continuous  from  one  margin  to  the  other,  except- 
ing a  few  on  the  anterior  margin, 

In  the  living  coast  species  the  lines  of  growth  are  less 
regular,  and  coalescent  near  both  margins ;  it  is  orbicular 
also,  being  as  long  as  wide.  The  fossil,  however,  closely  re- 
sembles the  living  one  of  the  coast,  though  it  differs  as  much 
from  it  as  Artemis  acetalubum  of  Conrard. 

Species  which  belong  to  the  miocene  and  which  remain 
undescribed :  A.  acetabulum,  A.  concentrica. 

FAMILY   TELLIMIDAE. — TELLINA   BIPLICATA. CON.       (Fig.    225.) 

Shell  rather  large,  thin,  sub-oval,  inequivalve,  sub-ventri- 
cose,  marked  with  rather  obscure  radiating  lines,  and  impress- 
ed with  an  oblique  fold  in  each  valve.  The  remaining  spe- 
cies of  Tellina  belonging  to  the  miocene  are  T.  Alternata,  T. 
Polita,  and  T.  Flexuosa. 


NORTH-CAROLINA   GEOLOGICAL   SURVEY. 
FIG.  225. 


297 


FIG.  225A. 


TELLINA   LTJSORIA. (Fig.  225A.) 

Shell  oblong,  narrowed  posteriorly,  slightly  gaping  or  re- 
flected; pallial  sinus  deep; 
concentrically  striate ;  pos- 
terior margin  marked  with 
one  or  two  folds;  surface 
still  brown  ;  concentric  striae 
are  in  the  form  of  raised 
sharp  lines,  not  impressed 
lines  of  growth.  The  Tipho- 
nal  inflection  is  in  contact  with  the  pallial  line,  in  which  re- 
spect it  agrees  with  P.  Sammobia,  but  its  hinge  teeth  are 
2 — 2  in  both  valves. 


GENUS   DONAX. 

"The  general  form  is  trigonal,  or  wedge  form,  valves  closed, 
front  produced,  posterior  short ;  margins  usually  crenulated ; 
hinge  teeth  2—2 ;  laterals  1—1  in  each  valve ;  pallial  sinus 
deep." 


NORTH-CAROLINA   GEOLOGICAL   SURVEY. 


DONAX.— (Fig.  226.) 
FIG.  226.  Shell  triangular,  rather  abruptly  truncate  be- 

hind, and  traversed  by  a  ridge  from  the  umbo  to 
the  base ;  surface  marked  by  obscure  radiating 
lines;  base  crenulated.  This  small  shell  differs 
from  the  variabilis  in  its  proportion;  it  is  more  triangular,  and 
is  not  produced  so  much  in  front. 

Donax  Variabilis  probably  occurs  in  the  marl  of  £Torth- 
Carolina,  but  has  hitherto  been  overlooked. 

FAMILY   MACTRIDAE. GENUS  MACTRA. 

"  The  shell  is  equivalve,  and  nearly  equilateral ;  the  ante- 
rior hinge  tooth  is  in  the  form  of  an  inverted  A ;  lateral  teeth 
doubled  in  the  right  valve." 

MAOTRA   CONGESTA. 

Shell  rather  small,  but  thick  at  the  umbo ;  triangular,  rath- 
er inflated ;  inequilateral ;  rounded  anteriorly,  and  posterior- 
ly it  is  produced.  Yery  common  in  the  marl  of  Wayne  and 
Edge  combe. 

MACTRA   LATERALIS. SAY.       (Fig.  227.) 

FIG.  227.  Shell  small,  rather  thin,  smooth,  sub-tri- 

angular ;  lines  of  growth  fine  ;  posterior  side 
elongated,  or  margins  sub-equal,  rounded 
before ;  umbo  rather  prominent.  A  very 
common  fossil  of  the  miocene. 

4* 

MACTRA   SIMILIS. SAY. 

.Shell  thin,  of  a  medium  size,  margins  sub-equal,  concentric, 
striae  very  fine,  at  intervals  deep,  beaks  nearly  central.  The 
living  ones  of  the  coast  have  a  longitudinal  rounded  ridge 
running  from  the  beaks  to  the  base  and  obscure  radiating 
lines,  though  only  visible  in  a  favorable  position. 

'   GNATUODON   GRAYII. (Fig.    226a.) 

Shell  rather  thick,  sub-triangular,  inflated,  inequilateral, 


NORTH-CAKOLINA   GEOLOGICAL   SURVEY. 


299 


anterior  margin  rounded ;  posterior  elongated  or  wedge  form. 
Rather  common  in  the  shell  marl  beds  of  Cape  Fear. 


FIG.  227a. 


FIG.  226a. 


FAMILY    SOLENIDAE. — SOLEN   ENSIS. 

(Fig.  227a.) 

This  common  shell  of  the  coast  is  sword 
shaped,  with  the  anterior  and  posterior 
margins  truncate. 

SOLECURTIS   SUBTEKES. — CON.      (Fig.    228.) 

Shell  rather  small,  thin,  somewhat  sword 
shaped;  anterior  and  posterior  margins 
rounded,  ventral  margin  concave,  or 
arched. 

FIG.  228. 


p.  CAKIBOETTS. — (Fig.  228a.) 
Is   common   in  the  miocene,  but  the 
FIG.  228a.  valves  are  rarely  en- 

tire. I  should,  how- 
ever, express  some 
doubt  respecting  the 
identity  of  the  speci- 
men figured  with  this 
species. 


300 


HORTH-CAROLIN'A   GEOLOGICAL   SURVEY. 


FAMILY   ANATINIDAE. PANOPEA   REFLEX  A.       (Fig.    229.) 

Shell  large,  thin,  oblong,  ovate ;  wrinkled  and  margin  gap- 
ing widely  and  reflected.  Common  in  the  shell  marl  of 
Edgecombe  county. 


FIG.  229. 


PHOLADOMYA   ABRTJPTA. (Fig.    231.) 

Shell  oblong,  oval,  substance  nacreous;  surface  ornament- 
ed with  from  three  to  five  radiating  ridges.  This  beautiful 
bivalve  is  quite  common  in  a  marl  bed  in  Edgecombe  county 
but  rarely  entire. 

FIG.  231. 


NORTH-CAROLINA   GEOLOGICAL   SURVEY. 


301 


FAMILY    PHOLADIDAE. 

These  species  of  Pholas  have  been  found  in  the  miocene  of 
this,  viz:  P.  Costata,  P.  Oblongata,  and  P.  Memmingeri. 
They  are  rarely  if  ever  entire,  but  their  fragments  are  not 
uncommon. 

FAMILY  CARDIDAE.* CAKDIUM  MAGNUM. CARDIUM  VENTRICOSUM. 

Shell  large,  inflated,  obliquely  cordate,  radiately  ribbed, 
ribs  flattened,  anterior  ones  crenulated. 

This  magnificent  fossil  is  found  occasionally  in  the  miocene. 
It  is  quite  common  in  the  pliocene,  and  is  now  very  abun- 
dant upon  the  coast,  near  Beaufort. 

CARDIUM   MURICATUM. — (Fig.  232-'3.) 

The  specimen  given  in  the  figure  resembles  the  muricatuin, 

FIG.  232-'3. 


*  The  families  cardidae  and  cardiiidae  should  have  preceded  veneridae. 


NOETH-CABOLINA   GEOLOGICAL   SURVEY. 

but  it  is  more  elongated,  and  its  crenulations  appear  to  differ. 
I  have  obtained  only  one  specimen ;  and  hence,  cannot  speak 
of  the  permanence  of  its  characters.  It  occurs  in  Walker's 
Bluff,  on  the  Cape  Fear. 

Cardium  sublineatum  is  a  common  fossil  of  the  Cape  Fear 
and  Neuse  marl  beds. 

FAMILY   CAEDITIDAE. — CAEDITA   AEATA. — (Fig.  234:.). 

FlG-234-  Shell  rather   thick,  oblong, 

and  ornamented  with  fifteen  or 
sixteen  elevated  scaly  ribs ;  an- 
terior side  very  short;  poste- 
rior margin  oblique:  inner 
margin  crenate. 

.    C.    PEEPLANA. — (Fig.  235.) 

Shell  small,  rather  thick,  triangular,  inequilateral,  radiately 
ribbed,  striated ;  posterior  side  produced,  anterior  short. 
Common. 

FIG.  236.  A. 

FIG.  235. 

FIG.  236. 


C.    ABBEEVIATA. (Fig.  236.) 

Shell  small,  thick,  triangular,  oblique;  ribs  strong  and 
crenate ;  umbones  acute.  Common. 

CAEDITA   TEIDENTATA. — (Fig.  236.  A.) 

Shell  round,  triangular,  thick ;  ribs  strong  and  crennlate  : 
beaks  turned  forward ;  valves  with  two  teeth  in  the  left,  and 
one  in  the  right  valve. 

CAEDITA   CAEINATA. 

Shell  small,  thick,  wide  on  the  abdominal  side ;  ribs  strong 
and  radiating  ;  muricated ;  anterior  side  short. 


NORTH-CAROLINA   GEOLOGICAL   SURVEY.  303 

CHAPTEE  XIX. 

RADIATA. 

Considerations  relative  to  animals  belonging  to  this  type. — Aberant  forms 
of  the  Echinodermata. — Species  described. — Bryozoa,  Polyparia,  etc. 

Echinodermata  comprehends  a  class  in  the  Kingdom,  Ra- 
diata,  whose  organization  belongs  to  the  stellate  type.  Thit 
sub-class  derives  its  name  from  the  character  of  the  integu- 
ment, and  its  appendages,  which  remotely  resemble  that  of 
the  hedge-hog.  Some  are  called  sea-urchins,  others  star-fishes. 
In  most  of  the  families  of  this  great  class,  the  integument  is 
protected  by  calcareous  spines.  The  integument  itself  is  co- 
riaceous, but  it  takes  into  its  composition  a  large  quantity  of 
lime  which  imparts  to  it  firmness  and  durability.  The  skin 
is  complicated  in  its  structure.  It  is  made  up  of  an  immense 
number  of  plates  of  a  polygonal  form.  They  amount  to  600 
pieces  in  all.  These  are  dove-tailed  together  in  the  most  per- 
fect manner,  and  yet  they  are  so  invested  in  living  membrane, 
•  that  additions  of  carbonate  of  lime  are  constantly  made  to 
each.  By  this  arrangement,  the  animal  within  grows  without 
inconvenience  to  itself,  which  it  could  not  do,  if  the  integu- 
ment or  dwelling  was  composed  of  one  piece. 

The  forms  of  the  Echinoderms  differ  much  among  them- 
selves, and  yet  it  is  apparent  that  they  all  belong  to  one  type, 
and  are  constructed  upon  one  plan.  One  of  the  most  aber- 
rant of  this  type  is  the  sea  cucumber,  (Holothuria,)  which  is 
a  firm  fleshy  bag,  destitute  of  plates,  composed  of  carbonate 
of  lime.  In  another  upon  our  coast,  we  find  the  star-fishes 
with  five  arms  extending  from  a  common  center ;  and  in  an- 
other^  the  globular  sea-urchin,  in  which  the  five  arms  are 
folded  and  soldered  together  so  as  to  form  a  ball.  Another 
interesting  form  has  the  stellate  type,  but  differs  considerably 
from  the  star-fish,  and  most  strikingly  in  the  fact  that  the  stel-1 


304  NORTH-CAROLINA   GEOLOGICAL   SURVEY. 

late  head  is  supported  on  a  jointed  foot-stalk.    These  are 
called  Encrinites. 

These  different  families  have  a  special  geological  interest. 
The  last  for  example,  the  Encrinite,  lived  in  the  earliest  pe- 
riods of  the  planet,  and  are  known  principally  in  the  oldest 
palaeozoic  rocks.  In  the  lower  silurian  system,  beds  are  of- 
ten composed  mainly  of  their  disarticulated  remains.  In  mod- 
ern rocks  and  seas,  they  are  unknown.  On  the  contrary,  the 
star-fishes  without  pedicels  or  jointed  supports,  are  known 
mostly  in  modern  rocks,  only  two  or  three  species  being 
known  in  the  earlier  formations.  Now,  the  sea-urchins,  or 
the  globular  forms  of  this  class,  lived  in  great  numbers  in  the 
Mesozoic  or  Jurassic  period.  This  type  or  form  has  come  to 
us,  though  none  of  the  species  of  the  Mesozoic  period  live  in 
our  present  seas. 

I  have  spoken  of  the  complicated  structure  of  the  star-fishes 
and  the  provision  which  has  been  made  for  their  growth,  both 
of  which  are  worthy  of  our  highest  admiration.  But  nature 
had  not  exhausted  all  her  resources  when  she  had  provided 
for  their  growth  and  made  them  the  most  beautiful  objects  in 
the  seas.  She  has  in  this  elaborate  structure  made  their  or- 
namental work  subordinate  to  their  instruments  of  locomotion 
and  reproduction.  The  flowers  which  are  sculptured  upon 
their  integuments  form  a  part  of  their  organs  for  moving  from 
place  to  place.  These  flowers  which  represent  the  five  petals 
of  a  rose,  are  formed  by  punctures  through  the  outer  envel- 
ope. Through  them  the  urchin  protrudes  fleshy  suckers  or 
tubes.  If,  for  example,  a  sea-urchin  is  placed  in  a  glass  filled 
with  sea-water,  it  is  soon  seen  to  protrude  a  multitude  of  slen- 
der fleshy  threads,  each  of  which  is  tipped  with  a  little  knob. 
These  soon  come  in  contact  with  the  glass  to  which  the  knob 
adheres,  on  the  principle  of  an  exhausted  receiver.  By  means 
of  this  adhering  apparatus,  it  moves  itself  forward  or  back- 
ward. In  technical  language,  the  surface  from  which  these 
fleshy  threads  protrude,  are  called  ambulacral  areas,  and  the 
spaces  between,  interambulacral  areas.  Nothing  can  be  seen 
of  these  threads  when  the  animal  is  dead.  All  its  soft  parts 
are  strictly  encased  in  a  box  of  kard  shell  substance,  which 


NORTH-CAROLINA   GEOLOGICAL    SURVEY. 

has  received  the  name  of  Test,  or  Shell.  The  patterns-  of  these 
different  areas  vary  in  form  and  proportion,  and  hence  are 
used  as  characteristics  of  genera  and  species.  The  test  is  also 
covered  with  spines  of  different  forms  and  sizes.  These,  too, 
are  formed  after  different  patterns,  their  shafts  being  sculp- 
tured differently  in  every  species.  Their  spines,  and  the 
mode  they  are  attached  to  the  shell,  the  character  of  their 
surfaces,  the  position  of  their  oral  and  excretory  orifices,  fur- 
nish the  characters  upon  which  the  families,  and  lesser  sub- 
divisions of  this  class  are  founded. 

FAMILY   CIDARIDAE. CTDARIS    MITCHELLII. N.    S.       (Fig.    237.) 

Test  thick,  circular  or  turban  shaped  ;  flattened  above  and 
below;  ambulacral  areas  narrow,  and  provided  only  with 

minute  tubercles,  in  double  rows, 
FIG.  237.  and  three  in  each ;  interambulacral 

areas  nearly  four  times  as  wide  as 
the  former,  and  furnished  with  two 
distinct  rows  of  large  primary  tu- 
bercles, with  about  eight  in  a  row, 
including  the  smaller  ones  upon  the 
disks;  tubercles  perforated;  inner 
rim  surrounding  the  tubercle, 
smooth ;  outer,  bearing  small  sub- 
ordinate spines,  giving  it  a  crenulated  appearance ;  miliary 
zones  wide,  and  covered  with  small  close  set  unequal  granules  ; 
poriferous  zones,  unigeminal,  and  separated  by  nearly  plane 
ridges ;  spines  unknown ;  apical  disk  unknown ;  mouth  open- 
ing, appears  to  be  large,  but  too  much  broken  to  determine 
its  characters. 

Belongs  to  the  eocene,  and  accompanies  the  remains  of  the 
Zeuglodon. 

Dedicated  to  the  lamented  Prof.  Mitchell  of  the  University 
of  Chapel  Hill. 

CIDARIS   CAROLINEN8IS. — N.    S.       (Fig.    238.) 

Test  rather  thick,  circular  and  somewhat  oval.    Ambulacral 
areas  narrow ;  somewhat  undulating,  supporting  two  row*  of 
21 


306 


NORTH-CAROLINA   GEOLOGICAL   SURVEY. 


FIG.  238. 


small  tubercles  with  two  in  a  row,  and  interspersed  with 

minute  ones,  which  appear  in 
some  places  to  be  arrayed  in  sub- 
ordinate rows ;  interambulacral 
areas  wide,  covered  with  small 
subequal  and  rather  prominent 
tubercles,  among  which  minute 
granules  are  scattered ;  area  about 
four  times  as  wide  as  the  former  ; 
plates  pentagonal,  supporting  two 
rows  of  large  perforated  primary 

tubercles,  surrounded  by  plain  circular  zones ;  miliary  zone 
concave  or  depressed.  Poriferous  zones  narrow  ;  pores  uni- 
geminal;  outer  oblong;  the  inner  circular;  margin  of  the 
small  plates  between  them  marked  with  an  elongated  depres- 
sion. The  upper  and  lower  sides  crushed. 

Belongs  to  the  eocene,  and  accompanies  the  former. 
Figure  105  represents  the  jaws  of  an  Echinoderm,  p.  246. 
The  separate  pieces  of  the  test  and  jaws  are  quite  common 
in  an  eocene  bed  in  Craven  county.  They  belong  to  the 
upper  part  of  the  bed,  and  seem  to  be  confined  to  a  space 
about  two  feet  thick. 

FAMILY    CIDARITAS. ECHINUS   RUFFINII. ED.  FORBES.    (Fig.  239.) 

u  Body  sub-depressed  ;  ambulacral  and  interlambulacral ; 

plates  with  several  primary  tu- 
bercles on  each  closely  ranged, 
having  circles  of  secondary  tu- 
bercles surrounding  their  bases; 
rows  of  pores  very  oblique, 
with  three  pair  of  pores  in  each 
row,  the  uppermost  distant  from 
the  other  two.  Beneath  con- 
cave ;  mouth  broad ;  widely 
notched  opposite  each  avenue.'7 
Ed.  Forbes.* 


FIG.  239. 


Journal  Geological  Society,  vol;  i,  p.  426< 


NORTH-CAROLINA   GEOLOGICAL   SURVEY. 


307 


FIG.  240-'l. 


Found  in  the  miocene  beds.  Four  views,  #,  Echinus  Ruf- 
finii,  viewed  from  above ;  5,  mouth  ;  c,  spinegerous  tubercles  ; 
d,  ambulacral  plates,  and  arrangement  of  pores :  #,  J,  natural 
size,  <?,  dj  enlarged. 

FAMILY  CLYPEASTARIDAE. — ECHINOLAMPAS  APPENDICULATUS. N.  S. 

(Fig.  240-'l.) 

Test  thin;  body  oval,  depressed; 
margin  thick  or  rounded ;  somewhat 
elongated,  wider  anteriorly  than  pos- 
teriorly ;  ambulacra  narrow,  open  at 
their  extremities ;  sub-petaloid  ;  pores 
connected  by  furrows  ;  mouth  trans- 
verse ;  excretory  orifice  horizontal, 
marginal ;  inadriporiform  plate  ex- 
centric  ;  apical  disk  occupied  by  a 
sub-cordate  sculptured  plate,  furnish- 
ed with  a  pentangula  opening,  in  the 
centre  of  which  there  is  a  pore ;  are- 
olse  more  numerous  below  than  above  • 
area  around  the  mouth  inflected. 


ECHINOCYAMUS   PARVUS. — N.    S.       (Fig. 

Test  small,   oval,   with    rounded   sides;    avenues   dorsal; 

mouth  sub-central,  rounded,  large,  with  a  crenulated  margin; 

vent  between  the  mouth  and  hinder  margin  ;  genital 

FIO.  244.    p0res  apparently  four.     Figure   natural  size.     The 

£7}       mouth  is  large  in  proportion  to  the  size  of  the  body 

arid  the  vent  is  situated  half  way  between  the  mouth 

and  margin.     Eocene  of  Craven. 


308 


NORTH-CAROLINA   GEOLOGICAL   8URVET. 


FIG.  246. 


SCUTELLA   LYELLH.-— (Fig.  246.) 

Shield  small,  sub-circular,  flat,  scarcely 
convex  above  ;  below  slightly  concave ; 
ambulacra  open  towards  the  margin  and 
terminating  in  four  pores ;  in  that  direc- 
tion mouth  small ;  vent  near  the  margin. 
Eocene,  Wilmington. 


SCUTELLA. — (Fig.  247-'8. 

Figures  247-' 8  represent  a  common  fossil  of  the   eocene 
of  Craven  county.     247  inferior  face,  showing  the  relative 

FIG.  247-'8. 


position  of  th^  mouth  and  excretory  orifice.  Figure  248  is 
profile  view  of  the  same.  The  apical  summit  is  before  the 
genital.  Since  its  discovery  no  opportunity  has  been  furnish- 
ed by  which  I  could  obtain  a  comparison  with  the  forms  al- 
ready known  and  described  by  the  palaeonlologists  of  this 
country.  Wadsworth's  eocene  marl,  Craven  county. 


NORTH-CAROLINA   GEOLOGICAL   SURVEY, 


309 


FAMILY   SPATANOIDAE — GONIOCLYPEUS  SUBANGULATUS. — N.  G. 

(Fig.   242.) 

Test  thick,  sub-conical,  covered  with  small  spines,  anterior 
and   posterior  areas  somewhat  unequal;   margin   and  base 


FIG.  242. 


FIG.  243. 


somewhat  pentangular;  posterior  or  anal  orifice  lateral,  or 
upon  the  superior  face ;  interambulacral  area  grooved,  with 
the  continued  area  beneath  projecting  ;  interambulacral  areas 
sub-angulated ;  mouth  rather  narrow  or  small,  central;  peris- 
tome  angular,  and  surrounded  by  five  angular  prominences, 
which  terminate  in  the  interambulacral  areas,  between  which 
is  a  rosette,  perforated  by  seven  pairs  of  pores,  with  three  odd 
ones  at  the  end  of  each  petal ;  ambulacra  petaloid  and  closed; 
the  prolonged  zone  provided  with  alternating  pores  as  far  as 
the  base;  pores  connected  by  oblique  grooves;  interambu- 
lacral wide ;  plates  large,  and  nine  or  ten  in  a  column. 
Figure  243,  rosette  enlarged. 

OBSERVATIONS. — The  ambulacral  areas  are  narrow,  but  the 
poriferous  zones  are  rather  wide ;  and  the  interambulacral 
areas  are  about  four  times  as  wide  as  the  ambulacral.  The 
genital  plates  are  indeterminate,  but  the  pores  are  large  and 
the  occular  small,  and  appear  to  be  mere  indentations ;  buc- 
cal  area  ornamented  with  a  rosette;  petals  transversely 
wrinkled ;  pores  elongated ;  the  anterior  lateral  plates  appear 
to  have  eleven  pairs  of  pores  instead  of  seven.  The  genus  is 
closely  related  to  Cassidulus  of  Lamark,  but  the  pores  are 
united  by  grooves.  Eocene,  Wardsworth  marl,  Craven  co. 


310 


NORTH-CAKOLINA   GEOLOGICAL   STTKVEY. 


AMPHIDETUS    VIRGINIANUS.* FORBES.       (Fig.  245.) 

"Body  broadly  ovate,  elevated  and  truncate  posteriorly ; 
back  oblique  ;  dorsal  impression  lanceolate  ;  scutab  area  very 
slightly  excavated ;  ambulacral  spaces  broad,  triangular,  de- 
pressed ;  interambulacral  spaces  slightly  convex ;  anteal  fur- 
row broad  and'  shallow,  sides  slightly  gibbous ;  sub-anal  im- 
pressions broadly  ob-cordate  ;  post-oral  spinous  space  broadly 
lanceolate. — Edw.  Forbes." 

FIG.  245. 


<r,  lower  area ;  &,  upper  area ;  c.  posterior  area,  showing 
the  relation  of  the  sub-anal  impression.  Usually  found  in 
fragments  in  the  miocene  of  North-Carolina. 


*  Journal  of  the  Geological  Society,  Vol,  1,  p.  425, 


NORTH-CAROLINA   GEOLOGICAL   SURVEY. 


311 


FIG.  246.    FIG.  247. 


ORDER    CRINOIDEA. — MICROCRINUS    CONOIDEUS. — N.    G. 

(Figs.  246  &  247.) 

Body  conical ;  sub-pentangular  at  base  ; 
areas  five,  oblique ;  pores  six  or  seven  to 
J^       ^p        each,  alternating  and  arranged  in  rows, 
separated  by  a  ridge ;  apical  pores  five, 
base  wide ;  beneath  concave;  concavity  intersected  by  five 
bars,  which  descend  and  meet  in  the  center ;  spaces  between, 
triangular,  terminating  above  in  the  apical  pores. 

Figure  247  shows  the  base  with  the  intersecting  bars  and 
triangular  spaces  between. 

I  am  unable  to  determine  whether  the  head  is  supported  on 
afoot-stalk;  the  joints  of  a  crinoid,  however,  are  numerous 
in  the  marl  in  which  this  curious  species  is  found. 

Eocene  of  Craven  county,  and  associated  with  Echinocya- 
mus  Parvus. 


FIG.  248. 


FIG.  249. 


BRYOZOA. LUNULITES    BENTICULATA. —  (FigS.    248    &   249.) 

"  Conical ;  cells  inalternate,  oblong  externally,  interior  coni- 
cal, nearly  vertical  to  the  two  surfaces  of  the  polypidom ; 

margin  of  the  cell  in  its  immature 
state  open  and  denticulated ;  when 
mature,  covered ;  mouth  near  the 
distal  extremity;  semicircular  when 
imperfect,  circular  when  perfect ; 
gemmuliferous  chamber  at  the  dis- 
tal end  of  the  cell,  opening  round, 
concave  surface  furrowed,  irregular 

and  minutely  granulated."  *     Miocene,  and  common  to  most 
of  the  beds  upon  the  Neuse  and  Cape  Fear. 

Fig.  249,  enlarged  view  of  the  fossil,  showing  the  arrange- 
ment of  the  cells,  and  the  small  Figure  its  natural  size. 

LUNULITES   CONTIGUA. — FigS.    250    &   251. 

The  figures  exhibit  casts  of  the  concave  surface  of  the 


0 


*  Lonsdale,  miocene  corals  from  N.  America,  Journal  Geol.  Society,  TO!.  1,  p. 
Ml. 


312 


NORTH-CAROLINA   GEOLOGICAL   SURVEY. 


coral.  Fig.  251,  cast  of  the  concave  surface  natural  size; 
Fig.  250,  magnified  view  of  a  portion  of  the  surface.  Eocene 
Wilmington. 


FIG.  250. 


FIG. 


FIG.  253. 


LUNULITES   OBLONGUS. N.    8.       (FigS.    252    &    253.) 

Polypidom  small,  conical ;  cells  arranged  along  a  straight 
line,  from  the  base  to  the  margin ; 
open  cells  show  that  they  are  near- 
ly quadrangular;  the  closed  cells 
do  not  show  an  orifice ;  there  is  a 
simple  film  spread  over  the  cell, 
and  the  margins  are  simple  and 

£ ^  unlike  den ticulata.  Fig.  253,  great- 

ly enlarged  view  of  the  cells ;  small  figure  shows 
the  natural  size  of  the  fossil. 

DISCOPORELLA   UMBELLATA. — (FigS.    254:    &   255.) 

It  is  impossible  to  discover  any  difference  between  our 
Discoporella  and  that  of  the  miocene  of  France ;  the  cells 

have  two  orifices  at  op- 
posite acute  angles,  and 
the  same  arrangement 
of  cells.  Fig.  255  great- 
ly enlarged.  This  figure, 
however,  fails  to  give  a 
clear  and  correct  view  of 
the  fossil.  A  reference 
therefore,  to  Pietet's  PL  XC,  page  15,  is  necessary. 

The  small  lunulites  begin  to  form  at  the  apex,  and  for  this 


FIG.  255. 


FIG.  254. 


tfORTH-CAROLINA   GEOLOGICAL   STJEVEY. 


313 


Fio.  256. 


purpose  they  attach  themselves  to  a  grain  of  sand,  which  will 
generally  be  still  fqund  at  the  point  of  growth  ;  some  of  the 
miocene  ones  are  nearly  half  an  inch  in  diameter. 

POLYPAIRA. ASTRAEA    BELLA. (Fig.    256.) 

The  stars  are  polygonal,  variable,  rather  deep,  lamellar 
lamellae  twelve,  with  alternating  ones,  denti- 
culated, contiguous,  or  separated  by  their  par- 
titions. 

Common  in   the  miocene  incrusting  shells, 
and  various  bodies  found  in  a  marl  bed. 

ASTKAEA. — (Fig.  256a.) 

Irregularly  branched  ;  stars  deep  and  rather  distant,  though 
in  some  places  contiguous  as  in  the  Bella;  intermediate  spaces 
without  pores,  but  bordered  by  lines  to  which  the  lamellae 
extends ;  lamellae  denticulated,  as  in  A.  Bella,  and  provided 
also  with  the  same  number,  and  similarly  arranged.  Miocene. 

Fro.  256a. 


314  NORTH-CAROLINA   GEOLOGICAL   SURVEY. 

The  foregoing  sketch  of  the  fossils  of  the  marl  beds  of  the 
eastern  counties,  is  far  from  being  complete.  Numerous  spe- 
cies still  remain  unnoticed  and  undescribed.  It  seemed  to  be 
desirable,  however,  on  many  accounts,  to  illustrate  some  of 
the  interesting  contents  of  these  beds,  which  are  truly  the 
only  historical  mementoes  which  now  remain  to  us  of  the  ages 
during  which  they  lived.  It  will  appear,  on  examination,  that 
I  have  placed  by  far  the  largest  number  of  species  in  the  mi- 
ocene.  I  have  thus  placed  them  because  the  shell  marl  beds 
contain  so  large  a  number  of  the  acknowledged  miocene  fos- 
sils of  Virginia ;  and  besides,  there  are  many  which  replace 
miocene  fossils  of  Europe. 

In  conclusion,  it  is  due  to  myself  to  remark,  that  the  cir- 
cumstance under  which  many  of  the  determinations  have  been 
made,  rendered  it  impossible  to  consult  authorities,  and  hence 
it  may  turn  out  that  many  species  which  have  been  marked 
as  new,  will  prove  to  be  old  ones  already  described.  The 
course  I  have  pursued  may  have  been  injudicious,  and  hence 
may  open  the  way  for  censure;  still,  under  the  circumstances, 
I  deemed  it  the  best  I  could  pursue. 


ADDITIONS    AND    CORRECTIONS. 


I.— FIGURES  AND  NAMES  OP  SPECIES:— 
Page  205,  for  otololite  read  otolite. 
"    242,  fig.  90,  read  Galeocerdo  Egartoni. 
"    241,  fig.  84a  is  Sphyrna  denticulata. 
"      "     "    82a  and  83a,  Galeocerdo  contortus. 

"    243.    It  is  possible  Trygon,  fig.  94,  should  be  referred  to  Myliobatis. 
«    245.— Fig.  105  is  the  valve  of  the  genus  ScalpeUum  of  the  class  Oir- 

ripedes. 
"    261,  fig,  139.— This  is  not  Erato  laevis,  but  is  closely  allied  to  E. 

Maugeriae,  of  the  coralline  crag. 

"    268. — Fig.  159  resembles  Cerithium  adversum  of  the  English  crag. 
"    290. — For  J/ucenidae  read  Lueinidae. 
"    291. — Place  a  period  before  Brugiere. 
«      «  — For  Pennsylvania  read  Pennsylvanica. 
"      "      sencond  line  from  bottom,  for  multiMneata  read  multilineato. 
"    292. — For  Venerlda  read  Veneridae. 
«      "  — For  Tridaenoides  read  Tridacnoides. 
"    293.— For  Cribrari  read  Cribraria. 
"      "  second  line  from  bottom,  for  pramagna  read  permagna ;  for  metta- 

striata  read  metastriata. 
"    29'4.— For  Cytherca  read  Cytherea. 
"      "      For  reporta  read  reposta. 
"    295. — For  Artemes  read  Artemis. 

"    296.— Fig.  224  shows  the  hinge  of  Artemis  tranversus ;  and  read  Ar- 
temis for  Artemus. 
"      "     sixth  line  from  bottom,  for  TELLIMIDAB  read  TELLINIDAE;  and 

ninth  line,  for  Tiphonal  read  Siphonal. 
"    297. — For  P.  Sammobia  read  Psammobia. 
"    306. — For  Cidaritas  read  Cidarites. 

"    307,  second  line  from  top,  for  Spinigerom  read  Spinigerous. 
"    311. — Bryozoa  should  have  been  placed  under  an  independent  head, 

as  a  subdivision  of  Molusca  and  not  under  Radiata. 

Certain  figs,  have  been  placed  wrong  side  up,  particularly  Scutella.  fig.  247 — '8. 
In  the  Eocene  of  Craven  county,  I  have  found  the  palatine  teeth  of  the  Saurodon,  or 
Saurocapalhus,  and  also  fragments  of  a  Xiphioid  fish,  as  the  prolonged  premaxi- 
lary  of  a  sword  fish. 
Rctinasphalt  occurs  in  the  marl  of  Duplin  county. 


THIS  BOOK  IS  DUE  ON  THE  LAST  DATE 
STAMPED  BELOW 


AN  INITIAL  FINE  OF  25  CENTS 

WILL  BE  ASSESSED  FOR  FAILURE  TO  RETURN 
THIS  BOOK  ON  THE  DATE  DUE.  THE  PENALTY 
WILL  INCREASE  TO  SO  CENTS  ON  THE  FOURTH 
DAY  AND  TO  $t.OO  ON  THE  SEVENTH  DAY 
OVERDUE. 


Wl     3    18ft 

Uwl  ,1  o  iy3>2  *»*' 

i  oraK'fiOR  3CT 

REC'D  LD 

APR  9      Wfifl 

LD  21-95wi-7,'37 

YC  &5880 


MO139  Q3dlAIViS  SV  300 


anp  o\  joud 
sAep  fr  apeuu  aq  ABLU  seBjEgoej  pue 

dlbN 
s>|ooq  6ui6uuq  Aq  pa6jeij09j  eq  ABLU  SUBOI 


6u!||BO  Aq  P9M9U8J  eq  ABLU  SUBOI 
SAVQ  L  d31dV  a311V03d  39  AVIA1  SXOOQ 


VO  ' 

JO  Al!SJ9AIUn 


AlHIOVd  AdVdan  1VNOI03d  Nd3HldON 
am  o\  jo 

AjBjqn  BjUJOJJIBQ  JO  Al!SJ9A|Un 

AUB  |0  >|S9p  uouB|nojp  egj  Qi  Ndni3d