Journal of the Institute of Petroleum, Vol. 29, No. 234

Vo l.

29. N o. 234.

1943.

FRACTIONAL DISTILLATION ON BINARY MIX­

TURES. NUMBER OF THEORETICAL PLATES AND TRANSFER UNITS.*

B y A. J . V.

D.Sc., M.I.Chem.E., A.M.I.Mech.E., F.I.C ., P .In st.F ., F .In st.P et.

Su m m a r y.

For binary mixtures with constant relative volatility, the number o f transfer units can be calculated in a form which readily permits o f comparison with the number of theoretical plates required for the samo separation. The computation can in both cases be greatly facilitated .by using a simple graphical method. A ny separation w ith finite reflux can bo exactly converted into an equivalent separation with total reflux by using a modified relative volatility and modified compositions.

I

is useful to be able to m ake a ready comparison of th e num ber of theoretical plates an d of tran sfer u n its required for a given separation by fractional distillation. F o r binary m ixtures, in which th e com ponents follow R ao u lt’s Law and th e relative volatility m ay be assum ed constant, a num ber of analytical m ethods of calculation are available. F or the p articular case of to ta l reflux, Chilton an d Colburn 1 have derived an equation giving th e num ber of transfer units, while Fenske 2 an d U nder­

wood 3 have given equations for th e num ber of theoretical plates. F or finite reflux, Colburn 4 has given equations for th e num ber of tran sfer units, and Smoker 5 for th e num ber of theoretical plates. Dodge and Huffm an 8 an d H ausen 7 have calculated the num ber of theoretical plates required, assum ing a differential change in composition from p late to plate, and consequently their equations actually give the num ber of transfer units. F or th e general case of finite reflux, these various equations do not readily perm it of comparison between th e num ber of theoretical p lates and th e num ber o f tran sfer units, an d involve som ew hat laborious com puta­

tions. B y m aking use of the transform ation o f co-ordinates used by Smoker 5 in deriving his equation for th e num ber of theoretical plates, somewhat sim ilar an d readily comparable equations can be derived for th e num ber of transfer units. B y m eans of a simple graphical construction th e work of com putation in both cases can be g reatly simplified. Also, an y case of finite reflux can, by a fu rth er transform ation, be reduced to an equivalent separation w ith to ta l reflux.

The m ethod used by Smoker 5 for calculating th e num ber of theoretical plates is illu strated in Fig. 1. R eferring to axes O X and O Y, th e equili­

brium curve is represented by th e equation

OCX-

y = 1 + (a - I ) * ...( )

The operating line is represented by th e equation

y — mx

b ...

* R eceived 7th April, 1943.

1 4 8 U N D E R W O O D : FR A C TIO N A L D IS T IL L A T IO N O N B IN A R Y M IX T U R E S .

The origin of co-ordinates is now transferred to O', where th e operating line and the equilibrium curve intersect. I f th e co-ordinates of th e point of intersection, referred to the original axes O X and O Y , are (kv rnkx + b),

” * , + t ° r + ( f T i f e ... (3) '

or m(a — l)Iq2 + [wi + b(a — 1) — aJZq - f b = 0 . . (4)

E quation (4) has two solutions, representing the two points of intersection

of the operating line with the equilibrium curve (continued, if necessary,

beyond the norm al range of x = 0 to x — 1). In the following, k1 is tak en

as th e lower of the two values given by equation (4). F o r any practical

distillation problem, /j1 will lie between 0 and 1 for th e operating line of the

rectifying section, and will have a negative value for the operating line of th e

stripping section.

U N D E R W O O D

:

149 Taking new co-ordinates w ith th e point of intersection O' as origin

x' — x — iq

y' = y — (mky + b)

Substituting in equations (2) and (1), th e equation of the operating line becomes

y ' = m x ' ... (5)

and th e equation of th e equilibrium curve becomes

y' + m k i + h

= rqT(”(^t)(t')+jfc1j :••■■(«)

S ubtracting equation (3) from equation (6) and putting

1 (a — l)&i = c ...(7)

.

g l v c s ,J ~ - i K ...< 8 )

which is the equation of th e equilibrium curve referred to the new origin of co-ordinates, O'.

Using th e notation of Colburn,8 the num ber of transfer units based on change in vapour concentration is given by

N,oo

[y•

-ir J —

...(9)

where y* is the equilibrium value of y corresponding to x and y is the actual value.

Transferring to the new co-ordinates

y* - I f = y'* — (mk! + b) — {y' — {mkx + b)}

=

y'* - y' From equations (8) and (6),

,

y * - y

= — ^ + ^ -1),'

and

Equation (9) then becomes

1 5 0 U N D E R W O O D : FR A C TIO N A L D IS T IL L A T IO N ON B IN A R Y M IX T U R E S .

which gives the num ber of transfer units required to effect a change in composition from x„' to x 0'.

The equation derived bj*- Smoker 5 for th e num ber of theoretical plates,

N p = •

lost*

me“

log, 3 + log.

1

me“ ^■

( H )

For to tal reflux, the operating line is y = a;, and intersects the equilibrium curve a t the origin 0 , so th a t rn = 1, Zq = 0, c = 1, and x' = x. E quation (10) then becomes :

£ ; i ... < 1 2 >

- l0&S + l08- { r F q l + r h •k* { r = J ;}

. . . „ 3 ,

or

E quation (12) is the same as the one derived by Dodge and Huffm an 6 for to tal reflux, when calculating th e num ber of theoretical plates on the assum ption th a t there is a differential change in composition from plate to plate. E quation (13), after m aking th e substitution x — y, is the same as th a t given by Chilton and Colburn 1 for th e num ber of transfer un its for to tal reflux.

I f now a further transform ation of co-ordinates is m ade by p u ttin g

a — me“

then equation (10) becomes :

!) x , _ mc{a — l)(x — Zq)

mcr

(14)

$ ■ - U T S ? p r

Now, putting

mc~

N oo

(15)

(16)

E quation (16) is of exactly th e same form as the to ta l reflux equation

(12) w ith x" instead of x and ¡3 instead of a. In other words, th e calculation

for finite reflux by equation (16) is identical w ith the calculation for to tal

reflux if a modified relative volatility and a modified co-ordinate x " is used,

where and x " are defined b y equations (15) and (14).

U N D E R W O O D : FR A C TIO N A L D IS T IL L A T IO N ON B IN A R Y M IX T U R E S . 151

The same relation holds good for th e calculation of theoretical plates.

F or to ta l reflux, equation (11) reduces to :

iVP = p i - . log, { ^ ïü H

log, a c U „ ( l - . r 0)J (17)”

as given by Fenske 2 and Underwood.3

If, in equation (11), substitution is made by means of equations (14) and (15), it becomes :

N —

* low \ X° ^ Xn ^1

N l’

log, p ' l.rn" ( l - *o")J ’ '

E xactly as before, th e calculation for finite reflux is identical w ith a calculation for to tal reflux when th e modified relative volatility p and the modified co-ordinate x " is used.

The conversion of a finite reflux calculation to a to ta l reflux calculation can also be derived from equations (5) and (S) for th e operating line and equilibrium curve respectively. Substituting

<“ >

- i m

equation (5) then becomes :

y " = x " ... (20)

which is th e equation of an operating line for to tal reflux, and equation (8) becomes :

„ " =

, , f " • • • (21,

1

+ ( — 2 - i V ' 1 + ( p

Kmc2 )

which is th e equation of an equilibrium curve w ith relative volatility p.

There is a simple geometrical interpretation of th e transform ation from th e (x , y) system of co-ordinates to th e (x", y") system of co-ordinates.

In Fig. 1 th e operating line (for finite reflux) intersects th e equilibrium

curve a t O' (kv m kx + b) and D (Ic2, m k2 + b), where kl and k2 are the two

roots of equation (4). The area OACB represents th e usual range of values

(referred to origin O') are each transform ed by a m ultiplying factor so th a t

the lengths of O'E and D E become 1, th e slope of th e line O’D will become

45°. The equilibrium curve, which is a rectangular hyperbola, will be

transformed into another rectangular hyperbola. The area O'DCO'

between the operating line and the equilibrium curve will th en obviously

be of the same ty p e as th a t between an operating line and equilibrium

curve for a case of to tal reflux. The transform ations required to convert

the co-ordinates (x', y') to (x", y") are given by equations (14) an d (19),

which can also be dem onstrated geometrically. The length O'E is (k2 — Aq) As Aq and Aq are th e two roots of equation (4)

_ a — b(a — 1) — m

«'l i ^2 —

1 5 2 U N D E R W O O D : FR A C TIO N A L D IS T IL L A T IO N ON B IN A R Y M IX T U R E S .

and AqAq

m( a — 1) b

m( a — 1)

so th a t (Aq - AJ2 = {g

. (22)

1 m r y u . — I ) “5 m ( a — 1 )

From equations (3) and (7)

, (a — mc){c — 1) c(* - 1)

and, on substituting this value of b, equation (22) gives

h - 1

= ~ T mci i ... (23)

The length O 'F, which is (k„ — ¿q) in the (x y ' ) system of co-ordinates, has to become unity in the (x ", y") system of co-ordinates, giving the re­

lation

„

(a —

„

x " = =--- =- or---x = — -^ . x

«2

— Aq mc(a — 1) which is equation (14).

Similarly, in order th a t th e length D E , which is m (k2 — Zq) in the (x't y') system of co-ordinates, should become u n ity in th e (x", y") system of co-ordinates

,,

, (a — me2)

y = M k ^ n r )

or

which is equation (15).

A simple graphical construction can be used to facilitate th e num erical com putation of transfer units or theoretical plates from equations (16) and (18). The equilibrium curve and th e operating line being drawn, th e abscissae of th e points of intersection give iq and k2. Then

xn" — Xn

= ^ a n d sim ilarly for x0".

2 1 2 1

B = is then readily calculated from c = 1 + (a — l)Zq.

me2

The further calculation then becomes the same as for a case of to tal reflux.

H aving thus determ ined the factors required for conversion to the to tal reflux form, th e com putation of equation (18) can be readily m ade by using the appropriate nomograms published by Smoker 9 and by Underwood.3

Actually it is only necessary to determ ine Aq or k2, as, if one of these is

known, the other is readily found from th e relation kxk2 = . I t is

X

Fi g. 2 .

then being a t the point of intersection D. c and c' are th e roots of the equation

?nc2 + c{Z»(a — 1) — a — 7«} + a = 0

obtained by substituting for k in equation (4).

Thus cc' = — so th a t

m me

a

and a ~ mc~ _ _ a ~ ?»c'2

mc(a — ]) ?nc'(a — 1)

aq" can also be obtained purely geometrically as shown in Fig. 2. O'A convenient to take Aq as the lower value and A2 as th e higher value, so th a t (Aq — Aq) is always positive. F or the operating line for th e stripping section k1 will be negative. The value of c in equations (10) and (11) corresponds to Aq. E xactly similar equations can be obtained in which c is replaced by c' = 1 + (a — l)Aq, th e origin of the (x', y') co-ordinates

Y

YI

U N D E R W O O D : FR A C TIO N A L D IS T IL L A T IO N O N B IN A R Y M IX T U R E S . 1 5 3

1 5 4 U N D E R W O O D : FR A C TIO N A L D IS T IL L A T IO N ON B IN A R Y M IX T U R E S .

an d OE are draw n to intersect in J . G is any point on O'E, and has the abscissa x n' referred to origin O' or x n referred to origin 0 . I f a line joining

A simple geometrical construction, can also be used to give ¡3. J 1 is a

point on O 'Y ' having th e same ordinate as D, and a line is draw n joining it to E. The line F E intersects th e equilibrium curve a t L . The ratio of the distance of L from D E to 'its distance from O 'Y ' (i.e., E M : MO") is th en equal to V p . P u ttin g O'E (i.e., k2 — /q) equal to I, the line E F referred to axes O’X ’ and O 'Y ' is

The intersection of this line w ith th e equilibrium curve of equation (8) is given by :

For

O'G ~ O 'J ~ O'E

Since O'G = x n', A O = 1 and O'E = k2 - kv

//

y ' — m(l — x')

Also • • • (23)

(24)

S ubtracting equation (24) from equation (23) gives , c(a — me2) — mc2(a — 1).t' W/’-v' --

1______ 1_____ 1_

mCX = ( a -

l){c

mcx'

or, using equation (23),

(25) Dividing equation (24) by equation (25),

I — x '

a

I t m ay be noted th a t, if a tangent to th e equilibrium curve is draw n a t

equal to p. F or, differentiating equation (1), th e slope of th e tangent

/

U N D E R W O O D : FR A C TIO N A L D IS T IL L A T IO N ON B IN A R Y M IX T U R E S . 1 5 5

a t O' is equal to ; or -g. This is not a practically convenient {1 + (« — 1 )M 2 e*

m ethod of determ ining p, on account of the inaccuracy involved in drawing th e tangent. The point is, however, of interest, as C olbu rn8 has proposed an approxim ate m ethod for calculating the num ber of transfer units, which is based on th e assum ption th a t th e ratio of th e slope of th e tan gen t to the equilibrium curve to th e slope of the operating line is constant.

The equation given by Colburn 8 for the num ber of transfer units based on liquid concentration

; <26) can be integrated by th e same m ethod as was used for obtaining N

For x — x* — (x' + h) — {x‘* + h) = x' — x'* and from equations (5) and (8),

x' — e2y'

= x

c(a — 1 )y' ' a — mc(a Also dx = dx', so th a t equation (26) becomes :

mc( a — 1 )x'}dx'

l)x '

N,_OL [*•'

{;

Jx„' & {(a me*

me“

me2) — mc(a — l)æ'}

dx' mc(a. — 1 ). dx' \ x '

(a — me2) — mc(a — l)x ' ) { A Imc“

Nql — me“

me“

2) — mc(a — l)x ' 1 mc( a — 1)

-I Ô- •

1 - Using equations (14) and (15)

a — mc2

a — me2

f l - xn 1 U - V 'J

(27)

(

28

The corresponding equation for to ta l reflux is derived b y substituting a for p and x for x " . E quation (28),- together w ith equations (10) and (18), gives th e relation

Nog

+ N

- N r . | ± i . log, p . . . . (29) which holds for any separation. F or to ta l reflux p is replaced by a.

F or two components w ith low relative volatility, th e value of a approaches 1 and the values of c and m are also nearly equal to 1 so t h a t p is also nearly equal to 1. P u ttin g p = 1 -j- S where S is small, th en to th e first degree of approxim ation, equations (16), (18) and (29) all reduce to th e sam e form

- N r = i . .„g. ( g j i . .

(30)'

1 5 6 U N D E R W O O D : FR A C TIO N A L D IS T IL L A T IO N ON B IN A R Y M IX T U R E S .

References.

1 Chilton, T. H ., and Colburn, A. P., Industr. Engng Chem., 1935, 27, 255-260.

2 Fenske, M. R., ibid., 1932, 24, 4S2-185.

3 Underwood, A. J. V., Trans. Inst. Chem. Eng., 1932, 10, 112-152.

1 Colburn, A. P., Indnstr. Engng Cliem., 1941, 33, 459-467.

6 Smoker, E . H., Trans. Amer. Inst. Chem. Eng., 1938, 34, 165-172.

6 Dodge, B . P., and Huffman, J . R ., Industr. Engng Chem., 1937, 29, 1434-1436.

7 Hausen, H ., “ Der C hem ieingenieur,” Vol. I, Part 3, 117—11S (Leipzig, 1933).

8 Colburn, A. P., Trans. Amer. Inst. Chem. Eng., 1939, 35, 211-236.

9 Smoker, E . H ., Industr. Engng Chem., 1942, 34, 509-510.

1 5 7

RECENT DEVELOPMENTS IN WELDING.*

By C. W.

M .Inst.W -t

So great is th e pressure of present activities th a t there is little or no opportunity to study w hat certain other industries are doing. N everthe­

less, it is m ost helpful to look beyond the bounds of our direct interests, in order to ascertain if th e progress th a t others are m aking can be adapted to solve the problems w ith which we are faced individually.

One of the m ost rem arkable developments brought to the forefront by the stimulus of war is th a t of successful welding. I t is to the credit of a com parative handful of far-sighted men th a t research work in this direction had been intensified to an ex tent th a t enabled war-time needs to be m et w ithout delay, although m any requirements, were so novel and exacting th a t th ey had n o t been attem pted on a commercial scale before hostilities commenced.

Engineering progress has provided m any such achievements, b u t few things, if any, have been so widespread in their influence on industry as welding. I t is the key to th e incredibly rapid production of new m erchant ships in America. The same methods are applied w ith equal success to m aintenance and repair, not only to hulls and m arine engines, b ut also to plant of all kinds where com ponents require re-conditioning on account of breakage, wear, or corrosion. A t th e outset it is m ost desirable to emphasize th a t no form of welding calls for greater skill or a more varied background of experience th a n does repair work, since requirem ents are seldom identical.

Even when the m ethod of procedure is agreed between th e operator and metallurgical chemist, and th e requisite strength assured, it is skill alone by which perfect alignm ent is preserved. The usual tolerance allowed is 0-001 plus or minus.

U ntil quite recently only identical m aterials could be welded together.

Then came certain changes in technique which enabled some degree of variation within th e ferrous group. Malleable and cast iron were welded to steel and so on, b u t the limits were somewhat rigidly defined, and it was thought th a t greater variations in the respective coefficients of expansion offered an insuperable barrier.

N ot long ago a process was perfected which m ade a clean sweep of most preconceived notions on the subject; it resulted in th e achieving of a tru ly welded union w ith alm ost any com bination of metals. This im portant fact is not generally realized, and so far only a few industries, among them the electrical trad e in particular, have grasped its im portance. Quite a p a rt from repair work, for which th e system was prim arily evolved, its influence on new production will be potent, to say th e least. Aluminium can be welded to steel if needs be, b u t however striking th e dissim ilarity between the two metals, tests to destruction, by exerting an increasing pull until something m ust break, always cause the weaker p arent m etal to fail first whilst the weld invariably holds fast.

* Received 30th March, 1943.

t

1 5 8 B R E T T : R E C E N T D E V E L O P M E N T S IN W E L D IN G .

Corrosion is an influence all too well known to those who handle petroleum . I t m ay be th a t m etal spraying, which is an off-shoot of welding, will be used far more widely to com bat this trouble in the future. The spraying of zinc in place of the ordinary m ethod of galvanizing is a most useful possibility, particularly in cases of erection work in which the act of riveting lias dam aged th e original zinc coating locally.

The m ethod by which spraying is done is relatively simple. A “ gun ” is th e m ost complex item of th e apparatus. I t is rath er larger th a n th e h an d ­ piece of a p ain t spray, and is provided w ith an autom atic feed for wire (or sometimes powder) of th e m etal it is desired to spray. As th e wire or powder is passed through the oxy-acetylene flame it is in stan tly m elted, the molten particles being ejected by a stream of compressed air.

A m a tt surface can be coated w ith th e same facility as th a t common to paint. The surprising point about this process is th a t by th e tim e th e particles of m etal reach their m ark th ey are relatively cold. Tliis is shown by the ability to coat th in paper w ithout scorching it. Although adhesion is mechanical and n o t fusive, as in true welding, adherence is extrem ely good—so good, in fact, th a t this m ethod is used to build u p certain worn components, b u t scientific welding engineers prefer to apply new m aterial in the more usual form of feed-rods, contending th a t in this way the grafted and p arent m etals have a homogeneous u n ity th a t is otherwise impossible.

So far as ferrous m aterials are concerned, gas welding is m ainly confined to cast and malleable iron, whilst the electrical m ethod is reserved for steel, although th e dem arcation is n o t inflexible. In this connection certain lines of progress are im portant and should be remembered, because each has opened up new resources, particularly in regard to m aintenance and repair.

I t seems to have escaped general notice th a t th e thickness of th e m etal th a t can be handled offers no deterrent to repair. ' A t one tim e, n o t so very long ago, a section of m etal 2 inches in thickness was alm ost a borderline case for welding, b u t nowadays a massive casting having a section of 10 inches in, need of repair would not be regarded by the skilled operator as being a t all unusual.

To some exten t this is linked up w ith the fading practice, of pre-heating before welding is started. The object of this time-wasting practice was to avoid strains being set up owing to sudden local expansion brought ab ou t by the heat necessary to accomplish a weld. New m ethods for controlling the heat-flow have been devised, so th a t work can be sta rte d a t once, even to th e exten t of dealing with th e dam aged p a rt whilst it is still in position.

Obviously accessibility m ust be reasonably good and subsequent m achining unnecessary. To those unfam iliar w ith this process it m ay seem impossible to confine th e h eat strictly to the point of repair, b ut it is done so effectively th a t it 'is possible to touch close up to th e seat of fusion w ithout fear of being burned.

I t is hardly surprising, in view of the alm ost infinite variations in modern welding, th a t a very great deal of it, as applied to repair work, is undertaken by specialists who are willing to work to th e closest lim its on a guarantee basis.

Most re-conditioning work is needed for one or more of tfiree reasons—

corrosion, wear, and breakage. A great deal of work is done in correcting

C O M B I N E D C Y L I N D E R - B L O C K A N D C R A N K C A S E O F A L A R O E D I E S E L E N G I N E . P a r t o f th e c ra n k c a s e , a t o n e e n d , w a s b r o k e n o ff . T h e e n g in e b e lo n g e d t o a le a d in g f i r m o f e n g in e e rs , a n d t h e y d e c id e d im m e d ia t e ly t o h a v e th e d a m a g e r e p a ir e d b y w e ld in g .

Fi g. 2.

C Y L I N D E R - B L O C K A N D C R A N K C A S E S H O W N I N F I G . 1 A F T E R I T H A D B E E N R E S T O R E D T O P E R F E C T C O N D I T I O N B Y S C I E N T I F I C W E L D I N G .

ITofacep. 158.

Fi g. 3 .

A G E A R W H E E L , S I M IL A R T O M A N Y O T H E R S U S E D I N A L L K I N D S O F P L A N T .

A number o f teeth have been broken away.

Fi g. 4 .

T H E M IS S IN G T E E T H W E R E B U I L T U P B Y S C I E N T I F I C W E L D I N G , R E - M A C H I N E D , A N D T H E W H E E L W A S A S G O O D A S E V E R , S A V I N G T H E O W N E R S W E E K S O F D E L A Y A N D T H E G R E A T E R P A R T O F T H E C O S T O F A N E W G E A R .

Fi g. 5 .

O N E O F T H E C Y L I N D E R - H E A D S O F A L A R G E D I E S E L E N G I N E .

I t was cracked on the water- jacket; two other heads were cracked in a similar way. The ship to which they belonged was hold up. A temporary repair had already been made by patching—th at oxplains the pre­

sence o f the numerous holes in the jacket.

Fi g. 0 .

T H I S I S T H E R E P A I R E D H E A D A F T E R A L L T H E C R A C K S A N D H O L E S H A D B E E N W E L D E D .

Unlike a “ p a tch ” job, this'was a permanent repair, and in a very short tim e the ship was at sea again. The other heads were re­

paired in the same way.

Fi g. 7 .

T H I S I L L U S T R A T I O N S H O W S H O W E X T E N S I V E I S T H E D A M A G E S U S T A I N E D B Y S O M E E N G I N E P A R T S .T H A T A R E S E N T F O R R E P A I R B Y S C I E N T I F I C W E L D I N G .

I t is a three-cylinder block—part o f tlio main engine o f a powerful tractor. The whole o f the damage to the engine is not shown—the starting engine was also smashed.

I t will ho seen that both sides of the crankcase of the main ongino are broken out.

Fi g. S.

H E R E I S T H E R E P A I R E D C Y L I N D E R - B L O C K A N D C R A N K C A S E , R E A D Y F O R R E T U R N T O T I I E O W N E R S .

These difficult and extensive repairs were completed w ithin a few days, and apart from the very big saving the owners made on the cost o f a now block, tlioy did not lose the use o f the tractor during the m any weeks th ey would have had to w ait for a replacement.

B K E T T : R E C E N T D E V E L O P M E N T S IN W E L D IN G . 1 5 9

machining, errors and making faulty castings dependable, b u t such m atters are likely to be of lesser interest to readers of this Journal.

Few components illustrate th e dependability-of welding so strikingly as do crankshafts. F o rtunately a breakage in this connection is no t common­

place, b u t it does happen from tim e to time. In these days particularly, prom pt replacem ent is impossible. On the other hand, th e reunion of a fractured shaft by scientific welding is no tem porary measure, b u t a per­

m anent and absolutely dependable repair. So much so, in fact, th a t for years it has been th e standard practice of m any large tran sp o rt concerns to have broken shafts welded under guarantee even when replacem ents were im m ediately available.

The care tak en w ith work of this kind m ust be seen to be appreciated fully. Alignm ent is specially im portant, b u t it is th e hydrogen “ field ” process which is largely responsible for the strength secured. I ts m ethod of operating is interesting. F or a long tim e th e welding of any steel com­

ponent which had subsequently to receive a high degree of finish was beset with difficulties owing to the alm ost inevitable pit-m arks in th e weld metal.

These blemishes, whilst not detracting from th e strength of th e work, could hardly be tolerated on a heavily loaded bearing surface. The reason for them was obvious. Surface m arks are due to the affinity of th e oxygen in th e atm osphere for th e m olten steel. F a r less evident was th e cure, although ultim ately th e complete rem edy tu rn ed out to be quite sim ple; for the point of repair was enveloped in hydrogen, and mirror-like surfaces could be obtained a t will. A t first th e supply of hydrogen to cloak th e weld was obtained from high-pressure cylinders, b u t th e m ost favoured m ethod to-day is to use a rod coated w ith a chemical compound th a t gives off th e required volume of gas directly it is heated in th e norm al process of repair. This m ethod of welding is so successful th a t it is freely used to build up worn crank-pins, journals, and other p arts. Sometimes oversize diam eters are specified, so th a t after a further long period of d u ty all th a t is necessary is precision grinding to restore the shaft to its original limits. This m ethod prevents th e need for reducing a worn shaft to below its designed dimensions.

X -ray apparatus of a ty p e specially developed for this work is an im ­ p o rta n t aid tow ards assuring absolutely reliable results. In th e unlikely ev en t of there being a flaw in th e weld, this would be revealed an d elimin­

ated, however small it m ight be.

A nother point th a t arises, sometimes in th e form of unfounded criticism, is the influence of th e weld on h eat-treated steel. This is a m atter th a t is always compensated. I f th e characteristics of th e m etal are likely to be influenced locally, th en restoration is merely a m a tte r of fu rth er treatm ent.

A point a p t to be overlooked in this connection is th e autom atic correction th a t takes place. I t is th a t one layer of electric welding has a valuable normalizing effect on th e previous run. This explains w hy a m ulti-layer weld has a m uch finer grain and is far more reliable th a n heavy single layer weld, which has a coarser crystalline structure. As th e final ru n of welding is not norm alized in the m anner of th e previous runs, an adequate am ount of surplus m etal is left on* th e shaft or other com ponent, thus, when finally machined, th e weld consists of norm alized m etal throughout.

So great is th e variety of illustrations th a t can be chosen th a t it is n ot

easy to select those which will indicate m ost fully the field of opportunity

1 6 0 B R E T T : R E C E N T D E V E L O P M E N T S I N W E L D IN G .

open to the m any branches of th e petroleum industry. In the transp ort of oil from its source to the delivery of derivatives th e valuable applications of welding are legion. B y this means pipe-lines are m ade continuous, and whilst damage and loss from leaking joints are avoided, it is no more difficult to remove a length of piping and insert a new one should this step be desir­

able a t any time.

The construction of tanks and pressure containers of all kinds purely by welding methods is well known, and pum ping p lan t is quickly re-conditioned in th e same manner. In m any examples of corroded p arts a welded com­

ponent has proved superior to a replacem ent of th e original type. The explanation of this is found in th e latitude th a t is possible to th e scientific welding engineer in the choice of weld m etal. The m anufacturer m ay have found it necessary to compromise in the choice of m aterial, on account of conflicting characteristics, whereas, when the tim e comes for re-conditioning, the new m etal to be welded in position is chosen w ith a view to com bating corrosive influence or abrasion more actively, or to bring about an increase in strength. In this way the new work helps all concerned by supplem ent­

ing th e producers’ efforts to reach a more ideal item of equipm ent. In view of this objective, of which there are countless examples in service, there is no exaggeration, in the claim th a t a welded item is often far better th a n a new one. A t the same tim e it is much less costly. Rigid com pari­

sons are, of course, impossible, b u t it is seldom th a t re-conditioning by welding does not show an overwhelming saving on th e price of renewal.

N aturally the figure varies, for the welding engineers m ay be required to carry out several machining operations after th e actual welding has been completed. Often th e comparison of cost, as it relates to a new part, can be disregarded in th e face of a far greater loss of tim e and o utpu t caused by the breakdown of some im portant item of plant. I f it were n o t for welding, m onths of delay m ight ensuet whereas th e resources outlined can be p u t into operation and completed within a rem arkably short time. I t is surprising how even elaborate repairs which involve th e norm al processes of welding followed by machining are commonly completed w ithin tw enty- four hours.

I t is not often realized th a t there would be enormous m aintenance difficulties in regard to foreign-built vessels now sailing with th e Allied F leet were it not for th e benefits of welding. W ith so m any ships cut off from the p aren t yards and factories, aid in refitting and furnishing replacem ents m ust come from other sources. Y et these vessels, m any of them tankers, are m aintained splendidly.

Steam-engine cylinders are welded from tim e to time. Sometimes fractures are so inaccessible th a t portions of sound m etal are cut away tem porarily so th a t a perfect weld can be m a d e ; th en th e section removed is replaced w ithout leaving a visible trace of w hat has been done.

N ot long ago an im portant undertaking handling petroleum urgently

required a large compressor. U nfortunately, in course of delivery th e cast

base, weighing 8 tons, was damaged, a large portion being broken aw ay a t

one corner. To have replaced th e casting, involving elaborate machining

operations, would have taken several m onths. Actually th e original casting

was made ready for installation within five days by skilful and unrem itted

welding.

B R E T T : R E C E N T D E V E L O P M E N T S IN W E L D IN G . 1 6 1

This job was no t exceptional in the dem ands it m ade upon th e welding engineers, b u t the results were m ost im portant, as they united a broken link in a chain of urgent production.

Sometimes an oil-storagetank fails w ithout warning and leakage occurs along a riveted seam. Trouble of this kind arose not long ago when a large container developed this type of fault. The tank, which had been in use only a few years, is used to hold the fuel needed for the central heating system of a large block of flats. Owing to surrounding brick-work it was alm ost impossible to move such a large container, and as the trouble developed in mid-winter, no tim e could be lost. I t was decided to weld the faulty seam, thus making a perm anent seal. There was ju st room to reach the seat of the trouble, b u t the restricted quarters, together with fumes from th e tank, now tem porarily drained, made it impossible for welding operators to work for more th a n ten m inutes consecutively. The welding was therefore carried on in relays, b u t a life-line was attached to each m an, who was under constant observation. The repair was completed in a m atter of hours, and the tan k refilled. I t has not given th e slightest trouble since it was welded.

Bearing in mind present-day road transport problems, there is little or no excuse for th e immobilizing of vehicles, on account of replacem ent p arts being in short supply, i n alm ost every case components which are th o u g h t­

lessly throw n on a scrap-heap impose further strain on supplies of raw m aterial. These item s can not only be restored to a condition th a t is equal to a new p art, b u t th e job can be done quickly—often w ithin tw enty- four hours. Gear wheels from which teeth are broken and lost can be made whole and dependable again, whilst the problem of an ugly smash is straig h t­

ened out in a m anner th a t is incredible to all those who have no experience of the resources of modern welding. Cylinder-blocks broken into m any fragm ents, by reason of road accident or frost fracture, are pieced together invisibly and w ith perfect alignm ent, whilst worn king-pins and other p arts are restored to usefulness and dependability again. This is an aspect of welding which can only be touched briefly upon. I t should be emphasized, however, th a t deeply undercut valve-seats can be built up to their original level again w ith heat-resisting m etal th a t is also capable of w ithstanding valve ham m er far b etter th an cast iron. The increased efficiency resulting from this work is reflected in reduced fuel consumption, am ounting to so much as 20 per cent, in m any instances. Building up worn splines and keyways is com paratively simple to the expert.

Most industries m ust undertake a great deal of experim ental work, and this is particularly tru e of those who handle petroleum and its derivatives.

From tim e to tim e special plan t is required for which there is no general demand, w ith the result th a t th e cost is liigh. In regard to this m atter welding can help a great deal, since it avoids thp use of castings, and, in addition, weight is reduced and, still more im portant, th e need for pattern - making is eliminated. I t is not suggested th a t th e foundry is likely to be made unnecessary by welding. This is not th e case where repetition work is concerned, b u t when one or even two or th ree duplicate pieces of m achinery are required welding is less costly.

The same m ethods can be applied even to gear-cutting. This can be

done with such accuracy in flam e-cutting machines th a t gears, pinions, or

sprockets formed in this way can be p u t straig h t into service an d w ithout

1 6 2 B R E T T : R E C E N T D E V E L O P M E N T S I N W E L D IN G .

th e necessity of any additional finishing. One ra th e r elaborate machine recently p u t into commission incorporates ra th e r more th a n fifty gear-wheels an d sprockets, each one of which has been flame cut from a steel blank.

I n th e examples quoted there is considerable m aterial for th ou ght and practical application. There is, however, a pitfall. The craftsm an always creates an impression of simplicity in his work and th e welding repair worker is no exception to this rule. Ambitious work should therefore never be a ttem p ted w ithout th e essential backing of experience and th e advice of th e specialist.

Welding goes a very long way tow ards solving the difficulties and easing th e responsibilities of m aintenance engineers. A t th e same tim e it can offer alternatives of the greatest value in new construction, and provide a solution to complex m etallurgical problems which hinder new developments.

I

Ju n e 19 4 3 1 9 9 a

ABSTRACTS.

Geology and D evelop m en t

P A G E . . 1 9 9a

G eophysics ... . . . 2 1 1a

Drilling . . 2 1 3a

Production ... . . 2 1 7 a

Transport and Storage . . 2 2 6 a

Gas ... ... ... . . 2 2 7 a

H ydrogenation .... . 2 2 7 a

P olym erization and A lk ylation .. . 2 2 7 a

Synthetic Products . . 2 2 8 a

Refining and R efinery P la n t . . 2 2 8 a

Fire Prevention . . 2 2 9 a

P A G E

P h ysics and C hem istry o f H ydro- c a r b o n s ... ... ... ... 230 a

A nalysis and T estin g 2 3 2 a

M otor F u e l s ... ... ... ... 2 3 2 a

L ubricants and L ubrication ... 233 a A sp h alt and B itu m en ... ... 2 3 5 a

Special P rod u cts ... ... ... 2 3 5 a Coal an d Shale ... ... ... 2 3 6 a E conom ics and S ta tistics ... ... 237 a

P u b lication s R eceiv ed ... ... 2 4 0 a

A U T H O R I N D E X .

T he num bers refer to th e A b stract Num ber.

The o rig in a l pap ers referred to in th e a b stra cts m ark ed w ith a n asterisk m a y be borrow ed by M em bers from th e In stitu te Library.

Amero, R. C., 576 Anglo-Iranian Oil Co.,

589

Baker, W. L., 511 Bakus, H., 601 Baldwin, R. R., 585 Ballard, N., 506 Bamford, C. H., 585 Bolton, L., 598 Briggs, F., 552 Bunch, E. S., 525 Cabeen, W. R., 561 Canesco, 0., 581 Capell, R. G., 576 Cobb, J. W., 598 Coumbe, A. T., 603 Cozzens, F. R., 559 Cragg, J. C., 586 Culüngworth, J . E., 598 Dawson, L. R., 542 Decker, C. E., 505 Devon, J., 599 Dudley, R. L., 512

Ï, G., 596 Evans, E. A., 586 Faith, W. L., 584 Fancher, G., 549 Faust, P. H ., 582 Freeman, L. I., 541 Friedmann, W., 581 Ghosh, B. P., 598 de Golyer, E., 526 Hake, B. F., 535 Hines, J . T., 593 Hobson, G. D., 508 Hopkins, G. R., 601, 602 Hunter, A. L., 561 Hunter, G. M., 536 International Catalytic

Oil Processes Corp., 589

Jenny, W. P., 538 Kalmanovsky, E., 587

Kemler, E. N .,551,565,568 Kirkpatrick, G. M., 569 Korafeld, J . A., 558 Lacey, W. N., 550 Lahee, F. H., 537 Layng, E. T., 589 Logan, L. J., 513 Lott, F. S., 602, 603 Maxwell, T. A., 591 Milam, H. S., 564 Morian, S. C., 575 McCamey, J . D., 580 Nickell, F. A., 507 Olds, R. H ., 550 Ortynsky, R. L., 593 Parlette, W. H., 577 Payne, T. G., 504 P ott, A., 572 Price, P. H., 509 Redfield, A. H ., 600

Reed, P., 562 Rolhner, T. G., 592 Ryder, H. M., 556 Sage, B. H., 550 Shaw, S. F., 566 Simons, H. F., 548, 560,

563

Smith, R. O., 540 Standard Oil Develop­

m ent Co., 589, 597 Sterrett, E., 512, 543,

544, 547, 578 Stockman, L. P., 527 Sudholz, L. H., 592 Wassermann, A., 573 Weiss, J ., 583 Weizmann, C., 579 Whiting, R. L., 549 Wildams, N., 546, 557 Winterbum, R., 553, 554 Woodward, H . P., 509 Zabel, H. W., 584 Zinszer, R. H., 539

Geology and Development.

504.* Stratigraphical A n a ly s is a n d E n v ir o n m e n ta l R e co n stru ctio n . T. G. P ayn e.

Bull. Amer. A ss. Petrol. Oeol., N ovem b er 1 9 4 2 , 26 ( 1 1 ) , 1 6 9 7 —1 7 7 0 .— I n stratigraphy the naming of form ations, correlation charts, and faunal lists are n o t enough, bu t must be su p plem ented b y palaeoecological stu d y of sed im en tary ty p e s and fossils.

A scheme for rapid a n alysis of d a ta bearing on m ode and en viron m en t of origin is outlined w ith special reference to th e crinoidal p h ase of th e Grand Tow er lim estone (Middle D evonian) of Ozora, S te. G en evieve C ounty, Missouri.

Field description, th in sectio n s, p olish ed section s, binocular stu d y of fragm ents, chemical te sts, siev e and p ip e tte m eth od s, all con trib u te to petrographical analysis.

Genetic classification th e n d istin gu ish es b etw een au toch th on ou s and allochthonous components. T h e form er m ay be eith er sy n g en etic or ep igen etic, according as th ey are formed contem poraneously w ith or after th e rest of th e rock. A ccount m u st also be taken of th e m anner in w h ich each com p on en t h as b een form ed. A ttrib u tes of the com ponent fragm ents to be n o te d are : c o m p o s itio n ; internal structure ; size ; percentage of to ta l volum e ; sh ap e and degree of rounding ; surface textu re— w hether rough, sm ooth, frosted, etch ed , striated , p it t e d ; rela tiv e position of c o m p o n e n ts;

random, parallel, or im bricate orien tation of fragm ents.

Mass properties to be considered are colour, chem ical com position in b u lk, acid R

2 0 0 a A B S T Rc u jx o.

solu b ility, in terstitia l cem en tation , m ass tex tu re, packing, p orosity, perm eability.

P la sticity , hygroscop icity, fu sib ility, and specific g ra v ity are of less stratigraphical im portance. M ass tex tu res are eith er fragm ental or crystallin e. I n th e la tter case, th e effects of chem ical precip itation , secondary recrystallization, and secondary crystalline grow th m u st be distinguished. F or in stan ce, orthoquartzite, due to grow th of secondary quartz, is to b e discrim inated from m etam orphic, recrystallized p araquartzite.

Syn gen etic sedim entary structures to be looked for include cross-bedding an d o th er ty p e s of stratification, bioherm s and biostrom es, m arkings b y rain-drops, ice-crystals, w ave-ripples, sw ash es, and anim al tracks, ten sion al m ud-cracks, com pressional ice- rucking, slum ps. E p igen etic structures are solutional, like s t y lo lit e s ; com pressional, like folds, fau lts, cone-in-cone or tensional, like fa u lts and jo in ts.

Palaeontological analysis, for w h ich Sim pson and R o e ’s Q uantitative Zoology (1939) m ay serve as an in troduction, begins b y d istin gu ish in g life assem blages an d d eath assem blages. I n different sed im en ts, p revalen t genera and ty p e s of gross extern al m orphology (c/. L am ont, A n n . M ag. N a t. H ist., Ser. 10, v o l. x iv , 1934) should be recorded. T he size of th e organism s is also u sefu l as an in d ex to factors of sa lin ity , tem perature, tu rb id ity , d issolved gases, e.g. abundance or p o v erty of oxygen , in th e environm ent. D w arf faunas m ay be due to stu n tin g or to w eed in g o u t of large sp ecies b y selection (cf. Shim er, Am er. N a t., vol. xlii, 1908). P resence of broken m aterial, space distribution, an d estim a tes of th e num bers of different sp ecies in faunulos should also b e studied, w ith th e caveat th a t short-lived anim als, or th o se w ith an annual m ou lt, m ay produce a larger num ber of p reservable hard p arts, th a n th eir contem poraries.

P a y n e also an alyses th e environm ental factors in th e provision, transport, and d eposition of sedim entary m aterials. H e propounds rules such as th a t “ th e sea w ill te n d to produce, under given environm ental conditions, sed im en ts of sim ilar tex tu ra l character regardless of differences as to source and ty p e of m aterial b ein g su p p lied ,”

and th a t “ under g iv en environm ental conditions th e sea w ill ten d to produce organic assem blages of similar nature regardless of differences as to species and genera a v a il­

a b le.” I t is desirable to know th e character, slope, and degree of com paction of th e substratum , also th e direction of m ovem en t of th e sedim ent and of th e m edium . B o tto m currents can be estim a ted from th e range o f size of particles b y m ean s of H ju lstrom ’s size-velocity curves for erosion and deposition (Recent M arine Sediments, A .A .P .G ., 1939, p. 13). C onjectures should be a tte m p te d concerning d ep th of w ater, lig h t in ten sity , and geographical position. L im iting d a ta can be obtained from th e “ ecologic valen ce ” of different fossils. T hus, th e valen ce of a sp ecies of coral m igh t be : salin ity, rather high ; m ean tem perature, 60-85° F . ; current v elo city under 100 c m ./se c o n d ; dissolved o x y g en over 4-0 m l ./ L ; lig h t in te n sity , ab ove som e m easurable value.

T he chem ical changes undergone after deposition, in th e zones of w eath erin g, diagenesis, and anam orphism , fall also to be considered, togeth er w ith th e h istory

of th e sedim ent during diastrophic m ovem en ts. A . L.

505.* T w o M ore O rdovician W ell-co re G raptolites, Crane C ounty, T exas. C. E . D ecker.

B ull. Am er. A ss. Petrol. Qeol., N ovem ber 1942, 26 (11), 1771-1775. F rom M cK night N o. 4, Crane County, a t 6458 ft. an d 6479 ft. resp ectively Callograptus tenuissim us D ecker sp. nov. and Didymograptus cf. bartrumi B enson and K eble are described.

T he former w ould probably fall in th e sub-zone (a) of th e D . extensus zone, as E lies indicates th a t dendroid graptolites are m o st abundant in th a t sub-zone w h ich is about th e m iddle of th e Skiddaw slates. T he nearest counterpart, is C. tenuis from th e U pper A renig. D . cf. bartrumi com pares w ith D . bartrumi w hich in th e Low er Ordovician of N ew Zealand is associated w ith Tetragraptus Jruticosus. I t also

approaches D . nitidus. ^ t

506.* Stratigraphy o f N orth D a k o ta . N . B allard. B ull. Am er. A ss. Petrol. Qeol., N ovem ber 1942, 2 6 (11), 1776. I n th e Carter te s t and Prairie Oil and G as C om pany’s A rm strong N o. 1 ( = Steele o f D r. K line), beds w ith red and green sa n d y sh a les and a chert conglom erate, ascribed b y K line (Bull. A .A .P .Q ., March 1942) to th e D evon ian , are probably p a rtly Jurassic and Triassic. Perm ian or P en n sy lv a n ia n m a y also be

represented. A L

A B S T R A C T S . 2 0 1 A

5 07.* D e v elo p m en t and U s e o f E n g in ee r in g G eology. (W ith a Forew ord b y P aul W eaver and others). F . A . N ickell. B ull. Am er. A ss. Petrol. Qeol., D ecem ber 1942, 26 (12), 1795—1826.— T h e hade and direction o f fa u lt p la n es o ften depend on th e natu re of th e rock form ation. F a u lte d areas u su a lly con tain a v a s t num ber of in ­ d ivid u al fa u lts. A structural contour m ap w ith unbroken lines, is, therefore, a t b est an idealization.

S tu d ies of th e site s o f th e B oulder, Grand C oulee, S h asta, F rian t, and M arshall Ford dam s in N o rth A m erica are u sed to illu strate such problem s. T h e factors tak en in to accou n t are :

(1) L ocal geology, including seism icity.

(2) Character o f form ations a t site.

(3) T h ick n ess of cover or overburden.

(4) P h y sic a l s ta b ility of rocks.

(5) P ossible seepage and d etailed featu res requiring special treatm en t.

(6) A vailable con stru ction m aterials.

Trial h oles, seism ographic m eth od s— seism ic v e lo c ity is related to ela sticity — m igration of tracing fluid b etw een h oles, d eterm ination of com pression and shear stren gth s, and petrographical and siev e a n alyses are all of significance.

Boulder D am , Nevada and Arizona.— T he dam of arch-gravity p lan is b u ilt on th e D am breccia, a T ertiary tu ff w ith m onzonitic p orp h yry fragm ents, th e m inim um u ltim a te stren g th of w h ich is 9260 lb ./sq . in ., and m axim u m 22,400 lb ./sq . in . U p p er parts of th e a b u tm en ts rest on la tite flow -breccia w h ich can carry reduced stresses.

T h e site is jo in ted and fa u lted , w ith a ch annel 80 ft. deep, due to la te Glacial reju ven a­

tion, alon g th e centre of th e floor of th e gorge. F illin g o f th e reservoir resu lted in sligh t iso sta tic settlem en t, an d certain fa u lts w ere revitalized. M icroseism s of in ­ te n s ity IV on th e M ercalli scale (1931) h a v e b een recorded on th e fa u lted m argins of depressed blocks som e m iles from th e dam.

Grand Coulee D am , Washington.— T his straigh t-gravity dam , containing 10 m illion cu. yd s. of concrete, w hich d evelop s pow er partly u sed to pum p w ater in to a high-level sy stem of irrigation canals, is on Jurassic coarse-grained granite w ith intrusions of fine-grained granite an d o f T ertiary b asalt. There are crush-planes. O verburden is of G lacial clays and sand, an d of stratified alluvium . T here w a s m uch slum ping w hen, after th e glaciation , th e C olum bia river e x ca v a ted th e valley-fill, th e rem aining part of w h ich h ad to b e stab ilized b y drainage, p artial unloading, and dum ping of rock along th e to e. F a u lts are w id e ly spaced and probably of T ertiary d ate. T h ey influence w eath erin g an d erosion, and h ad to b e clean ed o u t below th e le v e l o f th e surrounding floor, filled w ith concrete, and grouted.

M arshall Ford D am , Texas.— T his straigh t-gravity structure controls flooding on th e Colorado river and serves hydro-electric generation. I t rests on nearly h orizontally bedded Com anche (Low er Cretaceous) sedim ents, including sandy lim estone, lim estones, and shales. S h aly layers slacked on exposure and h ad to be trim m ed im m ed ia tely before concrete w as p u t in position . Som e stra ta ten d ed to slide under th e d o w n ­ stream th r u st o f th e dam . T o p rev en t squeezing of so ft m aterial an d to ob viate seepage, pairs of tu n n els w ere constru cted th rou gh th e d ou b tfu l b ed s in each a b u t­

m ent, and th ese drifts w ere th e n filled w ith concrete and grouted. G routing w as also used to fill cracks an d ca v ities under th e foundations. A rtesian flow in th e H en sel

“ sands ” a t a d e p th o f 200 ft. in d icates relative im perviousness of th e beds flooring th e reservoir.

Shasta D am , California.— A curved-gravity dam , to be com p leted in 1944, is in ­ ten d ed to control floods on th e Sacram ento river an d regu late th e sa lin ity in San Francisco b a y , and to serve for pow er generation. T he fou n d ation s are on a g g lo ­ m erate, conglom erate, an d la v a of th e C opley form ation (pre-M iddle D evon ian ).

In tru sion s of D a c ite p orp h yry and quartz-diorite are of Jurassic age. L ocal fau lts are in a ctiv e, b u t slig h t earthquakes from M ount L assen h a v e b een calcu lated for.

W eathering is far ad van ced in th e D ev o n ia n m eta-an d esite, and over 3 m illion cu.

y d s. of p o ro u s m aterial h ad to b e strip p ed aw ay. F a u lts and schistose zones w ith gouge had also to b e trea ted to av o id differential settlem en t and seepage. D rainage and in sp ection tu n n els ha v e b een driven for exam in ation of crush zones and concreting w h en th e reservoir is filled. S tren g th of th e m eta-an d esite varies from 535 to 35,400 lb ./sq . in.

2 0 2 a A B ST R A U X S.

F riant D am , California.— W here th e San Joaquin river leaves th e w estern foothills of th e Sierra N evada, th is straigh t-gravity dam im pounds irrigation w ater. The foundation is m ainly quartz-biotite sch ist w ith m any Jurassic intrusions of p egm atite and dioritic dykes. In general, planes of sch isto sity dip dow nstream and strike at a sm all angle to th e axis of th e dam. W eathering becom es more in ten se as one ascends th e v a lley slopes. F a u lts and seam s w ith unsound rock w ere in v estig a ted b y shafts, tunnelled, concreted, and grouted. S tren gth of th e sch ist varies b etw een 1460 and 20,680 lb ./sq . in. U nw eath ered specim ens break along planes according to th e orientation of th e strain ellipsoid, regardless of th e direction of sch isto sity . A . L.

5 08.* C alculatin g th e True T h ic k n ess o f a F old ed B ed . G. D . H obson. B ull. Am er.

A ss. Petrol. Geol., D ecem ber 1942, 26 (12), 1827.— H obson g iv es an accou n t of th e errors w hich m ay arise in calculating true th ick n ess of a bed from vertical th ick n ess m ultiplied b y th e cosine of th e angle of dip. T he calcu lation becom es less accurate th e greater th e dip and th e greater th e ratio of th e tru e th ick n ess of th e b ed to it s radius of

curvature. A . L.

509.* G eology and W ar. P . H . Price and H . P . W oodw ard. B ull. Am er. A ss.

Petrol. Geol., D ecem ber 1942, 26 (12), 1832.— T he strategic ap p lication s o f geology are num erous. I t can be used in th e interp retation of terrain for defence and a ttack , since th e geologist understands earth-features b e tte r th a n an yon e else, and know s how to forecast th em in unfam iliar areas from geological p u blications, photographs, and m aps. M ost overseas regions h a v e been exam ined b y geologists now resident in th e U .S . T he g eologist can also p oin t o u t te stin g grounds w here dup licate con ­ ditions e x is t for an ticip ated cam paigns. H e can advise on off-shore, beach, and riverine topography, on soil and lik ely v eg eta tio n , on sites for roads and road-m ending m aterials. H e can help in appraising th e value or vu lnerability of fortifications, cam ps, batteries, aerodrom es, and on th e accessib ility of w ater and m ineral resources.

Geological advice sh ou ld be ob tain ed in th e conversion of ground to m ilitary use for trenches, ta n k traps, tu n n els. Quarrying and th e p reven tion of slum ps and debris-slides are also in th e geological am bit, and th e q uestion of w hich sw am ps and quicksands can b e reclaim ed and w hich n o t. E ffect of bom bardm ent on different rock ty p es, sa fety of underground w orkings as refuges, reten tio n of gas in different rocks and soils, restoration of sh elled territory and of oil-w ells, are all m a tters for geological investigation.

G eologists can prospect for brick clays, m arl ballast, sand, aggregate for concrete, do m ap-m aking and surveying, solve w ater-supply problem s and th o se of sa n itation and interm ent.

Camouflage and its d etection are a new sphere for th e geom orphologist w ho know s so w ell th e natural appearance of a landscape.

G eophysicists can be em ployed to d etect m ines, pip es, and m etallic underground installations.

T he authors recom m end th a t a geological organization should b e se t up for th e application of geological know ledge “ in th e p ost-w ar settlem en t, and in possible future m ilitary and n a v a l undertakings.” There should be sustained stu d y o f geology in relation to fortification. H andbooks of M ilitary G eology should b e prepared. A more in ten sive search for new sources o f raw m aterials b o th a t hom e and abroad is necessary now , and w ill be equally necessary in th e future. A . L.

510.* W a r T rim s W orld Crude P ro d u ctio n T en P ercen t. A non. Oil W kly, 1.2.43, 1 0 8 (9), 44.— T h e 1942 w orld oil production w a s a b ou t 10% below th e 2,226,836,000 brl. produced in 1941. T his decline reflects in particular th e d eclin es in U .S .A ., R u ssia, and V enezuela, as w ell a s th ose in countries w ith sm aller o u tp u ts. T h e 1942 estim a tes are rather u n certain ow ing to th e ab sence of official sta tistic s on m an y countries. I t is p ossible th a t b o th R u ssia and V enezuela h a v e experienced sharp production declines. Subm arine w arfare g reatly curtailed V enezuelan oil exp orts early in 1942, b u t th e exp orts probably rose later in th e year.

T he M aikop field in R u ssia w a s p u t o u t of action in 1942, and norm al oil m ovem en ts from Grozny and th e B aku d istrict were disrupted b y th e Germ an invasion. In th e