Investigations upon Sugar Boiling, Clarification, and Quality of Raw Sugar By W. H. BASS. With the advance of science and technical ideas sugar mills have of late years improved their methods and efficiency in almost every direction, but in most mills no great advance has been made in t h e study and improvement of sugar boiling as this art was found not t o lend itself easily t o such met,hods. Recognising the need for study in order t o bring about the improvement of refining quality of raw sugars, the Mackay 3Tanufacturers' Association inaugurated these researches and this paper is submitted to the Conference through their courtesy and generositv. At the outset a visit was paid t o the U.S.R. Co.'s Research Laboratories and assistance rendered by this Company was found invaluable for this work. The investigations were carried out a t each of the seven mills in the Mackay district, a t which i t was found that the co-operation by Xanagements and Staff was largely responsible for the rapidity of the progress made.
BOILING. Graining. This most important operation, which from the outset reflects largely upon the quality of the final grain, was first studied and i t was found that uneven grain was due t o a great variety of causes. The varying causes were made quite clear t o each mill by microscopic examination, ant1 i t is pleasing t o note that they now each have a microscope upon the pan stage in constant use. It was obvious in most cases that the unevenness of the grain was due t o : (1) too long a period by the waiting method for the required number of grain t o appear : this wolxld often take as long as 20 minutes with grain coming in all the time. with the result that the crystals which appeared a t first were much larger than those towards the end : (2) a far worse fault was new grain appearing throughout the skip : thus, although a n initial uneven crop remained the same in number throughout the boil, it would have a tendency t o even up considerably in size a t the latter stages, ~vllereasgrain coming in all along would make far greater unevenness. The maintenance of ~ t e a d ysteam and vacuum conditions on the pan stage is a prime necessity for good grain, and a recording vacuum and/or temperature gauge for the graining park was found t o be a most vahxable guide.
Conglomerate Grain. Dnring the early part of these investigations it was found that; cry;ital complexes, conglomerate. or curly grain were very easily forn~ed,t o the extent of as mueh as 80 per cent,. by weight, and this type of grain was very difficult t o avoid under the then existing con-
.
60
April,-1934 -
FIETR ANNUAL CONFERENCE .
PROCEEDINGS
ditions (Plate 111, Figs. 26, 26, 27). This type of crystal is easily distinguished from the aggregates of crystals, being generally formed on twin cryatals with crystals growing from near the centre i n all planes, and consequently star shaped when viewed on edge. Further, as opposed t o aggregates, the crystals of true conglomerate grain cannot be separated without breaking the crystals.
It is presumed that the refiners' objection t o conglomerate grain is based on the syrup or molasses film which is enclosed in the very centre of t h e crystal, and which cannot be washed away during t h e affining process because the crystal remains complex after washing. This condition is borne out by analysis. During grist analyses of raw sugars i t is generally found t h a t the conglomerate grain is more concentrated in the larger crystals, as is seen in the following tabulated result of a n analysis : "
Grain Size (Determined by Sieving)
--
> 14 meshes
14 > 20 meshes
Mixed Sugar
Total
(>onglomerate
---
p----------.
Coarae
B " Sugar
1 i
%
1
1.9
1
16.9
45
1
30
1
Truo
1 /
I
%
1
4.5
1
24.9
60 35
20 > 28 meshes
28 > 35 meshes Pine
=
< 35 meshes
Satisfactory Grain (20 and 28 mesh fiWtiom) Total Conglomerate
1 I
Apparent
1
6196
12.5
1 /
11% 17.40,;
1
Apparent
1 I
True
64%
l
50%
10.3
18.4%
l
The figures shown under the sub-heading " conglomerate " represent the percentage of each fraction which is present as conglomerate grain ; by deducting this proportion from the " apparent " satisfactory grain a n arbitrary figure, " true satisfactory grain," is obtained. An approximate figure of- total per cent. conglomerate grain is obtained by adding the weights of conglomerate grain on each sieve fraction. The diameters of openings in the 14, 20, 28, and 35 mesh sieves were respectively 1.7, .83, .5, and .42 mm., while the a,verage size .of crystals retained by these sieves was respectively 1.41, L .O, .75, a n d . 5 mm. Hand picking of conglomerate grain from separate sugar samples o n the same sieve gave the following results : (1) Well washed good filterability sugar. Ash conglomerates (Plate 111, Figs. 232-3) 10 per cent. higher than non-conglomerate. (2) Medium filterability sugar. l . l 1 mm. fraction. Ash conglomerate 0.27 per cent., non-conglomerates 0.21 per cent. (Plate 111, Figs. 234-5.)
A
9
QUEENSLAND SOCIETY SUGAR CANE TPOBNOLOCISTS -
61
U
(3) Medium filterability sugar. Ash conglomerates 0 . 2 6 per cent. non-conglomerates 0 . 2 1 per cent.
It is assumed t h a t the impurities will be present in proportion t o t h e ash, and records have shown improvements in filtrability with diminution of the proportion of conglomerate grain. Conglomerate grain appears t o be formed mostly in t h e early stages of a boil. The higher the purity, the greatmr the likelihood of crystal complpxes and, particularly when bringing together, a n y new small false grain with a tendency t o grow upon twins eventually develops into conglomerate. Conglomerate grain i n bags occupies more volume than ordinary grain, and i t has been noticed t h a t t h e weight per sack increases when this is eliminated. Conglomerate grain appears t o separate t o the outer edge when fugalling (Plate TII., Pigs. 293-5) and although the resulting sugar is always darker, i t has not been established whether i t hinders fugalling.
Purity of Graining Charge. The following advantag- were found when graining upon purities of 80-87 obtained by a blend of liquor and syrup.
(1) The crop comes in more suddenly and more evenly : within limits the lower the purit,y the greater the number of grain in t h e crop. (2) Conglomerate grain is minimised. There appears t o be a seasonal and district variation in this maximum purity required t o keep out conglomerate grain and often a slight drop in purity makes a very appreciable difference. 8ee Plate I., Figs. 374-377, 380, 384-387.
(3) When cutting liquor seed or seed of high purity, t'he tendency t o form fine and conglomerate grain can be minimised by adding a little lower grade syrup before cutting. The only apparent disadvantage of using blonds for graining is the likelihood of increasing viscosity and the greater ease of decomposition of syrup. By the adoption of blend graining the amount of conglomerate grain has been reduced from about 80 per cent. t o 10 per cent. in the Mackay district.
Fornation of Crops. It was found that the best grain was made when the biggest "shock" was introduced t o the charge a t about graining point. Where pans were not capable of being sufficiently cleaned o u t between skips, there was generally sufficient " soldier " grain t o make the addition of icing or castor sugar superfluous, as the sugar added is not considered t o be the nucleus for grain formation. It is interesting t o note that the presence of any sugar crystals whatsoever prior t o graining point gives a very good indication of the correct time t o shock, since as shown by microscopic observation, the crystals which were eroded sharpen up fairly suddenly a t about graining point. NO effect was observed by varying the quality or quantity of " shocking " sugar, although this sphere has not been closely investigated.
6%
April, 1934
P R O C E E D I N G S P I F T R ANNUAL CCONPEREYCE
_ _ X _ _ _ _ - _ _ _
__
-.__ll...-_____
"-
The best method st~xdieclt o date is a combination of shocking methods, as follows, in which a recording thermometer (see Chart 1) is used : Just before graining point ( T l ) a small qnaatity of icing or castor sugar is introduced into the body of the charge ; a t the same time air is bubbled through, so that the vacuum quickly drops about 2" and the temperature on the recorder rises suddenly .to T2, after which it begins t o fall suddenly. At this moment the steam is shut off and
t h e head water increased, so t h a t a sudden cooling takes place. J u s t about the time when the original T1 is reached the crop appears, and, when in sufficient number, is checked a t temperature T3. The resultant chart will appear similar t o that illustrated in Chart 1, and if the shock was induced a t the correct t,ime there will be only about two minntes between T2 and T3. I n fact this time can be regarded as a measure of the speed with which the crop appears and its resultant evenness, and is, t o the writer's knowledge,the only methot1 of obtaining a record of the sugar boilers'accwracy in shock seeding at the correct time.
Establishment of Grain. The time between checking and establishment, or "bringing together," is the most critical period, as i t is a t this time t h a t the supersaturation is more likely t o show large variations, on account of the greater rate of crystal growth and temperature variation. Once a pan is put upon continuous liquor or s37nlp feed t,here is but little variation in temperature, or only a very gradual variation as the massecuite heavies up. After checking, however, the temperature may fall as much as 4O0P., which represents a large change i n supersaturation and if this takes place too sudtlenly, false, uneven and conglomerate grain is bound t o appear. Independently of the temperature, if during "bringing together," the supematuration is not steadily maintained by occasional drinks, false, uneven, and conglomerate grain may also occur. So far the best improvement has been obtained by using a recording thermometer in the graining pan and maintaining a gradual return t o normal boiling temperat'ure after checking, with the addition of short drinks of water every fernminutes ; the latter is presumed t o be a n advantage over liquor
"
drinks as it results in the time of bringing together being shortened. d good example of this is seen in C'harts 2 and 3 which represent charts obtained fur the same graining. After a little practice, t h e sugar boiler has no difficulty in maintaining the required vscnxuas and can produce charts of quite uniform appearance. It Inay be again emphasised here that when using blends for graining, the eveness or otherwise of the initial crop is by no means as irnportallt as t h e elimination of further grain, particularly during the period of bringing together or establishment of t h e grain.
I t was observed that ternpcrature ha,.: a great effect upon t h e characterisbicn of the grain and a t the higher temperatures a thicker crystal was formed. 'Phis may be a n explanation for the term " hard " ancl " soft " grain--the thicker crystals feeling harder t o t h e touch. This may also serve t o substantiate the claim that "soft " grain seed exhaust fiyrup Inore as, with crystals of the same flat size, the thinner ones might eventnallp grow t o the same size and thickness ns the thicker crystals and thus exhaust more sugar from the syrup.
_
64
PROCJEEDIN(:Y
__
-
FIPTH A N N ~ T A LCONFERENCE
- --
- l ~ ~ r i1934 l. ---A-
Fig. l.-Dot diagram representing the relation between filterability and the quantit,y of lime used for the clarification of raw sugar liquors. Data collected from experiments at three mills.
Glaritication, Molasses Inclusion, and Pilterabilily. I t has been found that the filterability of the raw sugar depend& largely upon the clarification and when this process was improved t h e filterability rose a8 is seen from the following figures:
Eilterability ( X i 1 1 A ) raw sugar i?z gallons y e y square foot y e s hour. Keek ending 23/.9/33 Poor clarification A
B
5.7 2 .0 0.6
Mixed
3.6
AR
D7eek ending 28/11/33 Good clari,fication A AB B
8.6 5.2 4.6
Mixcd
6.6
I n the above instance, fall in filterability with grade of sugar is most marked ; in some cases the p H of the different grades of sugar falls witjh the filterability as is brought out by the following results : Class of Sugar
Mill R
I{ i
Mill C
c
pH.
Filterability
6.8
9.6
6.8
7.5
6.6
6.7
6.1
very low
7.0
7.5
6.9
5.7
6.4
very low-
The greater concentration of colloidal matter in the lower grades of syrup is probably the determining factor in hhis relation. and not the pH, which, however, does show the decomposition produced by continued boiling. x-as greatly improved b p the introI n some mills ~larificat~ion duction of more accurate p H control, arltl a little assistance was also tons per week) in the afforded bv adding j)hosphoric acid (up t o forin of superphoppliate. The introduction of baffles, a shalt distance from the overflow of the subsiders, was found of great assistance in retaining floating scum and giving a good clean juice. At the end of the s e a o n almost every mill had satisfactory filterability, which was a considerable advance over previous years' results.
In the dot diagrams set out in Fig. 1 are shown the relation between t h e pH of the mills' mixed weekly sample of shipment sugar, and filterability. The abscissaere present amount of lime required t o bring sugar t o p H 8.0 and is, therefore, proportional t o the p H of t h e sugar. The ordinates represent filterability in galls. per sq. foot per hour. These charts demonstrate fairly clearly the improved clarification and filterability a t higher p H values. Molasses Inclusion. Molasses inclusion (impurities embedded in the
grain), is
aggravated by poor clarification, and the former was found to have a great effect upon filterability, particularly t,hat of the affined sugar, as the impurities were not all washed away during affination. The
66 --
PIFTH ANNUAIA CONFEREXCE ----PROUEEDINUS -- -- -- - ---- -
April, 1934
cause, nature, shape, position and extent of molasses inclusion was found t o vary to a very considerable degree. I t may occur a t t h e very beginning or end of a boil, and sometimes even in a jelly " which is crystallising out, and the different types are clearly seer1 i n the photographs. "
&Iicroscopic observation clearly demonstrates that the larger faces of a crystal are very easily etched or eroded in nn-saturated solutions, which might conceivably form a pocket for syrup. One type of this, termed "late inclusion," takes place towards the end of the boil when the circulation in the pan is very poor, and with "stagnant " ~xn-saturatedsolution on the surface the massecuite might easily erode the crj-stals, which, when falling into supersaturated solution would sharpen up snddcnly without the enclosed syrup being able t o escape. dnalj~sisof the impurities, both inside and olitsicle the crystal, gave some interesting figures :( l ) A raw B. sugar with late and central inch~sion( l.00 mm. fraction) was taken : i t had two distinct zones of inclnsioncentral and late in almost every crystal. Results of analysis were-Filterability 1 . 3 : Ash 0 . 3 2 per cent. (See photo Plate VI., Fig. 236.) ( 2 ) The above sugar was washed tlloroughy with saturated pure sugar solution, and finally with pure sugar saturated alcohol, and dried. With this washing all external syrup is presumably removed, hut the sugar was still brown in colour, and the ash was found t o be still as high as 0.11 per cent. This is, therefore, clue t o the syrup included in the crvstals, and. in this instance. as much as one-third of the total impurities is embedded in the grain. (See Plate T T . , Fig. 237.)
(3) The washed sugar from ( 2 ) was washed u~itlipure unsaturated sugar solution until the crystals were partially ditisolved and t8he late inclusion removed, when it was noticed t h a t the wash liquor became coloured owing t o the removal of the latter. The crystals were finally washed with sugar saturated alcohol and dried Analysis showed that the ash was reduced t o 0.09 per cent., so t h a t here the late inclusioil is responsible for only a small proportion of the total inclur' , 10x1. (Hee Plate VT., Fig. 238.) (4) The sugar from (3) was washed with unsaturated sugar solution
until more of the crystal tlissolved and the zone of the central inclusion was reached. The crystals were &hen washed with saturated sugar alcohol and dried and gaye a n analysis of 0.13 per cent. of ash. This figure leaves no doubt as t o the extent t o which impurities may X)e ernl~edcted in the grain of raw sugar. (See Ylate TT., Fig. 239.) Another R sugar with marlceii inclnsion, and with a filterability of only 0.6 (see Plate Vl., Fig. 309) gave the following anal>-sis :--
April, 1934. QL~EEXSI~ANL) SOCIETY -
SUCAR CAAE TECHKOLOGISTS
p p -
--
A"
67 P
I N external tnolnsses filrvtWater Sucrose Impurities I n cr!jstctls-Water Sucrose Impurities
.. .. .. .. .. ..
0.21 0.97 1.22 0.13 96.7 0.71
99.98 Internal Imyur. .37 Ratio : ---0 ' 7 1 - ---Total Impur. l . P2 0.71 -1.0 At one mill with filterabilities for raw sugar for 4 weeks as follows:--1.8, 1.$, 2.1, 1.'7 (see Plate III., Fig. 35%),thefilterabilityfor affined sugar was only 4.76 for a bulk sample of the above period. At a later date with the shifting of the inclusion towards the edge of the crystal, although the filterability of the raw sugar a t this mill wras only 1.9, t h a t of the saturated sugar washed raw, was 9. l, so that the position of the inclusion plays an important part in refining, the nearest t o the centre being the worst. Figs. 513-517, Plate W., show. the characteristics of late inclusion. As i t appears t h a t central inclusion would be formed early in the growth of the crystal, excessive amounts of water in the early stages of t h e boil should be avoided, and in some instances advantageous results were obtained by cutting down the size of the pan water supply t o 4 t o 4" pipe. The depth and position of inclusion within the crystal varies considerably as will be seen in the photograph. The position i s located by focussing upon any particular surface or edge of the inclusion, in relation t o a surface or edge of the crystal, and measuring the difference. When any measurements are taken within tho crystal. allowance is made for the refractive index of sugar. The writer has been unable t o discover in the literature any descriptions or photographs of inclusion, so t h a t with this paper a comprehensive series of photographs is included,
+
Laboratory attempts a t reproducing inclusion showed t h a t contrary t o the conchisions arrived a t from factory observations, erosion is not the cause of inclusion. I n the laboratory i t was found t h a t circulation was the main factor as. even with eroded crystals as seed, no inclusion was formed when the syrup was stirred during crystal growth. With either eroded or " perfect '' seed, when the syrup was not stirred, the crystals were found t o grow with " step " formation (see Plate VIX., Figs. 643-618) and that with any subsequent growth with stirring, inclusion invariably occlxrred. This step formation was noticeable in many mill raw sugars, particularly where filterability was not good. From the above few experiments i t would appear t h a t the best defence against inclusion is (1) good clarification with resultant decrease in impurities and s~iscosity: (2) good circulation and uniform boiling conditions. On this account intermittent feed, with its resulting sudden changes in viscosity and circulation, should be avoided, as also should sudden variations in steam or vacuum.
Magma System. The investrigation in Mackay has shown cleax-ly the advantage8 of nsing magma seed, in the eyennesa of the grain, and the saving
68
PROCEEDINGS
BIBTH ANNUAL CONFERENCE .--.-.-S---.
April, 1934
of time, steam and worry in reducing the number of graining~. As everything depends then upon the C sugar grain, attention t o t h i s operation determines the quality of the final shipment sugar, and t h e photographs show clearly the variation that may occur. As in many cases i t is difficult t o avoid fine grain in the C sugar, i t was found t o be of importance t o have this effectively removed shortly after t h e introduction of the magma seed into the pan, a.s otherwise great unevenness would result. This removal could be effected by drawing in liquor first, heavying ~xpslightly t o under graining point, and then drawing in the magma, but care mnst be taken not t o heavy up the liquor too much otherwise the addition of magma will cause t h e formation of false grain. The purity of the magma being of importance, not only on account of its effect upon the quality of the grain, but also upon the quantity of impurities returned into circulation, i t is hoped t h a t next season the installation of a C massecuite crystallizer cooling and heating system a t one of the mills will show some improved results i n this direction as has been the case in Hawaii. Regarding the grain size of magma seed, Thieme (" Studies on Sugar Boiling" published by Pacts about Sugar, New York, 1928) gives interesting figures for the relation between volume and grain size of magma seed footing. Although Thieme's tables are worked out for magma of a definite srystal content they can be applied t o any magma if its consistency a t any one mill is kept constant, and by triaI and error a factor which will give accurate and consistent results can be obtained. A glance a t these tables shows the enormous influence of the size of t h e magma grain upon the quantity of seed required. Thus, taking final grain of .6 mm., the percentage of footing required is 5.33,17.33, 40.66, and 80, for grain of . 2 , .3, .4, and .5 mm. respectively. As i t is impossible t o make the C sugar always of the same grain size, the magma will vary accordingly, but hitherto no allowance was made for this in taking in magma seed, resulting in shipment sugar with great variation in size. I n the writer's opinior~the shift chemist could be easily trained t o advise the sugar boiler upon the proportions of magma t o be used, with the expenditure of only 15 t o 30 minutes per shift upon the pan stage. I n this time the average grain size of the magma can be obtained correct t o . O 1 mm. by measuring, say, 100 crystals under the microscope, and calculating the proportions from tables prepared from Thieme's figures. This will greatly assist in securing a better chemical control of the sugar boiling section, where hitherto much has been left t o the discretion of the sugar boiler. As the rate of crystal growth varies with the purity, it will b e seen from Thieme's tables that for final massecuites of the same grain size the required proportion of seed decreases with lower purities of syrup, and here again the chemist may act as adviser t o the sugar boiler. Thus with magma of 0.4 mm., t o give final grain of 0.85 mm. according t o Thieme's tables, and assuming pans of equal sizes, seed should be apportioned as follows :If making A & B. 315 seed for A. 2/5 seed for B. ,, ,, AB & B. 6/11 for AB and 5/11 for B. ,, ,, A. & AB. 519 for A. 4/9 for AB. ), ,, A. AB & B. 15/37 for A, 12/37 for AB, 10/37 for B, The above are approximate and will vary according t o the relative purities of the different grades of syrups. The breaking u p of magma
p
. 1 3 -.
------
Q 17EENSL4 K I ) SOCIETY
- -- -
--
-
S f(iAR C A X E TECHNOTAC)(+ISTS ---
--
-69 -
grain in the screw or pump wa* founcl t o inflllence thc yaality of the final grain (see Plate IT, Figs. 96-8) ancl consequently this should be avoidetl as far as possible. As the grain size of individual C's will vary, keeping the magma separate will assist in maintaining unifornl grain.
instruments for Gontrol of Sugar Boiling. Owing t o trhe pressure of other work, not much time was devoted t o these instruncnts, but the following ronclusions may be of interest:Eleetrieal conductivity apparatus.-An instrument similar t o the Cuitorneter (Xntcrnat. Sugar Journal, F. Todt. Feb. 1933, $ W. F, Alewjin, Sept).1938) \v88 tried on a graining pan but probably oning t o variations of temperature and purity, consistent conctuetivities were not obtained for graining point, so t h a t the results t o date show n o superiority over the judgment of the sugar boiler. The simplicity ancl low cost of this instrument. however, merit further study for i t s application for boiling control. Temperatnre Difference Recorder (Smith, Faet* about Sugar. Feb. & March, 1932).-Two different makes of instrnment, Leecls, Northnipp & C'arnbridge, were studied when in nse with c*alandria pans for high grade sugars (A & AB). Althougll insufficient data have been collected t o enable the construction of ideal curve# for different types of massecuites (o>$-ingprobably t o variation of purity a11tl temperat~xres.size of seed grain,etc.) the writer belieres that the use of this type of instrument is of great assistance for (1) minirnisiilg the percentage of fine grain : and ( 2 ) recorcting the method of boiling each strike and the ps~chologicaleffect of improving the work of t h e sugar boiler, as well as being a gootl guide.
Although this work was carried out primarily in order t o improve the refining cpality of the raw sugars, from the miller's point of view i t was shown that many important advantages make such improvements and investigations thoroughly worth while. Improvement i n grain size a ~ l dr e g u l a r i t ~and filterability were found t o giire t h e :following ad~aiit~ages ( a ) Better exhaustion and fugalling ; (b) Less impurities returned t o milling operations and consequerlt better boiling ; (c) d higher net titre for the same polarisation owing t o elimination of impurities, At one mill the asEl of the raw sugar was redrlerd from 0 . 2 4 t o 0.11, which represents a considerably higher return. Most authorities also state that even grain miliixnises deterioratiori ; ((1) The introd~xction of closer control of sugar boiling makes for more s;vstematic scheciules of pan floor operations and in many cases this has res~xltedin the speeding up of the work owing t o the better utilisatio~lof the pans. The writer is greatly indebted t o the managements and staffs of the &lackay RIills for their unfailing co-operation, advire and assistance dtlriilg the progress of this work, and t o the C.S.R. Company for their valuable technical advice. To m?- assistant, Mr. N. Smith, Rl.I.Pc., I am greatly indebted for his indefatigable enthusiasm in carrying ouf. so a b l ~ all the laboratory work. and for the prodnetion of many of t h e photographs in this paper.
PLATEI. Explanation of PhcitomirvograpIi8.
Fig. 151.
10 minutes after check :
Fig. 153. 2.5 minutes after check ; Fig. 153. l hour 55 minutes ; Fig. 155. Xassecuite (AB) tlropprtl ;
Fig.
(M.
7 minutes after check ;
Fig.
67.
17 minutes after ;
Fig.
68.
Half hour after ;
Figs.
381-387.
Four skips A Sugar from graining chargr; 1XlM) Lquor: 7tM) AB Syrup.
Figs. 37$-375. Four s k i p d Sugar froln prainlng (har:.
.
"
Fig. 564. See 561 and 562 ; crystals grown furt,her to 0.7 mm. with stirring ; inclusion now appeam early. 1 X v . Fig. 580. Lahorat.ory attempt a t rcpeated inclu+m. Seed areraged 0.25 mm. in superjaturatcd AB syrup, grown altrrnately a t rest and in rapid motion. 1 8 ~ . Pig. 581. See 580 ; first inolusion. Pig. 582.
ASM 580 a-ld 581 ; after aucceasivc growths at rcit and in motion.
18 ..i
Pig. 616. Repeated indusion in mill raw R Hugnr : focus 0.22 mm. helow front orthopinitcoid. 27 ;4 Fig. 617. &me cryst,ala a.? 616 ; focus 0.33 mm. helow left pole. 27 X .
.
Fig. 618. Same as 616 and 617 ; focus 0.33 mm. below basal pinaroid. Tnolusion indicates different rate* of crystal gran-t h in different tlirectionfi. 27x .
d j ~ ? * d1954. .
QTEESSLASD SOC'IETY b f ( . d R C I X E TECIISOLO(;lbTS
--
-- - --
_____-I---__--
I'L~TE \.It.
--
I______-
83
