Search for Hidden Chambers in the Pyramids - MIT Laboratory for ...

39 downloads 193 Views 3MB Size Report
3 (left). The initial measurement (with zenith angle of counts from 20 to 40 degrees) of the .... The plane on the top o
Search for Hidden Chambers in the Pyramids The structure of the Second Pyramid of Giza is determined by cosmic-ray absorption. Luis W. Alvarez, Jared A. Anderson, F. El Bedwei, James Burkhard, Ahmed Fakhry, Adib Girgis, Amr Goneid, Fikhry Hassan, Dennis Iverson, Gerald Lynch, Zenab Miligy, Ali Hilmy Moussa, Mohammed-Sharkawi, Lauren Yazolino

The three pyramids 6f Giza are situated a few miles southwest of Cairo, Egypt. The two largest pyramids stand within a few hundred meters of each other. They were originally of almost exactly the same height (145 meters), but the Great Pyramid of Cheops has a slightly larger square base (230 meters on a side) than the Second Pyramid of Chephren (215.5 meters on a side). A photograph of the pyramids at Giza is shown as Fig. 1. Figure 2 shows the elevation cross sections of the two pyramids and indicates the contrast in architectural design. The simplicity of Chephren's pyramid, compared with the elaborate structure of his father's Great Pyramid, is explained by archeologists in terms of a "period of experimentation," ending with the construction of Cheops's pyramid (1). (The complexity of the internal architecture of the pyramids increased during the Fourth Dynasty until the time of Cheops and then gave way to quite simple designs after his time.) An alternative explanation for the sudden decrease in internal complexity from the Great Pyramid to the Second Pyramid suggested itself to us: perhaps Chephren's architects had been more successful in hiding their upper chambers than were Cheops's. The interior of the Great Pyramid was reached by the tunneling laborers of Caliph MaThe authors are affiliated with the Joint Pyramid Project of the United Arab Republic and the United States of America. They reside either in Cairo, United Arab Republic, or in Berkeley, California. The article is adapted from an address presented by Luis W. Alvarez at the Washington Meeting of the American Physical Society, 30 April 1969. 832

mun in the 9th century A.D., almost 3400 y'ears after its construction. Of our group only Ahmed Fakhry (author of The Pyramids, professor emeritus of archeology, University of Cairo, and member of the Supreme Council of Archeology, Cairo) was trained in archeology. As laymen, we thought it not unlikely that unknown chambers might still be present in the limestone above the "Belzoni Chamber," which is near the center of the base of Chephren's Second Pyramid, and that these chambers had survived undetected for 4500 years. [We learned later that such ideas had occurred to early 19th-century investigators (2), who blasted holes in the pyramids with gunpowder in attempts to locate new chambers.] In 1965 a proposal to probe the Second Pyramid with cosmic rays (3) was sent to a representative group of

cosmic-ray physicists and archeologists with a request for comments concerning its technical feasibility and archeological interest. The principal novelty of the proposed cosmic-ray detectors involved their ability to measure the angles of arrival of penetrating cosmicray muons with great precision, over a large sensitive area. The properties of the penetrating cosmic rays have been sufficiently well known for 30 years to suggest their use in a pyramid-probing experiment, but it was not until the invention of spark chambers with digital read-out features (4) that such a use could be considered as a real possibility. [Cosmic-ray detectors with low angular resolution had been used in 1955 to give an independent measure

of the thickness of rock overlying an underground powerhouse in Australia's Snowy Mountains Scheme (5)]. The favorable response to the proposal led to the establishment by the United Arab Republic and the United States of America of the Joint U.A.R.U.S.A. Pyramid Project on 14 June 1966. Cosmic-ray detectors were installed in the Belzoni Chamber of the Second Pyramid at Giza in the spring of 1967 by physicists from the Ein Shams University and the University of California, in cooperation with archeologists from the U.A.R. Department of Antiquities. Initial operation had been scheduled for the middle of June 1967, but for reasons beyond our control the schedule was delayed for several months. In early 1968 cosmicray data began to be recorded on magnetic tape in our laboratory building, a few hundred meters from the two largest pyramids. Since that time we have accumulated accurate angular measurements on more than a million cosmic-ray muons that have penetrated an average of about 100 meters of limestone on their way to the detectors in the Belzoni Chamber. Proof of the Method Before any new technique is used in an exploratory mode, it is essential that the capabilities of the technique be demonstrated on a known system. We gave serious consideration to a proposal that the cosmic-ray detectors be tested first in the Queen's Chamber of the Great Pyramid, to demonstrate that the King's Chamber and the Grand Gallery could be detected. But this suggestion was abandoned because the King's Chamber is so close to the Queen's Chamber and because it subtends such a large solid angle that earlier (low resolution) cosmic-ray experiments had already shown that the upper chamber would give a large signal. It was apparent that the only untested feature of the new technique involved the magnitude of the scattering of high energy muons in solid matter. (An anomalously large scattering would nullify the high angular resolution that had been built into the detectors, in the same way that frosted glass destroys our ability to see distant objects.) We had no reason to doubt the calculated scattering, but we were anxious to be able to demonstrate to our colleagues in the U.A.R. DepartSCIENCE, VOL. 167

Fig. 1 (top right). The pyramids at Giza. From left to right, the Third Pyramid of Mycerinus, the Second Pyramid of Chephren, the Great Pyramid of Cheops. [L National Geographic Society]

ment of Antiquities in a convincing manner that the techniquLe really worked as we had calculated. For this pturpose we required as our test objects not large features that were nearby but, instead, small featuLres separated from the detectors by the greatest possible thickness of limestone. Fortunately, such features are available in the Second Pyramid; the four diagonal ridges that mark the intersections of neighboring plane faces were farther from the detectors than any other points on the individual faces. (From now on, we will refer to these ridges as the "corners.") From the known geometry of the Second Pyramid, the trajectories of cosmic-ray muons that pass through a point on a face 10 meters from a corner and then down to the detectors can be shown to traverse 2.3 fewer meters of limestone than do muLons that strike the corner. They shoulld therefore arrive with 5 percent greater intensity than the muons from the corner. SuLch an increase in intensity, corresponding to suLch a decrease in path through the limestone, is abouLt half of what would be expected to result from the presence of a chamber of "typical size" (5 meters high) in the pyramid. Since such a chamber would necessarily be closer to the detectors, it wotuld for these two reasons be a muLch "easier object to see" than the corner. The detection equipment was therefore installed in the southeast corner of the Belzoni Chamber, with the expectation that it would first show the corners in a convincing manner, so that the presence or absence of unknown chambers could later be demonstrated to the satisfaction of all concerned. In September 1968 the IBM-1130 compuLter at the Ein Shams University Computing Center produced the data

(a) *

>2

t

I

=

E

G

Fig. 2 (bottom right). Cross sections of (a) the Great Pyramid of Cheops and (b) the Pyramid of Chephren, showing the known chambers: (A) Smooth limestone cap. (B) the Belzoni Chamber, (C) Belzoni's entrance, (D) Howard-Vyse's entrance, (E) descending passageway, (F) ascending passageway, (G) underground chamber, (fl) Grand Gallery, (1) King's Chamber, (J) QuLeen's Chamber, (K) center line of the pyramiid. 6 FEBRUARY 1970

833

I

I

Sum of NE +SW corners 0

0r- 2000

Un 4-

.F

-~

44

10

0)

i!

40-

0

c_2000

0 C) 0

E D

U)

z

I1I I1 -1 -80 -60 -40 -20 0 20 40 60 80 Distance from the corner (m)

135 180 225 270 360 Azimuthal angle 4 (deg) Fig. 3 (left). The initial measurement (with zenith angle of counts from 20 to 40 degrees) of the variation of cosmic-ray intensity Fig. 4 (light). with azimuthal angle, as observed from the Belzoni Chamber underneath the Second Pyramid of Chephren. Detection of the northeast and southwest corners of the pyramid obtained by plotting the second differences of the counting rate on the planes tangent to the corners as a function of distance from the corners. 90

for Fig. 3, which shows the variation of cosmic-ray intensity with azimuthal angle (compass direction). The expected rapid changes in cosmic-ray intensity in the vicinity of the corners were clearly shown, and the capability of the method could no longer be doubted. An analysis of more data was later made on the Lawrence Radiation Laboratory's CDC-6600 computer and is shown in Fig. 4. Here the "second differences" of the counting rate with distance from each corner are plotted on planes that are located symmetrically with respect to adjacent faces and that are tangent to the corner. Mathematically, we would expect to see a sharp spike at the corner of a sharply defined pyramid in the plot of the second derivative of counting rate with respect to distance. The second derivative becomes a second difference curve when we use bins of a finite size. The sharpness of the peaks in the second difference curves shows that the effect of the scattering of muons in limestone is somewhat smaller than the conservative estimate made in the original proposal. We were at first surprised by the large variations in maximum counting rate through the four faces of the pyramid. We knew that the Belzoni Chamber was not at the exact center of the base of the pyramid, but we had not appreciated what large changes in counting rate would be occasioned by the actual displacement of the detector from the center of the base; the equipment is 15.5 meters east and 4 meters north of the center. There are two independent ways to use cosmic-ray data

to determine the location of the detector with respect to the exterior features of the pyramid. 1) The difference in the maximum counting rate through the east and west faces gives the displacement of the detector toward the east, and similar measurements in the north-south directions give the displacement to the north. 2) The azimuthal angles of the dips corresponding to the corners give a second, quite independent, and more sensitive measure of the displacements. We can report that from cosmic-ray observations alone, "looking through" 100 meters of limestone, we can locate the position of our detectors to within 1 meter. To the best of our knowledge, no such measurement has ever been made before. Our cosmic-ray-derived position agrees to within less than 1 meter in the north-south direction with a recently surveyed position obtained by the U.A.R. Surveying Department, but it differs by 2 meters (that is, it indicates 13.5 rather than 15.6 meters) in the east-west direction. Simulated X-ray Photographs We have presented the cosmic-ray data in two different ways, one photographic and the other numerical. Both these methods involve the projection of each recorded muon back along its trajectory to its intersection with either a horizontal plane or a sphere that touches the peak of the pyramid. Figure 5a is a diagram representing the Second Pyramid with the horizontal "film

plane" touching the peak of the pyramid and with a dashed line (representing the path of a cosmic ray) passing from the detector through a hypothetical chamber to the image of the chamber on the "film plane." (The mapping of the pyramid structure by this technique is identical to what we would obtain by x-raying a small model of the pyramid, with an x-ray source in the Belzoni Chamber and with an x-ray film touching the peak of the model pyramid.) Figure Sb represents the spherical shell onto which cosmic rays were projected for numerical analysis. Figure 6 is a view of all the equipment, which occupied most of the southeastern part of the Belzoni Chamber. Figure 7 is a closer view of the detector. The two spark chambers, each 6 feet (1.8 meters) square, are separated vertically by a distance of 1 foot (0.3 meter). Above and below the spark chambers and just above the floor level were scintillation counters, which triggered the spark chambers when all three counters signaled the passage of a penetrating muon. The 4 feet (1.2 meters) of iron between the bottom two scintillators was installed to minimize the effects of muonscattering in the limestone. The simulated x-ray photograph of the pyramid shown in Fig. 13a is an uncorrected (raw data) scatter plot of 700,000 recorded cosmic-ray muons as they passed through the "film plane." The four corners of the pyramid are very clearly indicated. If a Grand Gallery and a King's Chamber were located in the Second Pyramid as they are in the Great Pyramid, the Grand Gallery SCIENCE, VOL. 167

834 d

tensity in the direction of a new chamber is simply that the integral range spectrum of the muons is represented by a power law with an exponent equal to -2. Therefore, if the rock thickness is changed by an amount AX, out of an original thickness X, the relative change in intensity is Al/I = -2AX/X. four known chambers in the two The a \VU f > large pyramids have an average height b Fig. 5. (a) Geometry of the of about 5 meters. Therefore AX/X Second Pyramid, showing the should be -5 percent, and the correprojection technique used to sponding value of AI/I should be +10 produLce a simulated x-ray photograph. The plane on the top of the pyramid can be thought of as the "film plane." (b) The spherical surface on which the events were percent.) projected for the ntumerical analysis of the data. Since the counting equipment was sensitive out to approximately +45 degrees from the vertical, our data were would have shown up clearly but the ways of the pyramid, each square plotted in a matrix with 900 entries, King's Chamber would probably have chamber comprised two chambers 3 30 X 30 bins, each 3 by 3 degrees. requLired some computer assistance to by 6 feet (0.9 by 1.8 meters) in area. Figure 5b illustrates this system of be made visible. There is one unex- Also, each of the large scintillation binning on a sphere that encircles pected feature in Fig. 13a: on the north counters was divided into sections. The the pyramid. We wrote a computer face, there appears to be a narrow inactive areas between the two pairs of program to simulate the counting rate north-south-oriented region that has spark chambers and between the sec- expected in each of these bins. As the a lower cosmic-ray intensity than is tions of the counters led in a predict- simulation program became more sofound in surrounding areas. We were able way to the unexpected signal phisticated with time, it took into account the most detailed features of the at first hopeful that the north-south shown in Fig. 1 3a. measured exterior surface of the pyrastreak indicated the presence of a mid, including the "cap" of original Grand Gallery above and north of the limestone casing blocks near the top, Belzoni Chamber, just as the Grand Numerical Analysis the surveyed position of the detectors Gallery is above and north of the We concluded from our study of the in the Belzoni Chamber, the positions Queen's Chamber in the Great Pyramid. But we later found a satisfactory simulated x-ray picture that no unex- of the walls and ceiling of the Belzoni explanation of this feature in the pic- pected features were discernible. But Chamber, and the sizes and positions tuLre that did not involve any interior since we had been looking for an in- of each of the four spark chambers structure in the pyramid. The region of crease in intensity of approximately 10 and the fourteen scintillation counters. lower cosmic-ray intensity resulted percent over a region larger than that An important control on the quality from the construction of the spark to which the eye responds easily, we of the experimental data being comchambers. Since we could not transport then turned to a more detailed numeri- pared with the simulated data came square chambers 6 feet (1.8 meters) cal analysis of the data. (The reason for from scatter plots showing the exact on a side through the small passage- expecting a 10 percent increase in in- x and y coordinates of each riecorded

Fig. 6 (left). The equipment in place in the Belzoni Chamber under the pyramid. Fig. 7 (right). The detection apparatus containing the spark chambers. 6 FEBRUARY 19708

835

found to be correlated with contaminated neon in the spark chambers; the log books show that whenever the chambers were flushed with fresh neon they recovered their substantially uniform sensitivity. By examining the scat-

passed through each of the five planes containing scintillators or spark chambers. Unsatisfactory operation of the spark chambers showed up as small blank areas in the scatter plots of muons passing through the chambers. Such unsatisfactory operation was muon as it

ter

plots

day-by-day basis,

on a

we

N

T

o 8 2 24 66

o O

1 5 11

8

1

5

18

54 114 173 219

191 269 364

54 112 163

67

0 29 110 213 315

73

w

56 78 95 76 92 78 72 88

Ie

201

367

843 639

698

210

627

832 935 922 972 1065 1119 1253 1259 1295 1184 1391 911 731 844

149

254 363 403 195 297 332 154 191 262 84 146 222 78 119 162 16 5t 175 6 11 0 8 0 0 9 0

426 303 277 233 149 95 49 0 3

545 491 433 354 229 159 62 6 9

472 515 635 562 625 628 624 648 708 791 538 668 799 046 545 782 908 946

165 230 262 302 318

312 361

364 338

591 604 536 513

786 810 726 665

949 873 896 836

941 988 926 831

225 409 532

958 294

53 29 32 21 5 8 8 o I

3 98 294

335 619 656 577 470 561 682 465 562

20 28 55

6

65

144 20 45 35 7 8 8 0

523

681 716 778 799 907 954 1111 1186 1186 973 966 886

34 23 195 259 381 511 560 724 737 899 936 1092 1013 1228 1078 1311 1118 1309 1105 1338 1082 1300 1191 1290 1325 1470 1358 1482 1 4371545 1250 1352 1138 1172 1035 1387 911 1741

59 298 559 838 1010 1206 1358 1520 1540

2606 1492 1464

1432 1492 1420 1313 1193 1156 1166

86 205 334 397 599 641 803 869 1103 1068 1208 1480 1413 1686 1634 1651 1538 1642 1457 1726 1486 1657 1404 1558 1357 1514 1102 1412 13 511315~ 1106 1266 117 1263 1159 133 1139 1265

121

136 1 2

663

393 742 034 322 573 720 847 909 931 B84 791

107 380 716 1002 1236 1461 1661 1808 1785 1780 1771 1688 1643 1448

716 975 1271 1513 1593 1761 1839 1763 1751 1704 1559

1316 1341

16 1430 1499 1354 1381

454 734 994 1263 1551 1660 1795 1870 1746 1826 1856 893 2083

15411180 1-343 g-3 1845 33

1347

2028 1890 1739 1651 1405

1730 1673 1473 1417

03 423 727 978 1249 1509 1679 1792 1810 10 35 1984 2038 2235

639 329 566 877 1280 1316 1463 1603 1604 1848 1983 2232 2 247

I4 3 24 99 276 252 159 591 450 363 298 855 675 562 39 3 040 807 680 555 1267 1045 825 759 1467 1240 1009 034 1511 1351 1283 1092 1721 1587 1417 13-6 2317 1919 1668 1533 2340 2073 1 771 1650 2396 2185 1987 1012 2688 2483 2231 2027

2339 2324 2023 2152 1867 1824 1739 0736 i5S,o 1526

2679 2524 22 78 2029 2411 2312 2077 2031 2221 2118 1981 1750 1958 1863 1635 1549 1692 1618 1510 1332

0 0 27 125

3 0 0 46

0 0 14 70

0 0 8 11 69

0 0 8 5 23

223 199 123 678 385 333 210 156

450

0 0 0 0 3 3 6 10

448 553

288 446

216 268

61 98 161

821

665

467

238

1579 1291 1259 1016 1 795 1406 1 336 1120

705 789

400 429

10776 1525 1462 1195 1731 1373 1288 1596 1523 1274 1213 981 1290 1366 1319 876 1166 936 931 717

909 795 696 159 555

481 462 415 338 320

611 591

363 292 207 147 75 23 2

231 185 103 68 38 3 8 I

49 89 28 13 8 3 0 0

566 732

507 639

096 720

1112 940 1234 1041 1430 1088

753 652 517 355 241 159 66 7

675 789 968 1774 1323 1145 1135 1198 1271 1285 1190 1118 1162 1146 1036 879 623 636 840 965 943 1372 1791 1129 1131 1173 1392 991 1000 888 834 734 517 648 789 875 035 972 1312 1089 1096 1051 1702 874 963 797 684 608 413 526 599 679 695 794 854 854 689 897 837 731 775 663 531 429 333 360 467 567 493 621 669 685 695 634 631 626 551 652 373 273 181 233 340 362 433 438 450 483 525 576 454 353 382 344 267 184 94 97 125 157 218 263 206 326 307 345 290 316 280 257 193 152 56 47 6 23 30 91 123 134 138 123 123 92 87 65 36 24 45 0 76 5 1 3 6 3 9 4 5

450

0 67 163 302

690 529 421 274 166 108 32 1 0

16

777 542 368 183

27 33

938 109 549 206 1731 914 618 307

644 518 376 258 174

519 464 322 201

14

133 49

17 0 0

6 0 0

8

53 60

E

88

Is 861 781

0

90

Fig. 8. An array containing the iinumber-s of events (uincorrected) observed during several months of operation in each 3- by 3-de-ree bin. N #84

8 8

4

4

4

8

8

8

026 6 9

16

67

8

4

0

4

2 65

195

124 262 316 88

118 182

3

644

14 274

30

8.S

0 13

0

9 27

2

2

1 74

367 573 721

3 0

3

o

33 0 33(

2

0

2

2

0

3801041ii8126 119 101 78 60 232 113 12 386 333 308 385 234167 331 663 408 706 677 595 572 455 357 19 264 565 584 696 723 j756 613 468 339 889 1029 1041 1177 1226 1237{1292 1239 1190 1078 1044 862 .699 549 35

133

62

0

0

8

8

0

0

6

6

111911954 1132 1254 1369 1458

194

w tf

319 526 728 287 673 652 231 618 570 216 353 689 178 293 610 165 236 333 189 183 268 73 138 196 68 89 165 27 60 185 0 21 51 o 8 5 8 8 0

86 82

77

I

88 62

13 39

25 17 8 1 o 8 8

0

.-r-

5

847 764

890

809 603 720 567 602 474 527 385 429 318 303 246 309 187 237 139 176 79 115 16 38 8 0

5

0

409

0 0

3

66

5

292 158 375 186 462 246 542 301 628 352

12 16 24 36 69 60

709 607

107201173 12702136551398 1197 1234 1335 1430 1743 1935 2081 212002456 2775 20890907160417961373084 9 78 1001 1154 1224 1241 11SS 1247 1297 1401 1617 86899 1883 1960 2180 203301981 1773 0535012038 1078 968 802 949 1019 1031 1085 0112 0339 1396 1 393 .466 1533 1695 1008 1938 18730170801528 1353 10111 036 856 259 828 899 1024 0157 1103'1305 1322 1388 1462 1372 1509 1510 1688 1634 1507 1328 0073 969 895 234 651 720 839 991 1116 1107 1241 1293 1301 .374 1329 132781303145981394 1285 1132 1001 807 746 611 541 659 789 931 1050 1040 1171 1214 1234 .29461249 1238 1141 1225 1168 10009 939 835 658 830 495 498 586 719 642 950 930 1047 1099 1109 1167 1132 1106 1002 1014 949 879 762 681 540 498 409 416 318 630 756 880 850 968 1014 1035 .106 1039 1017 918 917 795 707 604 532 415 377 300 326 487 512 618 704 698 802 839 849 888 852 826 741 729 628 525 440 '371 285 252 198 248 301 389 464 538 535 615 646 655 687 658 632 561 548 463 378 298 241 179 154 121 168 218 278 333 380 309 437 463 479 501 478 6498 393 383 323 265 211 162 111 85 57 9 71 206 149 194 234 241 187 309 315 333 319 301 258 242 197 009 138 73 438 23 8 0 39 35038 7 85 55 74 891058129 139 144 136 125 100 4 13270 . . I I 11 " I 3 0 7 3 3 33 79

71 79

356 287 205

215 167 113

133

69

78

37

29 6 8

9

7 11 71 67 62 55 41 27 16 7 0 0 0 0

8

901

-

-

0

i-3

446

702 414 361 6012 509 309 437 260

li 9Q0

450

8 0

459 19644372514121300310326518 822 466 ,32 246622630201817735014767140401147

14161511

-

6 0

61 27

185 121 85 46 17 438 339 298 215 137

11 99 192 297 413 478 653 820 1017 1230 1423 1418 1586 1666 1673 1757 1672 1607 1445 1427 1230 1012 889 743 613 561 427 15 139 246 356 464 530 709 881 1173 1334 1501 1507 1689 1766 1793 1882 1798 1728 1577 1556 1371 1221 1081 911 750 700 563 25 168 307 439 537 595 761 92771139 1370 1544 1529 1735 1782 1811 1903 1833 1811 1652 1746 1609 1414 1259 1087 887 835 668 38 209 371 527 633 663 815 936 1140 1373 1553 1 534 1729 1763 1641 1933 1804 1819 1796 1965 1831 1633 1438 124 1027 974 772 51 269 638 615 723 760 917 1013 1132 1339 1533 1537 1727 1803 1786 1256 1812 1175 1968 2219 2045 1828 1624 1406 1164 1105 884 66 292 699 600 820 844 1019 1121 1722 1324 0459 1450 1646 1700 1791 1978 1828 2012 2120 2309 2255 2021 1770 1570 1278 1232 993 2227 2595 2470 2199 78 3284171 552 161 882 925 1116 1234 1329 16391494 2 1821 2056 2009 2168 2245 2297 1474 1395 1309 1572 1662 1653 1851

66 f

0

0

9

4141433

243

.1350

1350

S

Fig. 9. An array containing the pr-edicted nuimber of events in each 3- by 3-degree bin for the best fit to the data.

N t

-4 -0

0

-1

3.-2 -3.

-3. 0 -3. 0 13 0 -1 0 -1 0 1I.-. -. 3 0 0

-3.

0 1

1

-1

I

-1-3

-3.

0 2

2

0

0

0

3. 0

0 3 13 1 1 0130 -.1 0 0 2 3. 2 3. 3 1 0 3. -3. -1 I

3.

2

2

0

2

1 -3.

3. 2

0 -3. 0 -3.

0

-3,0 1' 13

0

-3. -2 2 -3

13

0 " " 2 V v I I I -0 A -1 -3 0 -3. 0 -2 0 3. 0 3. -1 -1 -3. 0 0 0 -3. 0 0 -1 0 0 0

1

0

1 -2 -3.

450

-3. -3.

0 -1 -2 0 0 -2 3. -3. -3. 3. -20 2 -3. -3. -1 -1 -2 0 -3. -3. 0 0 -2 3.03. 0 -3. -3.

3.2 03 0 2 022 0 0 3. 2 3. 0 0 3. 0 -2 3 3. 2 3. 3. 0 2 -1 3. 0 2 0 3. -2 3. 0 3. 1 0 -3. 0 3 0 2 0 0 2 2 3. 2 -2 -1 0 -3. I~3 3 -2 -31 -3 -3 3. -3. 2 -1 -2 -2

i

0 3.

1 0

0 -3. 0 0 0 -2 0 -3 3. -3. -3. 0 -2 -3.

0 -3. I3 0 3. 3 2 -1 I 13 -3.1I 0 -1 -3. 3. -1 I3.-3. I3 0 3 2 -1 0 0 -2

2 -3. -2 0 -3.

-1I-

0 -2 -3.

0

3. 0 0

-3 -1

3.

0

0 3.

0

2 3. 0 -2

3.2 -3. 03.

0

-3.122

.9L 0

0

0 0 2 02

0 -2 -2 -2

o

900

3

-

-1

2

v --T-

.3.

3.-1 -3. 3. -0 0 3. -2 0 -2 0 -1 0 2 0- -3.3 -3.

0

1

0

0 -1 -1 -1

1 1

0

03.3. 2 -3-. -2 -3 -1 -3 -2 01 0 0 0 0 -2 0

0 2

0 -2

0

3.

0 -2 0 0

-2. -3.

0 -3.

3.

03. 0 0 3 0 0 2 23. -2 0 -3. -1 1 0 3. -3. 3. -3

0 -2 0 0 -3. -3. 0 2 0 2

-3.2 -3.0

-3. -3. -.1-1 0 0 -2 -2 Q 0 0-3. 2. 0-3. 0 2 002 3.0 0 3 0 -1 0 3. -3.0 0 2 0'-2 0 .0 0 1 23.1 -3. -3. 0 -3. 0 3. 0

-2 -3. 0 3. 0 3. 3. 0 3. 0 3.

0

0 -0

0 1 1 -1 -1 -1 -3 .-3. -1 -2 -1

S 836

3. -2 2 2

0 2 3 0 -2 -3. 0 2 0 1 3. -3. -3. 0 2 -3. 0 0 0 3.0 -3. 0 -2 0 -3. 0 -3. -3. 0 0 3 0 3. 0 -2 -2 3.3. 2 0 0 -2 3. 3.3. 0 2 ,. -I L-- --, A c, 1. -1 -3. I.-3.1I 0 -.-2 3. -3. -3. 0 3.3 0 -313 1 3 -3. -1 0 3 1 0 0

1 0 0 -1 3. 0 0 0 0 030

3. 3. -3. -3 0 0 0 3. 3 'I 0 0 -1 A1 0 0 0 3. -2 0 0 -3. 3. 0

0

-2 0

03. -2 -3. 03.

0 0 -1 0

0 0

0

-3. -3. -3.

-3.I3 0

0 0 0 -1 -3 0 -2 3 0 0 C 0 -2 0 2 -3. mC 7h% o 0 0 0 0 C

0 0

3. 3. 3. 3.

E

-1I3

pulw -T- v-

0O -13. -3 -1

2 2 3.

3. 0

Fig. 10 (bottom left). The differences be-

-32 -1

3. -1

tween the numbers of events measured

3 03.31 0.-3.0 0 3. o -i -3. 0 2 0 .3 -3. 0 -3.2 3. 0CC -3. .3. -2 -2 -3. -3. -2 23. -3. 3.

3.

eliminated from the data base about one-third of the measured muons. The scatter plo-ts also served as a check on the resolution and accuracy of the angle measurements. The edges of the counters showed up on these plots as sharp lines at positions that agreed well with the direct measurements of the counter locations. Neither the direct measurements of the counter positions nor the inferred positions of these counters as obtained from the data themselves were good enough to permit the program to make sufficiently accurate calculations. In a typical 3by 3-degree bin there are 1600 events. The statistical uncertainty in this numiber of events is 2.5 percent. It was necessary to make calculations to at least such an accuracy to make full use of the data. We first varied the assumed positions of the scintillat-ors by small amounts in an effort to fit the expected counts to the measured counts. This approach was unsatisfactory. Calculations of the desired accuracy were obtained only after we eliminated the events that passed near the edge of at least one of the counters. In effect, each counter was defined to be slightly smaller than it actually was, and only th-recorded muons that passed through these defined counter positions were accepted. This method eli-minated the problems associated with small displacements of the counters during the experiment, with small-angle scattering of muons in the iron, and with decreased sensitivity of the counters near their edges. About 15 percent of the events were eliminated in this way. We believe that the 650,000 muons in the final selected sample are free of important biases resulting from improper functioning of the equipment. In the course of the computer analysis, about 40 fits were made to minimize the difference between the matrices of actual and simulated counts. Although the matrices contain 30 x 30 bins each, some of the bins at the edges contain so few counts (or none at all) that the effective number of bins is close to 750. If we knew all the physi-

0

I I

-1350

I 35

and predicted expressed in integral numbers of standard deviations for the best fit to the data for which the XI was 905. (The bins for which the predicted number of events was less than 30 were not used in calculating the XI.) SCIENCE, VOL. 167

cal parameters of our detection equipm-ient, if we were equally sure of the cquations describing the cosmic-ray spectrum, and if we were, in addition, sure that the pyramid was made of solid limestone, then we would expect the x2 of the fit between the actual and the simulated data to be about 750. The carliest fits had X2'S of close to 3000, but this important parameter dropped to approximately 1400 by the time the

N 0

'A

0

1

2

I-

1

u

L

IL

-

1 0-1 0 1 1- -2 2 -3 1 0 1 01 2 1 1-1 2 00 -3 1 0 0 01 1 00 12 2-1 0 0 12 0 11-2 1-1 1 0 2 - -1 2 10-

210 - 0-1 o - 00-1 0 1 - -1 1 - 1-1 i2 1

1 -1 -1 -1 0 1 -1 1

0 E

1

0 090 0

2313-121011 1 0-3 10 00 0 110 00 0 0 01 10 3 1 0020 -2 -2-1 -1 10 -1 1 1

-10-.i21

stereophotographically determined contours of the pyramid exterior were made available to us through the courtesy of the U.A.R. Surveying Department. Figure 8 is a -matrix showing the total number of real counts recorded in each of the 900 3- by 3-degree bins. Figure 9 is one of the final simulation runs, and Fig. 10 -is the difference -between Figs. 8 and 9 expressed as the closest integral number of standard deviations. [For a bin in which the number of counts was 2500, an entry for +2 standard deviations means that the actual count exceeded the expected count by 2(2500)½2 = 100.1 If these deviations are only statistical in nature, one expects about 87 percent of the bins to have contents of - 1, 0, or ± 1, about 12 percent of the bins to have ±-2, and 1 percent of the bins to have ±-3. There is one chance in three of finding one bin having ±14, one chance in 200 of finding one bin with -+-5, and only one chance in 3 X 104 of finding one bin with ±t6, if the deviations are due only to statistical fluctuations. Thus no single bin has a significant effect unless its contents are at least ±+4. Figure 10 contains no bins

-

u

2

1 1 -2 2 2 0 -2 1 -1 -1 1 0 2 1 0 -1 1 2 -1 1 0 0 1 -2 0 -2 0 -1 1 1 0 0 0 1 1 1 0 1 -1 -1 0 1 0 0 -1 0 1 1 1 2 0 0 -2 -1 0 2 2 0 0 .2 -1 0 0 0 2 3 1 0 0 -1 2 -1 2 1 -2 0 -1 0 -2 -1 0 2 0 1 1 0 -2 0 1 0 1 -1 1 1-1 -1 0 1 1 -1 -1 2 0 -1 -1 0 0 0 -1 0 0 0 1 0 -1 -1 -1 0 2 1 2 -1 0 0 -1 0-2 0 -1 0 1 1 0 1 2 0 0 1 1 0 -1 -1 -1 0 0 2 0 2 0 -1 1 1 -2 -1. 1 0 -2 0 1- 1-1 0-2 -2 1 0 2 0 -2 -1 1 0 0 1 1 0 0 1 g0 00 1 0-1 21 1102 0 112 0-0119~,10 21 0 0 1 0 1 2 -1 -0 -0-2 1 0 -2-1 -0 1 -1 0 1 1 2' 1 10 652 010266 2-1 -1 1 -1 1 -1 -1-1 -1 0 0 0 2 -1 2 11 3010 1*1A1~.2~ 0 0-1 -2 1 1-11-2 -1 0 1-

0

2 2

0

1 0 0 10 0 1 0 0 0 1 0 0 1-2 0 0

0

2

0

1-1 0 - 3 11 2 1 21 1 0 1 1 13 0 0 2 0 00 1 -1 02 2 203 3-1 1 11 - -21 - -2 - -1 1 12- 2-

1

450

45 0

0 2

0 1

1 1 1

0 0 0 0 -1 -1 0 0 -1 -1 -2 -3

-1 0 1

1 1 0 0 0 1 2 0 1 -2 0 -1 -1

-2 -2 -1 -1

0

0 0

2 0 0 1 -1 -1 -1 -1 -2 -1 1 -1 0 -1 1 0 0 -1 -2 0 1 0 0 2 0 -1 0 3 0

-1

0

2-1 -1 0 0 0 1 1 1 0

0 -2

0

1 1 0 -1

1 0

-1 0 0 1 0 0 0 1 0

0 1 -1

0

-1

0

-1: -1 -1 -1

0

-1

-2

0

0

-1- -1 -2 1 0

1 -1 1 2 2 1

2 1 0

3

0

1

1 -1 -1 1 -1

1

1

0 1 0 -1 0 0 0 1 0 -1 2 1 0 0 0 0 -1 -1 -2 -1 -1

1350

90

Fig. 11. The display of Fig. 10 as it would have appeared had there been a "King's Chamber" in the pyramid 40 meters above the apparatus. The group of numbers larger than 3 at the center-left (shaded area) indicates the chamber's position.

simulation, the actual counts in this region were no longer systematically lower than predicted by the computer, and, the value of x2 dropped accordingly. Although the x2 of the fit between the actual and simulated data was lowere-d when the features of the cap were introduced into the simulation, this drop does not constitute the strongest proof that we were in fact detecting the cap. Figure 12 compares the measured and cosmic-ray-determined variation in thickness of the limestone cap in two 24-degree-wide strips that run over the top of the pyramid in the north-south and east-west directions. In the absence of a cap both on the real pyramid and in the simulation, we

would expect the experimental points to lie along the zero lines of deviation. The smooth curved lines are obtained from the simulation program by utilizing the recently determined contours of the pyramid. The generally good agree-ment of the data points with the prediction (Fig. 12) shows clearly that we have detected the presence of the cap through more than 100 meters of limestone. The detection of the cap was much more difficult than detection of the corners; together, these two "proofs of the -method" convinced us that we could have seen any previously unknown chamber that might exist in our "field of view."

showing ±+4. T ~North

I)etection of the Cap

2

The most distinctive feature of the Second Pyramid is the cap of original limestone casing blocks near the top. All the casing was removed fro>m the Great Pyramid in the Middle Ages, but the builders of Cairo, who "quarried" the pyramids at that time, stopped before completely stripping the Second 0

to South 2 ~t T

o

I

-2

I

I

E

West to East

0

6 FEBRUARY 1970

periment. The quantity plotted is the difference between the measured distance from the detector to the surface of the pyramid and the distance calculated under the assumption that the pyramid has no surface irregularities. The data tances indicated by the cosmic rays and are to

.3

'3)

0 0

Chephren's Pyramid Iof is observed in this ex-

teds onsrpeetteds ponsrersn

Pyramid.

Before the simulation program in the computer took account of the presence of the cap of limestone casing blocks on. the pyramid, the difference plots (like the plot given in Fig. 11) always contained a central region with a preponderance of negative entries. When the cap was properly allowed for in the

Fig. 12. This graph shows that the cap on the top

i

±

T2

-2

600

900

1200

be compared with the solid line, which represents the same distance measured by the aerial survey. These distances have been averaged over

24-degree-wide bands centered on the middle of the pyramid, one running north-south, the other east-west. 837

(ftthe ceniter of

Se:arch for ('avities

pvrI am id's

thc

baIse, the

a)pparenlt chaIIImbe mlapped us1onto Iil "Is pI 0oc Cerld ddiiiLInc-

A\s theIc

10 LIe t

di

s2I lx( to1 tbhri 2I11Is ox

h

pto led dx ltibhbeil- gceometIciria'1JIta'. oi-k \x e con ~ 1It thec coin, e)ot thixIx, sic crtax intelit Its\ In Ii meII d tb1itI thei L:))l tlie momenlcltuLli rliii_ce ot eroeic0 (4( to 7t ( IC\ 1" siIo ic id has aIn inltexitrd1 .-11IoC posse Ix~iwnrld\ ot1 2. .1 r)

ilnrist ot th1I s t ImeIc xx Wxeie e'\cited bx thei pi e-senlce of' t\\ o iposi tiI\ c rec-io,' (Itc ilaIter- couxIIots )I st i.ier thec IcilitsppaIei-l\ abidic ol i' Ii I' theC ipx\ ot thl' px riilidtir .iboiit () meiters abox c thc B3elzonii (haiti betl. BeCanse, rft theC displa"ICCletnet r)t theC Belzit C iabe to the no01ti~thann cist

oIItent Iil ot, thcse pr esecei lx tinIerI

missit iti"

the sonthic1n part ot' the \s \s,ici 0 lacc ot, the p\vrm iid I-lhe Ici atlI\ C nci C~cas 11pi et II LVnoJtjiIne ate 5\ as ab-out '1'Cpecterd. The an colail- size ot thec anomaix,1 ConIld bec elated to dis'tanlce ()III bx) aCS~i- nc11( a ccrtila si/c tar1 the ll(oli airca of, the habi7Ixxes SLimCd tha~t theC anomals~tk came1 Ii om1 a (oni"lhe size ot, C~heopsx, k nc's (hani ) It 3( meiters-, bei. it hadc to be abouh ixwxax aind its la poItion turneCd Ont to he almIos,t Ce\ar 'lx cenItra,l. L ntoi tona11~tel\V. thIsP In cc LinId pICIsisentsiea.to-cetheir \\ itli at L1i-cci ac. Ialler-anIcIilar SIIL snL 'Iat IcnaltI ox Cer aC d' ",isappea ed Is x\x Cle 1-inedIImic evict]I ,IIl theC diMenions0, ot theC apartisad olf the pxr amid that '\Cci e impoirtint InI the s I nIaILt ion p rocr!lail ntVC\chad niot atcipatecd the nieed l'oi sutch aIccur-ate

daIta. I Thec n11t itftcts 55 c obsci x ed tire fa romi mecinijone onlyv to sho\s thatIto seeing, nothing'" thrloughIfont the tinalsIs lpci-tor. xx\e hadtc thiree xcix\ e.\citinc s11iunals that disa"~ppearedf nlxv Aimte the -eC~tca,et cai-c hadL beenl taken to ma1,ke

aectlv

to

bfoth the 'ip-

thle 'e,omelllt\

WAhen the sim rLilat(1 ion pograml 5\Vas ats compleIte andI aPs coilCCi aIS 5\C couldI mIake- it. thec fit bhtxx ce thec recoi ded -)

at

01

11)1. -1 he loiml-1

abhout

nnenIxIocalls tha tIcl

st

t1

Ic f,I t cIsIIxc S

a

1istitico x.Bt a[ careu-l look 0I\ it the mati i\ otdiflicnce"s sossed thaIt I L beinc loserCl the eNpeCIedI 11.i,case inl aloeC 0ottabouIt 7 5 )) caLtmec pi- niai lx-1I "

Iim a-11tl t-itherI Lunit otlminlcrease inl dlit

tIteCe-IC Vx aLuIS root11 s,outh to nor1thl. II ssIlleCd thtit the cosmic-rtiv m1ienC IS sitS1 saie tilcfLs h ICOS 0. xv here Oi Isl 90) decree-Cs Inl ti Vcitictillxv ot lentedl eas,t est plane11. tindl O) degree f'or rtivs

ap1-

protarhing horizontalyIN riomi the nionI-i. the