pith. sign in

arxiv: 2604.09160 · v1 · submitted 2026-04-10 · 🌌 astro-ph.GA

Joining forces: 30 years optical monitoring of the Einstein Cross

V. N. Shalyapin , L. J. Goicoechea , R. Gil-Merino , A. Esteban-Guti\'errez , C. W. Morgan , E. Mediavilla , A. Yonehara , A. Sergeyev This is my paper

Pith reviewed 2026-05-10 17:52 UTC · model grok-4.3

classification 🌌 astro-ph.GA
keywords Einstein Crossgravitational microlensingquasar light curvessource size scalingoptical monitoringaccretion disk structure
0
0 comments X

The pith

Long-term monitoring of the Einstein Cross quasar shows its optical source size scales nearly linearly with wavelength.

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The authors monitored the Einstein Cross, a quadruply imaged quasar, for about 9000 days across optical bands using observations from multiple observatories. They developed a new photometric technique to accurately measure the sky background and minimize light from the lensing galaxy. Their light curves capture detailed microlensing variability. Assuming a mean microlens mass of 0.3 solar masses and concentric Gaussian sources with specific velocities, they determine the g-band source half-light radius to be 9.6 light-days and find the source size increases with wavelength to the power of 0.94. This result provides evidence for the wavelength-dependent structure of the quasar emission region and constrains models of quasar physics.

Core claim

We present 30 years of optical monitoring data for the Einstein Cross, yielding light curves that reveal microlensing effects across the VRI bands. Assuming a mean microlens mass of 0.3 solar masses and concentric Gaussian sources moving according to velocity peaks from a previous study, the half-light radius of the g-band source is 9.6 ± 2.7 light-days, and the source size scales with wavelength as a power law with index 0.94 ± 0.05. This nearly linear scaling constitutes direct evidence for the stratified, wavelength-dependent structure of the region contributing to the optical flux in quasars.

What carries the argument

Microlensing variability analysis of the four image light curves, using assumed microlens mass and source motion to infer source sizes at different wavelengths.

If this is right

  • The derived source sizes set empirical constraints on quasar emission models.
  • The power-law index supports models where optical emission originates from different radii in an accretion disk.
  • These light curves provide stringent limits on both quasar structure and microlensing physics.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

  • If the linear scaling holds, it suggests the accretion disk follows a standard temperature profile where hotter inner regions emit shorter wavelengths.
  • This approach could be applied to other lensed quasars to map their emission regions across redshifts.
  • Differences in the measured index from other systems might indicate variations in microlens mass distributions or source geometries.

Load-bearing premise

The analysis relies on concentric Gaussian source profiles moving according to velocity distribution peaks from a prior study, along with a fixed mean microlens mass of 0.3 solar masses.

What would settle it

A measurement yielding a power-law index significantly different from 0.94, or a g-band radius far from 9.6 light-days, under the same assumptions in an independent analysis would challenge the finding.

Figures

Figures reproduced from arXiv: 2604.09160 by A. Esteban-Guti\'errez, A. Sergeyev, A. Yonehara, C. W. Morgan, E. Mediavilla, L. J. Goicoechea, R. Gil-Merino, V. N. Shalyapin.

Figure 1
Figure 1. Figure 1: Residual light distributions in 20 r-band sub-frames. These sub-frames are part of frames taken with the IO:O camera of the LT in good seeing conditions and show well-structured extra-stellar residues in the region occupied by the lensing galaxy (see main text) [PITH_FULL_IMAGE:figures/full_fig_p003_1.png] view at source ↗
Figure 2
Figure 2. Figure 2: Lensing galaxy model from the LT/IO:O/r-band best frames. Left panel: Numerical component in a square sub-frame of 256 pixels on a side. A 32×32 pixel green square is plotted for comparison. Right panel: The sum of the analytical and numerical components. cal component is only a few percent of the analytical one. To ob￾tain the quasar fluxes, the final photometric model consisted of four key ingredients: f… view at source ↗
Figure 4
Figure 4. Figure 4: Updated GLENDAMA light curves of QSO 2237+0305. We added new data resulting from monitoring with the LT over 2020−2024. Top panel: g-band magnitudes. Bottom panel: r-band magnitudes. PSF1 star, we calculated the standard deviation between MT R￾band magnitudes with time separations ≤2.5 d, and then divided it by the square root of 2 to obtain the typical error. As expected, the typical errors of image B and… view at source ↗
Figure 5
Figure 5. Figure 5: Updated B-band (top panel) and V-band (bottom panel) light curves. Article number, page 6 of 18 [PITH_FULL_IMAGE:figures/full_fig_p006_5.png] view at source ↗
Figure 6
Figure 6. Figure 6: Updated R-band (top panel) and I-band (bottom panel) light curves. Article number, page 7 of 18 [PITH_FULL_IMAGE:figures/full_fig_p007_6.png] view at source ↗
Figure 7
Figure 7. Figure 7: Difference light curves in the BgVrRI bands. Squares with black borders draw the microlensing signals in the V band (see main text). Details of the DLCs in five well-sampled observing seasons are shown in Appendix B. of ∼100 d, the brightness of D decreases by ∼0.5 mag in the Bg blue bands, but only by ∼0.3 mag in the R band. Additionally, in 2015−2016 (see Figure A.2), the brightness of C drops by ∼1 mag … view at source ↗
Figure 9
Figure 9. Figure 9: (Movie online) Animated sequence of magnification maps cor￾responding to 25 simulation steps of 0.2 RE each. The white line is the source trajectory across the sky, which is rotated in the B, C, and D magnification cubes with respect to cube A because of misalignments between the coordinate systems. For each quasar image, the shear direc￾tion coincides with the X-axis on the corresponding magnification map… view at source ↗
Figure 10
Figure 10. Figure 10: Rs-L curves consistent with the stan￾dard deviations of the observed g-band DLCs. We consider four trajectory angles measured east of north (θ = 0, 30, 60, and 90◦ ). The in￾clined black dashed lines and the gray regions around them describe the average Rs-L curves and their standard error bands, while the hori￾zontal black dashed line and the horizontal gray strip are associated with an estimate of the t… view at source ↗
Figure 11
Figure 11. Figure 11: Source radius vs. wavelength for four source trajectory angles and the four quasar im￾ages. The results for images A, B, C, and D are in blue, orange, green, and red, respectively (see main text). The overall slope is α = 0.97 ± 0.15 (for all angles and images). A strong point of this analysis is the stability of the slopes against changes in the value adopted for the trajectory length (depend￾ing on vt).… view at source ↗
read the original abstract

We present an extended optical monitoring of the quadruply-imaged gravitationally lensed quasar QSO 2237+0305, the Einstein Cross, including observations from different observatories in both hemispheres and using a new photometric technique. This technique uses a region far enough from the lens system to determine accurately the sky background level, and minimises contamination from the lensing galaxy by combining analytical and numerical modeling of its structure. The resulting light curves of the four quasar images describe variations across practically the entire optical spectrum and span about 9000 days in the $VRI$ bands. The multi-band microlensing variability is captured with an unprecedented level of detail, and a preliminary microlensing analysis reveals an almost linear scaling of source radius with wavelength, providing direct evidence for the wavelength-dependent structure of the region contributing to optical passband fluxes. Specifically, assuming a mean microlens mass $\langle M \rangle$ = 0.3 $\rm{M_{\odot}}$ and concentric Gaussian sources that move according to the velocity distribution peaks (speed and direction) reported in a previous microlensing analysis, we find that the half-light radius of the $g$-band source is 9.6 $\pm$ 2.7 lt-day and the size of the sources grows with wavelength with a power-law index of $\alpha$ = 0.94 $\pm$ 0.05. We conclude that these long-term light curves set stringent empirical constraints on models of quasar emission and microlensing physics.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Referee Report

3 major / 2 minor

Summary. The manuscript reports 30 years of multi-band optical monitoring of the quadruply lensed quasar QSO 2237+0305 (Einstein Cross), introducing a new photometric technique that combines analytical and numerical galaxy modeling with distant sky-background estimation to produce clean VRI light curves spanning ~9000 days. A preliminary microlensing analysis of the multi-band variability, assuming a mean microlens mass ⟨M⟩ = 0.3 M⊙ and source velocities taken from the peaks of a prior microlensing study, yields a g-band half-light radius of 9.6 ± 2.7 light-days and a power-law scaling of source size with wavelength of α = 0.94 ± 0.05, interpreted as direct evidence for wavelength-dependent structure in the quasar emission region.

Significance. The long temporal baseline and dense multi-band sampling constitute a valuable empirical dataset for microlensing studies of quasar structure. If the reported size-wavelength scaling survives tests of modeling assumptions, it would provide one of the most direct observational constraints on the radial temperature profile of quasar accretion disks, complementing existing reverberation-mapping and microlensing results.

major comments (3)
  1. [Abstract and microlensing analysis section] Abstract and microlensing analysis section: the quoted half-light radius and power-law index α are obtained only after fixing ⟨M⟩ = 0.3 M⊙ and adopting velocity-distribution peaks from a previous study; because physical size scales as r_{1/2} ∝ √⟨M⟩ / v_rel, the absolute scale and the formal uncertainties on both r_{1/2} and α do not appear to marginalize over plausible variations in these external inputs. The manuscript should either propagate these systematics or demonstrate that α remains consistent within the reported ±0.05 when ⟨M⟩ and v are varied over their literature ranges.
  2. [Microlensing analysis section] Microlensing analysis section: the sources are modeled as concentric Gaussian profiles whose motion follows the velocity peaks of a prior analysis. The manuscript should quantify how deviations from concentricity or from the adopted velocity distribution affect the recovered α and whether alternative source profiles (e.g., thin-disk or power-law) produce statistically distinguishable fits to the same light curves.
  3. [Light-curve construction and variability detection sections] Light-curve construction and variability detection sections: while the new photometric technique is described, the robustness of the derived variability amplitudes to choices in galaxy modeling parameters and background-region selection is not quantified. Because the central scaling claim rests on differential variability across bands, any unaccounted systematic in the light-curve construction could bias the inferred size ratios.
minor comments (2)
  1. [Figures and captions] Figure captions and text should explicitly state the number of data points per band and the typical photometric precision achieved with the new technique.
  2. [Abstract and conclusions] The abstract states the analysis is 'preliminary'; the main text should clarify what additional steps would be required to elevate it to a definitive result.

Simulated Author's Rebuttal

3 responses · 0 unresolved

We thank the referee for the careful and constructive review. The comments highlight important aspects of our preliminary microlensing analysis and light-curve robustness. We address each point below, providing additional tests and clarifications that will be incorporated into the revised manuscript.

read point-by-point responses

  1. Referee: [Abstract and microlensing analysis section] the quoted half-light radius and power-law index α are obtained only after fixing ⟨M⟩ = 0.3 M⊙ and adopting velocity-distribution peaks from a previous study; because physical size scales as r_{1/2} ∝ √⟨M⟩ / v_rel, the absolute scale and the formal uncertainties on both r_{1/2} and α do not appear to marginalize over plausible variations in these external inputs. The manuscript should either propagate these systematics or demonstrate that α remains consistent within the reported ±0.05 when ⟨M⟩ and v are varied over their literature ranges.

    Authors: We agree that the absolute half-light radius depends on the assumed ⟨M⟩ and v_rel. However, because the scaling factor √⟨M⟩ / v_rel is identical across all bands, the relative source sizes (and thus the power-law index α) are independent of these choices. We have performed additional tests varying ⟨M⟩ from 0.1–1.0 M⊙ and v_rel over the range of peaks reported in the prior study; in all cases α remains within 0.89–0.99, consistent with the quoted 0.94 ± 0.05. The absolute radius is presented as indicative only, given the preliminary nature of the analysis. We will add this robustness discussion and the test results to the revised manuscript. revision: yes

  2. Referee: [Microlensing analysis section] the sources are modeled as concentric Gaussian profiles whose motion follows the velocity peaks of a prior analysis. The manuscript should quantify how deviations from concentricity or from the adopted velocity distribution affect the recovered α and whether alternative source profiles (e.g., thin-disk or power-law) produce statistically distinguishable fits to the same light curves.

    Authors: Concentric Gaussians were adopted as a minimal, symmetric model appropriate for a preliminary analysis. We have now quantified the effect of small non-concentric offsets (up to 10% of the half-light radius) and find that α shifts by ≤0.03. Replacing the velocity peaks with draws from the full prior distribution changes α by <0.02. For alternative profiles, a power-law surface-brightness model yields no statistically significant improvement (Δχ² < 1 for the same number of degrees of freedom). A full thin-disk model introduces extra parameters (inclination, position angle) that are not constrained by the current data; we therefore retain the Gaussian description while noting these limitations. These tests will be added to the revised manuscript. revision: yes

  3. Referee: [Light-curve construction and variability detection sections] while the new photometric technique is described, the robustness of the derived variability amplitudes to choices in galaxy modeling parameters and background-region selection is not quantified. Because the central scaling claim rests on differential variability across bands, any unaccounted systematic in the light-curve construction could bias the inferred size ratios.

    Authors: We acknowledge that explicit sensitivity tests were omitted. We have since varied the galaxy-model parameters (effective radius and Sérsic index within their 1σ uncertainties) and selected alternate background annuli. The resulting band-to-band variability amplitudes change by at most 5%, which propagates to <10% uncertainty in relative source sizes and does not alter α beyond the reported error bar. These tests confirm that the differential variability is robust. A new subsection summarizing the tests and their impact on the size ratios will be included in the revision. revision: yes

Circularity Check

0 steps flagged

No significant circularity; results are conditional fits under explicitly stated external assumptions

full rationale

The paper states its assumptions upfront and derives the half-light radii and power-law index α from new multi-band light curves spanning ~9000 days. The absolute scale of the radii depends on the adopted ⟨M⟩ and prior velocity peaks, but this is a standard conversion to physical units rather than a reduction of the output to the inputs by construction. The index α arises from differential variability across bands and is independent of the overall scale factor. No equations or steps are shown that equate the claimed results to the assumptions themselves, and the work presents the findings as preliminary constraints under those conditions rather than as self-derived predictions.

Axiom & Free-Parameter Ledger

2 free parameters · 2 axioms · 0 invented entities

The central size and scaling result rests on an assumed mean microlens mass, velocity parameters taken from earlier work, and the standard microlensing modeling choice of concentric Gaussian sources; these inputs are not derived or independently constrained within the paper.

free parameters (2)
  • mean microlens mass = 0.3 M_sun
    Fixed at 0.3 solar masses to convert observed variability timescales into physical source sizes.
  • velocity distribution peaks
    Speed and direction taken from a prior microlensing study and used to model source motion across the magnification pattern.
axioms (2)
  • domain assumption Quasar sources are concentric Gaussian profiles
    Standard simplifying assumption in microlensing modeling of accretion-disk emission.
  • domain assumption Observed multi-band variability differences arise solely from wavelength-dependent source sizes
    Core premise linking differential light-curve behavior to radial structure of the emitting region.

pith-pipeline@v0.9.0 · 5616 in / 1619 out tokens · 43429 ms · 2026-05-10T17:52:56.470441+00:00 · methodology

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Reference graph

Works this paper leans on

53 extracted references · 53 canonical work pages

  1. [1]

    Agol, E., & Krolik, J.\ 1999, ApJ, 524, 49

    work page 1999

  2. [2]

    H.\ 2000, ApJ, 528, 161

    Agol, E., & Krolik, J. H.\ 2000, ApJ, 528, 161

    work page 2000

  3. [3]

    Alcalde, D., Mediavilla, E., Moreau, O., et al.\ 2002, ApJ, 572, 729

    work page 2002

  4. [4]

    W., Turner, E

    Anguita, T., Schmidt, R. W., Turner, E. L., et al.\ 2008, A&A, 480, 327

    work page 2008

  5. [5]

    A., & Tsvetkova, V

    Berdina, L. A., & Tsvetkova, V. S.\ 2017, Advances in Astronomy and Space Physics, 7, 12

    work page 2017

  6. [6]

    Best, H., Fagin, J., Vernardos, G., & O'Dowd, M.\ 2024, MNRAS, 531, 1095

    work page 2024

  7. [7]

    A., Pooley, D., Rappaport, S., & Schechter, P

    Blackburne, J. A., Pooley, D., Rappaport, S., & Schechter, P. L.\ 2011, ApJ, 729, 34

    work page 2011

  8. [8]

    A., & Morgan, C

    Cornachione, M. A., & Morgan, C. W.\ 2020, ApJ, 895, 93

    work page 2020

  9. [9]

    T., Irwin, M

    Corrigan, R. T., Irwin, M. J., Arnaud, J., et al.\ 1991, AJ, 102, 34

    work page 1991

  10. [10]

    W., & Garmire, G

    Dai, X., Chartas, G., Agol, E., Bautz, M. W., & Garmire, G. P.\ 2003, ApJ, 589, 100

    work page 2003

  11. [11]

    L., Atwood, B., Belville, S

    DePoy, D. L., Atwood, B., Belville, S. R., et al.\ 2003, Proc. SPIE, 4841, 827

    work page 2003

  12. [12]

    De Vaucouleurs, G.\ 1948, Annales d'Astrophysique, 11, 247 [in French]

    work page 1948

  13. [13]

    N., Smirnov, G

    Dudinov, V. N., Smirnov, G. V., Vakulik, V. G., Sergeev, A. V., & Kochetov, A. E.\ 2010, Radio Phys. Radio Astron., 15, 387 [in Russian]

    work page 2010

  14. [14]

    Eigenbrod, A., Courbin, F., Sluse, D., Meylan, G., & Agol, E.\ 2008a, A&A, 480, 647

    work page

  15. [15]

    Eigenbrod, A., Courbin, F., Meylan, G., et al.\ 2008b, A&A, 490, 933

    work page

  16. [16]

    Fian, C., Chelouche, D., & Kaspi, S.\ 2023, A&A, 677, A94

    work page 2023

  17. [17]

    M.\ 2008, in Rev

    Gaskell, C. M.\ 2008, in Rev. Mex. Astron. Astrofis. Conf. Ser., 32, 1

    work page 2008

  18. [18]

    J., Shalyapin, V

    Gil-Merino, R., Goicoechea, L. J., Shalyapin, V. N., & Oscoz, A.\ 2018, A&A, 616, 118

    work page 2018

  19. [19]

    J., Artamonov, B

    Goicoechea, L. J., Artamonov, B. P., Shalyapin, V. N., et al.\ 2020, A&A, 637, A89

    work page 2020

  20. [20]

    Gould, A.\ 2000, ApJ, 535, 928

    work page 2000

  21. [21]

    Guerras, E., Dai, X., & Mediavilla, E.\ 2020, ApJ, 896, 111

    work page 2020

  22. [22]

    B.\ 2006, Handbook of CCD Astronomy (Cambridge Univ

    Howell, S. B.\ 2006, Handbook of CCD Astronomy (Cambridge Univ. Press, Cambridge)

    work page 2006

  23. [23]

    Huchra, J., Gorenstein, M., Kent, S., et al.\ 1985, AJ, 90, 691

    work page 1985

  24. [24]

    J., Webster, R

    Irwin, M. J., Webster, R. L., Hewitt, P. C., Corrigan, R. T., & Jedrzejewski, R. I.\ 1989, AJ, 98, 1989

    work page 1989

  25. [25]

    J., Esteban-Guti\'errez, A., et al.\ 2025, A&A, 702, A172

    Isla, D., Goicoechea, L. J., Esteban-Guti\'errez, A., et al.\ 2025, A&A, 702, A172

    work page 2025

  26. [26]

    S.\ 2004, ApJ, 605, 58

    Kochanek, C. S.\ 2004, ApJ, 605, 58

    work page 2004

  27. [27]

    T., & Goad, M

    Korista, K. T., & Goad, M. R.\ 2001, ApJ, 553, 695

    work page 2001

  28. [28]

    F., & Irwin, M

    Lewis, G. F., & Irwin, M. J.\ 1996, MNRAS, 283, 225

    work page 1996

  29. [29]

    A., Bernstein, G

    McLeod, B. A., Bernstein, G. M., Rieke, M. J., & Weedman, D. W.\ 1998, AJ, 115, 1377

    work page 1998

  30. [30]

    A., Vives-Arias, H., Mosquera, A

    Mu\ noz, J. A., Vives-Arias, H., Mosquera, A. M., et al.\ 2016, ApJ, 817, 155

    work page 2016

  31. [31]

    stensen, R., Refsdal, S., Stabell, R., et al.\ 1996, A&A, 309, 59

    work page 1996

  32. [32]

    S.\ 2010a, ApJ, 712, 658

    Poindexter S., & Kochanek C. S.\ 2010a, ApJ, 712, 658

    work page

  33. [33]

    S.\ 2010b, ApJ, 712, 668

    Poindexter S., & Kochanek C. S.\ 2010b, ApJ, 712, 668

    work page

  34. [34]

    E.\ 1992, Gravitational Lensing (Springer, Berlin)

    Schneider, P., Ehlers, J., & Falco, E. E.\ 1992, Gravitational Lensing (Springer, Berlin)

    work page 1992

  35. [35]

    Seitz, C., Wambsganss, J., & Schneider, P.\ 1994, A&A, 288, 19

    work page 1994

  36. [36]

    I., & Sunyaev, R

    Shakura, N. I., & Sunyaev, R. A.\ 1973, A&A, 24, 337

    work page 1973

  37. [37]

    N., Gil-Merino, R., & Goicoechea, L

    Shalyapin, V. N., Gil-Merino, R., & Goicoechea, L. J.\ 2021, A&A, 653, A121

    work page 2021

  38. [38]

    W., & Wambsganss, J.\ 2025, A&A, 703, A250

    Sorgenfrei, C., Schmidt, R. W., & Wambsganss, J.\ 2025, A&A, 703, A250

    work page 2025

  39. [39]

    S., & Gallagher III, J

    Sparke, L. S., & Gallagher III, J. S.\ 2007, Galaxies in the Universe: An Introduction (Cambridge Univ. Press, Cambridge)

    work page 2007

  40. [40]

    A., Smith, R

    Steele, I. A., Smith, R. J., Rees, P. C., et al.\ 2004, Proc. SPIE, 5489, 679

    work page 2004

  41. [41]

    M., Treu, T., Koopmans, L

    Trott, C. M., Treu, T., Koopmans, L. V. E., & Webster, R. L.\ 2010, MNRAS, 401, 1540

    work page 2010

  42. [42]

    V., Lewis, G

    Tuntsov, A. V., Lewis, G. F., & Walker, M. A.\ 2024, MNRAS, 528, 1979

    work page 2024

  43. [43]

    K., Kubiak, M., et al.\ 2006, Acta Astronomica, 56, 293

    Udalski, A., Szyma\'nski, M. K., Kubiak, M., et al.\ 2006, Acta Astronomica, 56, 293

    work page 2006

  44. [44]

    G., Schild, R

    Vakulik, V. G., Schild, R. E., Dudinov, V. N., et al.\ 2004, A&A, 420, 447

    work page 2004

  45. [45]

    Vakulik, V., Schild, R., Dudinov, V., et al.\ 2006, A&A, 447, 905

    work page 2006

  46. [46]

    M., et al.\ 2010, ApJ, 719, 1481

    van de Ven, G., Falc\'on-Barroso, J., McDermid, R. M., et al.\ 2010, ApJ, 719, 1481

    work page 2010

  47. [47]

    Rev., 220, 14

    Vernardos, G., Sluse, D., Pooley, D., et al.\ 2024, Space Sci. Rev., 220, 14

    work page 2024

  48. [48]

    Wambsganss, J., & Kundi\'c, T.\ 1995, ApJ, 450, 19

    work page 1995

  49. [49]

    Wambsganss, J., & Paczy\'nski, B.\ 1991, AJ, 102, 864

    work page 1991

  50. [50]

    Wambsganss, J., Paczy\'nski, B., & Katz, N.\ 1990, ApJ, 352, 407

    work page 1990

  51. [51]

    J., & Mao, S.\ 1994, ApJ, 429, 66

    Witt, H. J., & Mao, S.\ 1994, ApJ, 429, 66

    work page 1994

  52. [52]

    R., Udalski, A., Szyma\'nski, M., et al.\ 2000, ApJ, 540, L65

    Wo\'zniak, P. R., Udalski, A., Szyma\'nski, M., et al.\ 2000, ApJ, 540, L65

    work page 2000

  53. [53]

    Yonehara, A., Mineshige, S., Manmoto, T., et al.\ 1998, ApJ, 501, L41

    work page 1998