It will take a while (a few months) for us to send the paper to a reviewer, for the reviewer to write something about the paper, and for us to post the review to MathSciNet. So, to start, we have the bibliographic information in place:

In the old days, when there was only the paper *Mathematical Reviews*, nothing was “added” until we had the review in hand and could print it as part of one of the big orange volumes that some of us remember. Here is a photo of Robert Bartle hand stamping the item that became MR1000000. Everyone came out to help. Clearly this is a nice day, and not late February in Michigan. (The review was published in February 1990.)

For comparison, here are dates of other million milestones, and the gaps between them

MR0000001: January 1, 1940

MR1000000: February 22, 1990 — 50 years, 1 month, 21 days [18,316 days]

MR2000000: January 30, 2004 — 13 years, 11 months, 8 days [5,090 days]

MR3000000: November 06, 2013 — 9 years, 9 months, 8 days [3,569 days]

MR4000000: September 19, 2019 — 5 years, 10 months, 14 days [2,144 days]

At this rate, we could hit MR5000000 in about 3 years!

We had a little party to celebrate. Here is our cake:

Here is a panoramic photo of us getting ready to eat the cake.

This is a great opportunity to thank the people who made this possible. First of all, there are the great people who work for the AMS at the Mathematical Reviews offices in Ann Arbor, MI. We have a lot of talented people here! Then we have the publishers, whose cooperation makes it feasible to create a database of the research literature in the mathematical sciences. Finally, there are the 20,000+ reviewers, who enrich the database with their comments and insights, making MathSciNet a very valuable tool for mathematical researchers. Thank you all!

]]>$569936821221962380720^3 + (-569936821113563493509)^3 + (-472715493453327032)^3 = 3$.

As explained below, this is both amazing and predicted. Some earlier computer attempts turned up no new solutions from what Mordell had already found. Nevertheless, Heath-Brown expected an infinite number of solutions, and even estimated their density. However, it wasn’t until Booker and Sutherland had cracked the much harder nuts 33 and 42 that this new solution for 3 was found. It sure helps to have access to half a million cores!

Sutherland gave me permission to quote his message, which tells the latest story well.

[Note: I made minor edits, including some formatting to work in WordPress.]

---- Begin included message ----

While this might not be as exciting to Douglas Adams fans, it arguably has more mathematical significance, given the history of the problem. As Mordell wrote in 1953 [1,p.505],

“I do not know anything about the integer solutions of beyond the existence of the four sets (1, 1, 1), (4, 4, —5) etc.; and it must be very difficult indeed to find out anything about any other solutions*

This spurred several unsuccessful searches for other solutions to $x^3 +y^3 + z^3 = 3$, including [2,3], which led some to speculate that there were no further solutions. But in 1992 Heath-Brown conjectured that there should be infinitely many, and estimated that the density of such solutions was such that the next expected solution was simply beyond the range that had been searched to that point, writing [4,p.623]:

“A search for solutions to these equations (Gardiner, Lazarus, and Stein [3]) found only (1, 1, 1) and (4,4,-5), corresponding to $k = 3…$ Indeed, it was suggested that these equations have no more solutions. Inasmuch as the search range was only to $2^{16} \sim 6.5 \times 10^4$, the figures above indicate that any such conjecture is premature.”

Heath-Brown’s calculations suggested that one might expect the next solution to satisfy max(|x|,|y|,|z|) < 10^8, but later searches [5,6,7] yielded no new solutions even with $\max(|x|,|y|,|z|) < 10^{15}$.

In the solution we found, $\max(|x|,|y|,|z|) > 10^{20}$, but we were aided by the fact that $\min(|x|,|y|,|z|) < 10^{18}$, and by optimizations to the algorithm specific to the case $k=3$ (including an application of Theorem 1 of [4]).

From a second message from Sutherland:

I also should have mentioned this short paper by Cassels, which proves a congruence for the case $k=3$ that we exploited

MR0771049

Cassels, J. W. S.

A note on the Diophantine equation $x^3+y^3+z^3=3$.

Math. Comp. 44 (1985), no. 169, 265–266.

Heath Brown’s Theorem 1 in [4] is a generalization of Cassels’s result.

**References**

[1] MR0056619

Mordell, L. J.

On the integer solutions of the equation $x^2+y^2+z^2+2xyz=n$.

J. London Math. Soc. 28 (1953), 500–510.

[2] MR0067916

Miller, J. C. P.; Woollett, M. F. C.

Solutions of the Diophantine equation $x^3+y^3+z^3=k$.

J. London Math. Soc. 30 (1955), 101–110.

[3] MR0175843

Gardiner, V. L.; Lazarus, R. B.; Stein, P. R.

Solutions of the diophantine equation $x^{3}+y^{3}=z^{3}-d$.

Math. Comp. 18 (1964), 408–413.

[4] MR1146835

Heath-Brown, D. R.

The density of zeros of forms for which weak approximation fails.

Math. Comp. 59 (1992), no. 200, 613–623.

[5] MR2299795

Beck, Michael; Pine, Eric; Tarrant, Wayne; Yarbrough Jensen, Kim

New integer representations as the sum of three cubes. (English summary)

Math. Comp. 76 (2007), no. 259, 1683–1690.

[6] MR2476583

Elsenhans, Andreas-Stephan; Jahnel, Jörg

New sums of three cubes. (English summary)

Math. Comp. 78 (2009), no. 266, 1227–1230.

[7] Sander G. Huisman, Newer sums of three cubes, 2016. arXiv:1604.07746:

---- End included message ----

]]>

The problem is to try to represent each positive integer less than or equal to 100 as the sum of three cubes: $x^3 + y^3 + z^3=n$, where $n$ is the given integer between 1 and 100 and $x$, $y$, and $z$ are the integers you need to find. Some choices of $n$ are easy. Some are known to be impossible, such as any $n$ that is equal to 4 or 5 mod 9. Until recently, the answer was known for every $n$ except $n=33$ and $n=42$. Then, in early 2019, Andrew Booker announced a representation of 33 as three cubes, using some good ideas and an enormous amount of computing time. Hurray! Booker was rightly celebrated for his result. See, for example, the write-up in Quanta Magazine. But the question remained for one more number: 42.

To try to crack 42, Booker teamed up with Andrew Sutherland. Sutherland had a track record of breaking records using massively parallel computing. Given the amount of time needed to solve 33, it seemed like parallel computing would be necessary (or at least a good idea) for 42. They used Charity Engine, a global parallel computer that uses millions of idle personal PCs^{(*)}. After some millions of hours of computer time, Booker and Sutherland came up with the answer:

$ (-80538738812075974)^3 + (80435758145817515)^3 + (12602123297335631)^3 = 42$

Ta Dah!

The problem has a rich history, with lots of powerful ideas that produce integers with quite a few digits whose cubes add up to one- or two-digit numbers. Reviews of a handful of papers on the subject are provided below. There is also a nice video from Numberphile about the problem, which was posted to YouTube in November 2015, before Booker’s work on either 33 or 42, and a newer video posted just a few days ago, that discusses 42.

^{(*)} The use of a global network of otherwise idle computers is also used in the hunt for Mersenne primes. The GIMPS project has been finding record large primes this way since 1996.

**MR1850598**

Elkies, Noam D.(1-HRV)

Rational points near curves and small nonzero $|x^3-y^2|$ via lattice reduction. (English summary) *Algorithmic number theory (Leiden, 2000), *33–63,

Lecture Notes in Comput. Sci., 1838, *Springer, Berlin,* 2000.

11D75 (11G50 11H55 11J25)

The author gives an algorithm for effective calculation of all rational points of small height near a given plane curve $C$. Cases treated include the inequality $|x^3 + y^3 – z^3| < M \quad (0 < x \le y < z < N)$. Its solutions can be enumerated in heuristic time $\ll M\log^CN $ provided $M \gg N$, using $O(\log N)$ space. The algorithm is adapted to find all integer solutions of $0 < |x^3 – y^2| \ll x^{1/2}$ with $x < X$ in (rigorous) time $O(X^{1/2} \log^{O(1)}X)$. The strength of the algorithm, which involves linear approximation to the curve and lattice reduction, is illustrated by an example of integers $x,y$ with $x^{1/2}/|x^3 – y^2|$ greater than 46. The previous record was 4.87. The author discusses variations and generalizations. For example, fix an algebraic curve $C/\Bbb{Q}$ and a divisor $D$ on $C$ of degree $d > 0$. One can find the points of $C$ whose height relative to $D$ is at most $H$ in time $A(\epsilon)H^{(2/d)+\epsilon}$. Here $A(\epsilon)$ is effectively computable for each $\epsilon > 0$.

{For the collection containing this paper see MR1850596.}

Reviewed by R. C. Baker

MR1146835

Heath-Brown, D. R.(4-OXM)

The density of zeros of forms for which weak approximation fails.

*Math. Comp.* 59 (1992), no. 200, 613–623.

11G35 (11D25 11P55)

Let $f_k(x_1,x_2,x_3,x_4)=x^3_1+x^3_2+x^3_3-kx^3_4$. The author shows that if ${\bf x}$ is a primitive solution of $f_2({\bf x})=0$, then one of $x_1,x_2,x_3$ is divisible by 6. Thus $f_2$ has no rational zero close to both $(0,1,1,1)\in{\bf Q}^4_2$ and $(1,0,1,1)\in{\bf Q}^4_3$. Since these are zeros of $f_2$, the “weak approximation” principle fails for $f_2$. He obtains an analogous result for $f_3$.

Let $R_k(N)=\#\{{\bf x}\in{\bf Z}^4\colon\;\max_{i\leq 3}|x_i|\leq N$, ${\bf x}$ primitive, $f_k({\bf x})=0\}$. The author computes $R_k(1000)$ $(k=2,3)$ and observes some agreement with the hypothetical Hardy-Littlewood formula (1) $R_k(N)\sim {\frak S}_kN$. Subject to a conjecture on uniform convergence of certain products, he establishes that $\sum_{K<k\leq 2K}^*R_k(N)\sim N\sum^*_{K<k\leq 2K}{\frak S}_k$, where $\sum^*$ indicates omission of cubes. Perhaps, despite failure of the weak approximation principle, (1) is correct.

Reviewed by R. C. Baker

MR0000026

Davenport, H.

On Waring’s problem for cubes.

*Acta Math.* **71, **(1939). 123–143.

10.0X

There are very few results concerning Waring’s problem which may be called best possible. One of these is the main theorem of this paper: If $E_s(N)$ denotes the number of positive integers not greater than $N$ which are not representable as a sum of $s$ positive integral cubes, then $E_4(N)/N\to 0$ as $N\to \infty$. A simple argument shows that the theorem is false if $s < 4$.

The new weapon used here has been explained in detail in several papers by the author [C. R. Acad. Sci. Paris 207, 1366 (1938); Proc. Roy. Soc. London 170, 293–299 (1939); Ann. of Math. 40, 533–536 (1939)]. The remainder of the proof is similar to that given by Landau [Vorlesungen über Zahlentheorie, Bd. 1, 235–303] for the third Hardy-Littlewood theorem [Satz 346] as adapted by Davenport and Heilbronn [Proc. London Math. Soc. (2) 43, 73–104 (1937)] to the problem of representations by two cubes and one square. Interesting parts of the proof are the use of Poisson’s summation formula in Lemma 7 and the handling of the Singular Series.

Reviewed by R. D. James

]]>You may also check out the video of his talk at IAS on the work, available on YouTube.

**MR3814652**

Eskin, Alex(1-CHI); Mirzakhani, Maryam(1-STF)

Invariant and stationary measures for the ${\rm SL}(2,\Bbb R)$ action on moduli space. (English summary)

*Publ. Math. Inst. Hautes Études Sci.* 127 (2018), 95–324.

37D40 (22E50 37C85)

This monumental work has a deceptively simple objective. There is a natural action of ${\rm{SL}}_2(\Bbb{R})$ on the space ${\rm{GL}}_2(\Bbb{R})/{\rm{SL}}_2(\Bbb{Z})$; its ergodic and dynamical properties are well understood, and there is an extensive arsenal of tools from entropy theory, conditional measure techniques, measure rigidity, and Ratner theory available to study it. Here this action is thought of as the natural action of ${\rm{SL}}_2(\Bbb{R})$ on the space of flat tori, and this action is generalized to an action of ${\rm{SL}}_2(\Bbb{R})$ on the space ${{\mathcal H}}(\alpha)$ of translation surfaces, parameterized by a partition $\alpha=(\alpha_1,\dots,\alpha_n)$ of $2g-2$ for a fixed genus $g\geqslant1$. The main emphasis is on finding analogous rigidity and stationarity results in this setting, subsuming and generalizing much earlier work. While some of the results are inspired by the Ratner theory of unipotent flows on homogeneous spaces, much is different in this setting. In particular, the dynamical properties of the unipotent (upper triangular) flow are not understood well enough to be used, so the fundamental ‘polynomial divergence’ technique from unipotent flows on homogeneous spaces is not available. Instead, and in a setting where there is little control over the Lyapunov spectrum of the geodesic (diagonal) flow, new ideas are brought in to allow the ‘exponential drift’ technique of Y. Benoist and J.-F. Quint [Ann. of Math. (2) 174 (2011), no. 2, 1111–1162; MR2831114] to be used. This enormously understates the complexity of the work, which in fact makes use of many of the most significant results in the ergodic and rigidity theory of homogeneous dynamics. The authors have gone to great lengths to explain the overall view of the proofs, and take pains to explain where and why the main technical problems arise.

Reviewed by Thomas Ward

]]>Besides free MathSciNet at ICIAM in Valencia, it is possible to have demos of how to use MathSciNet.

Here I am talking with two mathematicians from Morocco about using the Mathematical Reviews Journal Database behind MathSciNet to help decide where to send a paper. We began by searching for all items in MSC 49 [Calculus of variations and optimal control; optimization] that were published in the last three years and that were published in journals (rather than in conference proceedings or as books). We then looked at the **Journals** facet in the sidebar to see what the most frequently occurring journals in the list were. The most frequent was *Journal of Optimization Theory and* Applications. The second most frequent was *SIAM Journal on Control and Optimization*. We clicked on the journal link from one of the items in* SIAM Journal on Control and Optimization*, which took us to the MR Journal Profile page for it. From there we could see all sorts of information about the journal: the number of articles published per year, the most commonly occurring subject areas, citation data, and a link to the journal’s website, where an author could find information about how to submit a paper. Next, we repeated the procedure for MSC 65 [Numerical analysis], which led us to look more closely at *Journal of Computational and Applied Mathematics* and *SIAM Journal on Scientific Computing*.

My earlier blog post, Journal Profile Pages, has more details about using the MR Journal Profile Pages. Take a look!

]]>**ICIAM Collatz Prize:**

Siddhartha Mishra

(ETH Zürich, Switzerland)

**ICIAM Lagrange Prize**

George Papanicolaou

(Stanford University, USA)

**ICIAM Maxwell Prize**

Claude Bardos

(Université Paris Denis Diderot, France)

**ICIAM Pioneer Prize **

Yvon Maday

(Sorbonne University, Paris, France)

**ICIAM Su Buchin Prize **

Giulia Di Nunno

(University of Oslo, Norway)

Mathematical Reviews will be at the ICIAM in Valencia next week, July 15-19. Naturally, we will be at the AMS booth, which is at locations 7, 8, and 9 in the Exhibit Area.

The AMS is sponsoring some extras at the coffee break on Tuesday, July 16, from 16:30 to 17:00 in the Exhibit Area. Stop by then – or any other time.

As with other important meetings, we have arranged for free access to MathSciNet at the conference sites and most of the affiliated hotels. You won’t need to do anything special, other than to connect using the internet or wifi service of the conference sites or hotels.

We hope to see you in Valencia next week!

]]>In what follows, I will highlight some of the new features. However, I also invite you to do some exploring of your own, since there are some new features that I will gloss over and seeing some of them on your own will be a lot more fun!

The main bibliographic information about the journal is in the first card, labeled **Journal Details**. Here is an example, using the *Proceedings of the London Mathematical Society*. I have chosen this journal because it has just enough features to make the example interesting without having so many features to make it complicated.

The title of the journal is at the top of the page. Some publisher information is at the top of the card, along with one or more links to the publisher’s website for the journal. The number after **Publications Listed** tells how many items (articles) from the journal have been included in the Mathematical Reviews Database and uploaded to MathSciNet. The number is actually a link that will take you to a listing of all 6,254 items, displayed as in the results of a MathSciNet publication search. The number for **Publications Cited** tells how many of the articles have been cited from the reference lists in MathSciNet. In this case, that number is 3,890, which is 62.2% of all the 6,254 articles we have for the journal in the database. The total number of citations in MathSciNet to *Proceedings of the London Mathematical Society* is 52,865. Apparently some of the items are receiving a lot of multiple citations! These 52,865 citations come from 42,917 items in the database. Since this number is smaller than 52,865, the pigeonhole principle tells us that some of these items are citing more than one article in *Proceedings of the London Mathematical Society**.*

The **Journal Details** card includes information about the **Latest Issue** we have, including a link to the contents of that issue in MathSciNet. It also includes information about the **Earliest Issue** that we have in the database, including a link to the contents of that issue in MathSciNet.

The astute observer will have noted that the first issue listed is not from the Third Series of the *Proceedings of the London Mathematical Society**. *Indeed, we have collected all the incarnations of the journal into a journal group. If you want to see the main constituents of the group, they are in the **Concise History** card. If a journal changes publishers or changes its title (from Second Series to Third Series, for instance), we catalogue that information. More details about the constituents, including notes on the changes from one to another, can be found by clicking the **Journal Title History** button in the **Concise History** card. See the note at the end of this blog post for a little more information about what is in the **Journal Title History**.

The **Mathematical Citation Quotient (MCQ)** for a given year is defined as the number of times the items published in the journal in the previous five years were cited by items in reference list journals published in the given year, divided by the number of articles the journal published in that same five-year period. The **All MCQ** is the MCQ computed as if every item (every journal article, proceeding article, book, and thesis) had been published in one big journal. It provides a benchmark by which to understand the MCQ of a journal. In the new journal profile pages, the MCQ for each year can be seen graphically all at once, along with the **All MCQ** for each year. (Note that we don’t compute MCQs before the year 2000.) Here is the graphical representation of the MCQ data for the *Proceedings of the London Mathematical Society**.*

If you prefer numbers, clicking on the table tab provides the yearly MCQ data for the journal: the MCQ, the number of relevant citations, and the number of relevant publications in the journal. You can also see the table for the MCQ via its tab.

**Pro Tip**: In the bar graphs, clicking the small camera icon in the upper right-hand corner will download a PNG image of the graph to your desktop. For the tables, you can highlight the table to copy and paste to a text document or a spreadsheet.

This card presents data on the citations to the journal. The first tab shows a bar graph of the citations, broken down by the publication year of the cited papers. You can choose whether the citing articles have been published in a particular year or look at the data over all time. The default is to use all time. Change it by selecting a year from the **Citation Year** box and clicking **Update**. The horizontal axis shows the publication years for the particular journal. The default range of years is from 1990 to the present. However, if the journal started publishing after 1990, the graph is adjusted accordingly. There is a slider below the graph that allows you to adjust the left and right endpoints of the year range being displayed. Here is the bar graph for citations from articles published in 2017 to articles in the *Proceedings of the London Mathematical Society** *in the year range 1900 to 2018:

If you want to see the data in tabular form, just click **Table**. Here are the top few rows of the tabular form of the data in the above bar graph:

Year |
Citations by Year |
Self Citations |
Total Publications |
% of Pub. Cited |
---|---|---|---|---|

2017 | 30 | 0 | 67 | 77.6% |

2016 | 53 | 0 | 59 | 83.1% |

2015 | 141 | 3 | 89 | 93.3% |

2014 | 112 | 2 | 95 | 94.7% |

2013 | 116 | 2 | 86 | 96.5% |

2012 | 109 | 2 | 78 | 96.2% |

2011 | 89 | 1 | 64 | 98.4% |

2010 | 66 | 0 | 56 | 92.9% |

The last column tells you what percentage of the articles in that year of the journal have been cited in MathSciNet. You can have a year like 2010 where there are more citations than papers published in the journal, but still have 7.1% of the papers uncited. Clearly this means that some articles have multiple citations. The entries in the second and fourth columns are links. The links in the second column takes you to the citing papers as listed in MathSciNet. You may encounter a list of fewer items than the number of citations in the table, since it is possible for one paper to cite more than one article in the journal. In the results list, those papers are noted with the phrase “Multiple citations from this item to Proc. Lond. Math. Soc.” (using the abbreviated name of the journal in question). When you click through to the *Proceedings of the London Mathematical Society **Proc. *are boxed. Note: The link specifies a particular publication year for the journal – citations to a different year of the journal will not be boxed.

You can also see the journals that are citing this journal most frequently, either in a bar graph or in a table. Clicking on the name of the journal takes you to its Journal Profile Page. Here is the table for the *Proceedings of the London Mathematical Society* for the citing year 2017:

Publication |
Citations |
---|---|

J. Algebra | 102 |

Adv. Math. | 100 |

J. Math. Anal. Appl. | 59 |

Trans. Amer. Math. Soc. | 53 |

Proc. Amer. Math. Soc. | 49 |

Comm. Algebra | 48 |

Int. Math. Res. Not. IMRN | 41 |

Math. Z. | 38 |

J. Number Theory | 37 |

J. Funct. Anal. | 31 |

Classifications

It is possible to see the distribution of MSCs (Mathematics Subject Classifications) in the journal, using either a three-year window or over all time. By default, just the top ten classes are displayed, but clicking **Show All** displays all of them. Here are the top ten for the *Proceedings of the London Mathematical Society* over all time:

Notice that **Other** is the most common class that occurs. The Mathematics Subject Classification has not existed for as long as the *Proceedings of the London Mathematical Society* has been publishing. Indeed, the journal began publishing papers in 1865, before there was an American Mathematical Society and long before there was a Mathematical Reviews! We received bibliographic data from the journal that allowed us to back fill those years. The items that come from the contributed data do not have classes attached to them and are marked as **Other**. Clicking on the two-digit MSC brings you to a full description of the subject class. Clicking on the number in the **Count** column takes you to a listing of all the items in the journal with that class as primary class from the time period selected.

The last feature I want to point out is the table of the authors most frequently appearing in the journal. As with the MSCs, you can look either at the most recent three years or over all time. Here is the table for the *Proceedings of the London Mathematical Society** *over all time.

Clicking on the author’s name takes you to his or her Author Profile Page in MathSciNet. Clicking on the number in the **Papers** column takes you to the MathSciNet listing of the author’s papers in the journal.

I have skipped over a few things, but I hope they will be sufficiently self-evident. Besides, I don’t want to steal your enjoyment of discovering what’s new with the MathSciNet Journal Profile Pages!

For cataloging and bibliographic reasons, different incarnations of the same journal are distinguished in the database. For instance, a journal may change publishers, as the *Proceedings of the London Mathematical Society* has done a few times, starting off with Francis Hodgson, moving to Cambridge University Press, and now being published by Wiley. It is also possible that a journal changes its title, say from *Bulletin of the American Mathematical Society* to *American Mathematical Society. Bulletin. New Series*. Or, a journal may split, as *Comptes Rendus Hebdomadaires des Séances de l’Académie des Sciences. Séries A et B* did, creating three journals, including *Comptes Rendus des Séances de l’Académie des Sciences. Série I. Mathématique.* * *The new journal profile pages provide information on these transitions in the **Journal Title History** area. For foreign-language journals, the profile page has information about translations that are also covered by Mathematical Reviews.

The new Journal Profile Pages were created by **Travis Smith** and **Erol Ozil**. They were guided by the tremendous staff of Mathematical Reviews, who provided great suggestions and feedback. They had particular help from **Kathy Wolcott** and **Norman Richert**.

We had a prototype of the Journal Profile Pages at the Joint Mathematics Meetings in Baltimore (January 2019). We received valuable suggestions from quite a few mathematicians, librarians, and publishers at the JMM. Some of the librarians took extraordinarily close looks at the prototype and provided insightful comments. We are especially grateful for the time they devoted to helping improve the pages.

]]>To find the list, login as a reviewer via the MR Reviewer page: https://mathscinet.ams.org/mresubs/. After logging in, you are at your reviewer home page. Here is a screenshot from mine:

The arrow points to the new link. Let’s follow it! We come to my reviewer history page. Here is a screenshot of the top of that page:

Now I can see a list of all my reviews. Moreover, we have set it up so that you can copy and paste the list into a text file. **Note**: if you copy into a WORD file, you may want to use the “Copy and Match Formatting” option in WORD.

The list in your Review History is **not** set up for easy pasting into a BiBTeX file. However, getting such a BibTeX list has been possible for a while now, but *requires that you do it while connected via a subscription to MathSciNet*. Here is how.

On your reviewer home page is a link to your Author Profile Page on MathSciNet, indicated by the red arrow in the screenshot below:

Clicking the link takes you to your Author Profile Page on MathSciNet. Below the basic information on the Author Profile Page, is a box with some links, including a link to a list of all of your reviews in MathSciNet:

The results list in MathSciNet looks like this:

To continue, click the link **Mark All**. The default format for batch downloads is the full review in HTML. However, we want to list all the items in BibTeX format. The arrow points to the button that allows you to change the format for batch downloading the results. Choose **Citations (BibTeX)**. Then click **Retrieve Marked**. The result is a list of all your reviews (well, in this case, all *my* reviews) in standard BibTeX format. Here is the top of that list:

@article {MR3185209, AUTHOR = {Olshanski, Grigori}, TITLE = {Projections of orbital measures, {G}elfand-{T}setlin polytopes, and splines}, JOURNAL = {J. Lie Theory}, FJOURNAL = {Journal of Lie Theory}, VOLUME = {23}, YEAR = {2013}, NUMBER = {4}, PAGES = {1011--1022}, ISSN = {0949-5932}, MRCLASS = {22E30 (41A15)}, MRNUMBER = {3185209}, MRREVIEWER = {Edward G. Dunne}, }

@article {MR3072155, AUTHOR = {Boyer, Adrien}, TITLE = {Semisimple {L}ie groups satisfy property {RD}, a short proof}, JOURNAL = {C. R. Math. Acad. Sci. Paris}, FJOURNAL = {Comptes Rendus Math\'{e}matique. Acad\'{e}mie des Sciences. Paris}, VOLUME = {351}, YEAR = {2013}, NUMBER = {9-10}, PAGES = {335--338}, ISSN = {1631-073X}, MRCLASS = {22E45}, MRNUMBER = {3072155}, MRREVIEWER = {Edward G. Dunne}, DOI = {10.1016/j.crma.2013.05.007}, URL = {https://doi.org/10.1016/j.crma.2013.05.007}, } etc.

You can now copy them and paste them into a BibTeX file!

]]>