A Prime To Remember

My diurnal age today is 27617 and the second interesting property that I discovered about this number (my first being that it's prime) was the fact that:$$ \begin{align} 2^1+7^1+6^1+1^1+7^1 &= 23 \\2^2+7^2+6^2+1^2+7^2 &= 139 \\2^3+7^3+6^3+1^3+7^3 &= 911 \end{align}$$The sums of the digits raised to the first, second and third powers (23, 139 and 911) are all prime. Numbers with this property constitute OEIS A176179:

A176179: primes such that the sum of digits, the sum of the squares of digits and the sum of 3rd powers of their digits is also a prime.

In the range up to 40,000, there are 322 numbers that are members of this sequence. They are:

11, 101, 113, 131, 199, 223, 311, 337, 353, 373, 449, 461, 463, 641, 643, 661, 733, 829, 883, 919, 991, 1013, 1031, 1103, 1301, 1439, 1451, 1471, 1493, 1499, 1697, 1741, 1949, 2089, 2111, 2203, 2333, 2441, 2557, 3011, 3037, 3307, 3323, 3347, 3491, 3583, 3637, 3659, 3673, 3853, 4049, 4111, 4139, 4241, 4337, 4373, 4391, 4409, 4421, 4481, 4603, 4663, 4733, 4919, 4931, 5303, 5503, 5527, 5639, 5693, 6043, 6197, 6337, 6359, 6373, 6719, 6733, 6791, 6917, 6971, 7411, 7433, 7691, 8209, 8353, 8803, 8887, 9091, 9109, 9341, 9413, 9419, 9431, 9491, 9901, 9941, 10099, 10103, 10141, 10211, 10301, 10499, 10909, 10949, 11003, 11047, 11113, 11117, 11131, 11159, 11171, 11173, 11243, 11261, 11311, 11317, 11399, 11423, 11443, 11489, 11519, 11621, 11731, 11777, 11821, 11939, 12011, 12101, 12143, 12161, 12211, 12347, 12413, 12437, 12451, 12473, 12541, 12583, 12611, 12743, 12853, 13001, 13049, 13171, 13241, 13313, 13331, 13339, 13421, 13441, 13487, 13711, 13933, 14011, 14051, 14071, 14107, 14143, 14251, 14321, 14327, 14341, 14387, 14431, 14543, 14549, 14723, 14783, 14891, 15241, 15401, 15443, 15823, 16097, 16361, 16631, 17041, 17401, 17483, 17609, 17627, 17977, 18121, 18149, 18211, 18253, 18523, 18743, 19009, 19139, 19319, 19333, 19391, 19403, 19777, 19841, 19913, 20023, 20089, 20333, 20441, 20809, 21011, 21101, 21121, 21143, 21211, 21341, 21347, 21611, 21767, 22003, 22027, 22111, 22229, 22447, 22481, 22483, 23417, 23581, 23741, 24113, 24137, 24151, 24247, 24281, 24317, 24371, 24443, 24821, 24977, 25057, 25183, 25253, 25411, 25453, 25523, 25583, 25589, 26111, 26177, 26399, 26717, 26993, 27143, 27431, 27479, 27527, 27617, 27749, 27947, 28111, 28351, 28513, 28559, 29989, 30011, 30307, 30323, 30347, 30367, 30491, 30637, 30703, 30763, 30853, 30941, 31247, 31333, 31393, 31847, 31991, 32141, 32303, 32411, 32969, 33023, 33113, 33203, 33311, 33331, 33353, 33391, 33533, 33863, 33931, 34019, 34127, 34141, 34211, 34217, 34439, 34703, 34721, 34781, 34871, 35069, 35083, 35281, 35803, 36037, 36073, 36161, 36299, 36307, 36383, 36389, 36697, 36833, 36929, 37003, 38053, 38639, 38693, 39041, 39119, 39133, 39191, 39313, 39443, 39667, 39863

However, 27617 has an even more exclusive claim to fame as shown below where the rightmost numbers represent digit sums. Here it is:$$ \begin{align}27617^1 &= 27617 \rightarrow 23 \\27617^2 &= 762698689 \rightarrow 61\\27617^3 &= 21063449694113 \rightarrow 53\\27617^4 &= 581709290202318721 \rightarrow 67\\27617^5 &=16065065467517436117857 \rightarrow 101\\27617^6 &=443668913016429033266856769 \rightarrow 127\\27617^7 &= 12252804370774720611730783389473 \rightarrow 131 \end{align}$$All the numbers on the right (23, 61, 53, 67, 101, 127 and 131) are prime. 27617 is the smallest prime with this property that qualifies it for membership in OEIS A131748:

A131748: minimum prime that raised to the powers from 1 to \(n\) produces numbers whose sums of digits are also primes.

In the case of 27617, \(n=7\) and the initial members are: 2, 5, 739, 47, 4229, 2803, 27617, 142589, 108271, 2347283, 1108739, 300776929, 300776929, 14674550173, 92799126239

It would appear that 27617 is the only prime that is a member of both these OEIS sequences.

Deceptive Numbers

It's not surprising that I've not come across deceptive numbers before as they are quite light on the ground so to speak. Today, 27613 is the number associated with my diurnal age and it is a deceptive number but the previous such number was 24661 and the next will be 29431. So what constitutes a deceptive number? Here what Numbers Aplenty has to say:

Let us denote with \(R_k\) the repunit \(111\dots 1\) made of \(k\) ones.

It is known that every odd prime \(p\) divides the repunit \(R_{p-1}\).

R. Francis & T. Ray call a composite number \(n\) deceptive if it has the same property, i.e., if it divides the repunit \(R_{n-1}\).

For example, \(91=7 \times 13\) is deceptive because it divides \(R_{90}\).

Francis & Ray have proved that there are infinite deceptive numbers since, if \(n\)  is deceptive, then \(R_n\) is deceptive as well.

Every number greater than 2980 can be written as the sum of deceptive numbers.

Here are the deceptive numbers up to 100,001: 

91, 259, 451, 481, 703, 1729, 2821, 2981, 3367, 4141, 4187, 5461, 6533, 6541, 6601, 7471, 7777, 8149, 8401, 8911, 10001, 11111, 12403, 13981, 14701, 14911, 15211, 15841, 19201, 21931, 22321, 24013, 24661, 27613, 29341, 34133, 34441, 35113, 38503, 41041, 45527, 46657, 48433, 50851, 50881, 52633, 54913, 57181, 63139, 63973, 65311, 66991, 67861, 68101, 75361, 79003, 82513, 83119, 94139, 95161, 97273, 97681, 100001, ...

These numbers comprise OEIS A000864:

 A000864: deceptive nonprimes: composite numbers \(k\) that divide the repunit \(R_{k-1}\).

Some Interesting Constants

There was an interesting tweet by Cliff Pickover today that is shown below as Figure 1.

Figure 1

The URL shown in the tweet leads to the Wikipedia article about it. First and formost however, who was Gelfond? As usual the MacTutor site at St.Andrews provides a brief biography.

*****************************

Born: 24th October 1906 in St Petersburg, Russia

Died: 7th November 1968 in Moscow, Russia. 

Summary

Gelfond developed basic techniques in the study of transcendental numbers.

Biography 

Aleksandr Osipovich Gelfond's father was Osip Isaacovich Gelfond who was a physician who also had an interest in philosophy. Gelfond entered Faculty of Physics and Mathematics at Moscow State University in 1924 and completed his undergraduate studies in 1927. He then began research under the supervision of Aleksandr Khinchin and Vyacheslaw Stepanov and completed his postgraduate studies in 1930.

During 1929-30 he taught mathematics at Moscow Technological College but already he had published some important papers: The arithmetic properties of entire functions (1929); Transcendental numbers (1929); and An outline of the history and the present state of the theory of transcendental numbers (1930). The second of these 1929 papers contained the lecture which Gelfond gave to the First All-Union Mathematics Congress held in Kharkov in 1930. These papers by Gelfond represent a major step forward in the study of transcendental numbers. The first of the papers examines the growth of an entire function which assumes integer values for integer arguments. In the second of the 1929 papers Gelfond applied this result to prove that certain numbers are transcendental, so solving a special case of Hilbert's Seventh Problem. We explain some of these ideas below.

In an article he wrote, Gelfond describes the four month visit which he made in 1930 to Germany where he spent time at both Berlin and Göttingen. He was particularly influenced by Hilbert, Siegel and Landau during his visit. After his return to Russia, Gelfond taught mathematics from 1931 at Moscow State University where he held chairs of analysis, theory of numbers and the history of mathematics. From 1933 he also worked in the Mathematical Institute of the Russian Academy of Sciences.

Gelfond developed basic techniques in the study of transcendental numbers, that is numbers that are not the solution of an algebraic equation with rational coefficients. In addition to his important work in the number theory of transcendental numbers, Gelfond made significant contributions to the theory of interpolation and the approximation of functions of a complex variable. He also contributed to the study of differential and integral equations and to the history of mathematics.

In 1934 he proved a special case of his conjecture namely that \(a^x\) is transcendental if \(a\) is algebraic (\(a \neq 0,1 \)) and \(x\) is an irrational, algebraic number. This result is now known as Gelfond's theorem and solved Problem 7 of the list of Hilbert problems. It was solved independently by Schneider. In 1966 Alan Baker proved Gelfond's Conjecture in general. Gelfond's papers in 1933 and 1934, which include his remarkable achievement, are:  

• Gram determinants for stationary series (written jointly with Khinchin) (1933) 

• A necessary and sufficient criterion for the transcendence of a number (1933) 

• Functions that take integer values at the points of a geometric progression (1933) 

• On the seventh problem of D Hilbert (1934) 

Gelfond addressed the Second All-Union Mathematics Congress in Leningrad in 1934) on Transcendental numbers.

 *****************************

The decimal expansion of Gelfond's Constant is:$$e^{\pi} = 23.14069263277926900572 \dots$$The Wikipedia article states that \(e^{\pi} \) also appears in the volumes of hyperspheres. The volume of an \(n\)-sphere with radius \(R\) is given by:$$V_n(R) = \frac{\pi^{n/2} R^n}{\Gamma (n/2+1)}$$where \(R\) is the gamma function. If we take \(R=1\) we get:$$V_n(1) = \frac{\pi^{n/2} }{\Gamma (n/2+1)}$$Any even dimensional \(2n\)-sphere now gives:$$V_{2n}(1) = \frac{\pi^{n} }{\Gamma (n+1)}$$Summing up all even-dimensional unit sphere volumes and utilizing the series expansion of the exponential function gives:$$ V_{2n}(1) = \sum_{0}^{\infty} \frac{\pi^{n}}{n!}=e^{\pi}$$The Wikipedia article also looks at some related functions and the first of them is Ramanujan's Constant:$$ \begin{align} e^{\pi \sqrt{163}} &= 262537412640768743.99999999999925007259 \dots \\ &\approx 640320^3 + 744 \text{ to within one trillionth}\end{align}$$The article goes on to say that:

This is an application of Heegner numbers, where 163 is the Heegner number in question. This number was discovered in 1859 by the mathematician Charles Hermite. In a 1975 April Fool article in Scientific American magazine, "Mathematical Games" columnist Martin Gardner made the hoax claim that the number was in fact an integer, and that the Indian mathematical genius Srinivasa Ramanujan had predicted it—hence its name. Ramanujan's constant is also a transcendental number.

There is also the transcendental constant \(e^{\pi}-\pi \) which approximates to:$$e^{\pi}-\pi =19.99909997918947576726 \dots$$The constant \(i^i\) can be evaluated as follows:$$ \begin{align} i^i &= (e^{i\pi /2})^i\\ &= e^{-\pi/2} \\&=(e^{\pi})^{-1/2}\\ &= 0.20787957635076190854 \dots \end{align}$$As for \( \pi^e \), it is not known whether it is transcendental or not.

Following the link above for Charles Hermite yields the following biography. I've included it in full because I believe it's important to promote awareness of the lives of famous mathematicians to whom we owe so much.

*****************************

Born: 24 December 1822 in Dieuze, Lorraine, France

Died: 14 January 1901 in Paris, France 

Summary

Charles Hermite's work in the theory of functions includes the application of elliptic functions to the quintic equation. He published the first proof that e is a transcendental number.

Biography

Charles Hermite's father was Ferdinand Hermite and his mother was Madeleine Lallemand. Ferdinand Hermite was a trained engineer and he worked in this capacity in a salt mine near Dieuse. After he married Madeleine he joined in the draper's trade in which her family were involved. However he was an artistic man who always wanted to pursue art as a career. He had his wife look after the draper's business and he took up art. Charles was the sixth of his parents seven children and when he was about seven years old his parents left Dieuse and went to live in Nancy to where the business had moved.

Education was not a high priority for Charles's parents but despite not taking too much personal interest in their children's education, nevertheless they did provide them with good schooling. Charles was something of a worry to his parents for he had a defect in his right foot which meant that he moved around only with difficulty. It was clear that this would present him with problems in finding a career. However he had a happy disposition and bore his disability with a cheerful smile.

Charles attended the Collège de Nancy, then went to Paris where he attended the Collège Henri. In 1840-41 he studied at the Collège Louis-le-Grand where some fifteen years earlier Galois had studied. In fact he was taught mathematics there by Louis Richard who had taught Galois. In some ways Hermite was similar to Galois for he preferred to read papers by Euler, Gauss and Lagrange rather than work for his formal examinations.

If Hermite neglected the studies that he should have concentrated on, he was showing remarkable research ability publishing two papers while at Louis-le-Grand. Also like Galois he was attracted by the problem of solving algebraic equations and one of the two papers attempted to show that the quintic cannot be solved in radicals. That he was unfamiliar with Galois's contributions, despite being at the same school, is not at all surprising since the mathematical community were completely unaware of them at this time. However he might reasonably have known of the contributions of Ruffini and Abel to this question, but apparently he did not.

Again like Galois, Hermite wanted to study at the École Polytechnique and he took a year preparing for the examinations. He was tutored by Catalan in 1841-42 and certainly Hermite fared better than Galois had done for he passed. However it was not a glorious pass for he only attained sixty-eighth place in the ordered list. After one year at the École Polytechnique Hermite was refused the right to continue his studies because of his disability. Clearly this was an unfair decision and some important people were prepared to take up his case and fight for him to have the right to continue as a student at the École Polytechnique. The decision was reversed so that he could continue his studies but strict conditions were imposed. Hermite did not find these conditions acceptable and decided that he would not graduate from the École Polytechnique.

Hermite made friends with important mathematicians at this time and frequently visited Joseph Bertrand. On a personal note this was highly significant for he would marry Joseph Bertrand's sister. More significantly from a mathematical point of view he began corresponding with Jacobi and, despite not shining in his formal education, he was already producing research which was ranking as a leading world-class mathematician. The letters he exchanged with Jacobi show that Hermite had discovered some differential equations satisfied by theta-functions and he was using Fourier series to study them. He had found general solutions to the equations in terms of theta-functions. Hermite may have still been an undergraduate but it is likely that his ideas from around 1843 helped Liouville to his important 1844 results which include the result now known as Liouville's theorem.

After spending five years working towards his degree he took and passed the examinations for the baccalauréat and licence which he was awarded in 1847. In the following year he was appointed to the École Polytechnique, the institution which had tried to prevent him continuing his studies some four years earlier; he was appointed répétiteur and admissions examiner.

Hermite made important contributions to number theory and algebra, orthogonal polynomials, and elliptic functions. He discovered his most significant mathematical results over the ten years following his appointment to the École Polytechnique. In 1848 he proved that doubly periodic functions can be represented as quotients of periodic entire functions. In 1849 Hermite submitted a memoir to the Académie des Sciences which applied Cauchy's residue techniques to doubly periodic functions. Sturm and Cauchy gave a good report on this memoir in 1851 but a priority dispute with Liouville seems to have prevented its publication.

Another topic on which Hermite worked and made important contributions was the theory of quadratic forms. This led him to study invariant theory and he found a reciprocity law relating to binary forms. With his understanding of quadratic forms and invariant theory he created a theory of transformations in 1855. His results on this topic provided connections between number theory, theta functions, and the transformations of abelian functions.

On 14 July 1856 Hermite was elected to the Académie des Sciences. However, despite this achievement, 1856 was a bad year for Hermite for he contracted smallpox. It was Cauchy who, with his strong religious conviction, helped Hermite through the crisis. This had a profound effect on Hermite who, under Cauchy's influence, turned to the Roman Catholic religion. Cauchy was also a very staunch royalist and Hermite was influenced by him to also become a royalist. We made comparisons with Galois earlier on in this article, but with royalist views, Hermite was now completely opposed to the views which the staunch republican Galois had held.

The next mathematical result by Hermite which we must mention is one for which he is rightly famous. Although an algebraic equation of the fifth degree cannot be solved in radicals, a result which was proved by Ruffini and Abel, Hermite showed in 1858 that an algebraic equation of the fifth degree could be solved using elliptic functions. He applied these results to number theory, in particular to class number relations of quadratic forms.

In 1862 Hermite was appointed maître de conférence at the École Polytechnique, a position which had been specially created for him. In the following year he became an examiner there. The year 1869 saw him become a professor when he succeeded Duhamel as professor of analysis both at the École Polytechnique and at the Sorbonne. Hermite resigned his chair at the École Polytechnique in 1876 but continued to hold the chair at the Sorbonne until he retired in 1897. In the 1890s Hermite became much less interested in the new results found by the mathematicians of the next generation.

The 1870s saw Hermite return to problems which had interested him earlier in his career such as problems concerning approximation and interpolation. In 1873 Hermite published the first proof that e is a transcendental number. This is another result for which he is rightly famous. Using method's similar to those of Hermite, Lindemann established in 1882 that π was also transcendental. Many historians of science regret that Hermite, despite doing most of the hard work, failed to use it to prove the result on which would have brought him fame outside the world of mathematics. Hermite is now best known for a number of mathematical entities that bear his name: Hermite polynomials, Hermite's differential equation, Hermite's formula of interpolation and Hermitian matrices.

For Hermite certain areas of mathematics were much more interesting than other areas. Hadamard, who unlike his teacher Hermite worked in all areas of mathematics, spoke of Hermite's dislike for geometry:

[Hermite] had a kind of positive hatred of geometry and once curiously reproached me with having made a geometrical memoir.

Hermite's great love was for analysis and, not surprisingly, he had a great respect for Weierstrass. When Mittag-Leffler arrived in Paris to study with him, Hermite greeted him warmly but said:

You have made a mistake, sir, you should follow Weierstrass's course in Berlin. He is the master of us all.

Poincaré is almost certainly the best known of Hermite's students. He once suggested that Hermite's mind did not proceed in logical fashion. He wrote:

But to call Hermite a logician! Nothing can appear to me more contrary to the truth. Methods always seemed to be born in his mind in some mysterious way.

Hadamard like Poincaré was very interested in the way that mathematics was discovered. He also had this to say about the way that Hermite made his discoveries:

Hermite used to observe [that biology] may be a most useful study even for mathematicians, as hidden and eventually fruitful analogies may appear between processes in both kinds of studies.

Hadamard had great respect for Hermite as a teacher. He said:

I do not think that those who never listened to him can realise how magnificent Hermite's teaching was, overflowing with enthusiasm for science, which seemed to come to life in his voice and whose beauty he never failed to communicate to us, since he felt it so much himself to the very depth of his being.

[Hermite] was making a deep impression on us, not only with his methods and those of Weierstrass, but also with his enthusiasm and love of science; in our brief but fruitful conversations, Hermite loved to direct to me remarks such as: "He who strays from the paths traced by providence crashes." These were the words of a profoundly religious man, but an atheist like me understood them very well, especially when he added at other times: "In mathematics, our role is more of servant than of master." It goes without saying that gradually, as years and my scientific work unfolded, I came to understand more and more deeply the aptness and scope of his words.

Cross, reviewing where 125 letters from Hermite to Mittag-Leffler are reproduced, writes:

So there stands revealed one of the most engaging and influential men in Parisian and French mathematics in the second half of the 19th century, one might even say the central character for the period in which he published, 1842-1901. What radiates from the text is [Hermite's] humility, his Catholicism, his concern for his (very extended) family, his willingness to fight for colleagues whose merit he discerns, and his devotion to family, merit, and principle rather than simple influence.

In terms of his family life Hermite had married Louise Bertrand, Joseph Bertrand's sister. One of their two daughters married Émile Picard. Struik writes:

Hermite lived a retired life, with his family. His working hours were devoted to mathematical research and teaching. His outlook on mathematics was realistic in the Platonic sense: a mathematician, like a naturalist, discovers an outside world, in his case a world of ideas. Hermite, therefore, disliked Cantor's world, in which a new mathematical world was created.

***************************** 

Xenodrome Probabilites

I posted about xenodromes in an eponymous post on the 20th October 2024. In this current post, I want to look at the probabilities of xenodromes occurring in the various number bases. Let's start with base 16 where we have 16 digits to choose from:

0, 1, 2, 3, 4, 5, 6, 7, 8, 9, a, b, c, d, e, f

An example of a five digit xenodrome in base 16 would be 56a3f where:$$56a3f_{16}=354879_{10}$$What I found was the following (permalink):

There are 491400 permutations of 5 digits in base 16 so that all are distinct with no leading zero. There are a total of 983040 possible permutations of 5 digits in base 16 with no leading zero. Probability of a 5 digit xenodrome in base 16 is thus 0.4999 or 50%.

The probability is thus effectively 50%. However, the five digit decimal numbers that I deal with when investigating the numbers associated with my diurnal age (10000 to 40000) require only four hexadecimal digits for their representation and so:

There are 40950 permutations of 4 digits in base 16 so that all are distinct with no leading zero. There are a total of 61440 possible permutations of 4 digits in base 16 with no leading zero. Probability of a 4 digit xenodrome in base 16 is thus 0.6665 or 67%.

So 5 digit base 10 numbers in the range between 10000 and 40000 will have a 2/3 chance of being xenodromic in base 16. 

Let's look at the other bases now. Here's the results for base 15 and 5 digits:

There are 336336 permutations of 5 digits in base 15 so that all are distinct with no leading zero. There are a total of 708750 possible permutations of 5 digits in base 15 with no leading zero. Probability of a 5 digit xenodrome in base 15 is thus 0.4745 or 47%.

 For base 14 and 5 digits:

There are 223080 permutations of 5 digits in base 14 so that all are distinct with no leading zero. There are a total of 499408 possible permutations of 5 digits in base 14 with no leading zero. Probability of a 5 digit xenodrome in base 14 is thus 0.4467 or 45%.

For base 13 and 5 digits:

There are 142560 permutations of 5 digits in base 13 so that all are distinct with no leading zero. There are a total of 342732 possible permutations of 5 digits in base 13 with no leading zero. Probability of a 5 digit xenodrome in base 13 is thus 0.4160 or 42%.

For base 12 and 5 digits: 

There are 87120 permutations of 5 digits in base 12 so that all are distinct with no leading zero There are a total of 228096 possible permutations of 5 digits in base 12 with no leading zero Probability of a 5 digit xenodrome in base 12 is thus 0.3819 or 38%.

For base 11 and 5 digits: 

There are 50400 permutations of 5 digits in base 11 so that all are distinct with no leading zero There are a total of 146410 possible permutations of 5 digits in base 11 with no leading zero Probability of a 5 digit xenodrome in base 11 is thus 0.3442 or 34%.

For base 10 and 5 digits:

There are 27216 permutations of 5 digits in base 10 so that all are distinct with no leading zero There are a total of 90000 possible permutations of 5 digits in base 10 with no leading zero Probability of a 5 digit xenodrome in base 10 is thus 0.3024 or 30%.

So the probability of a five digit xenodrome in base 10 is 30%. We may as well continue for the lower bases.

For base 9 and 5 digits:

There are 13440 permutations of 5 digits in base 9 so that all are distinct with no leading zero. There are a total of 52488 possible permutations of 5 digits in base 9 with no leading zero. Probability of a 5 digit xenodrome in base 9 is thus 0.2561 or 26%.

For base 8 and 5 digits: 

 There are 5880 permutations of 5 digits in base 8 so that all are distinct with no leading zero. There are a total of 28672 possible permutations of 5 digits in base 8 with no leading zero. Probability of a 5 digit xenodrome in base 8 is thus 0.2051 or 21%.

For base 7 and 5 digits:

There are 2160 permutations of 5 digits in base 7 so that all are distinct with no leading zero. There are a total of 14406 possible permutations of 5 digits in base 7 with no leading zero. Probability of a 5 digit xenodrome in base 7 is thus 0.1499 or 15%. 

For base 6 and 5 digits:

There are 600 permutations of 5 digits in base 6 so that all are distinct with no leading zero. There are a total of 6480 possible permutations of 5 digits in base 6 with no leading zero. Probability of a 5 digit xenodrome in base 6 is thus 0.09259 or 9%. 

For base 5 and 5 digits:

There are 96 permutations of 5 digits in base 5 so that all are distinct with no leading zero. There are a total of 2500 possible permutations of 5 digits in base 5 with no leading zero. Probability of a 5 digit xenodrome in base 5 is thus 0.03840 or 4%.

Of course, there are no 5 digit xenodromes in base 4 so we need to switch to four digit numbers. So for base 4 and 4 digits:

There are 18 permutations of 4 digits in base 4 so that all are distinct with no leading zero. There are a total of 192 possible permutations of 4 digits in base 4 with no leading zero. Probability of a 4 digit xenodrome in base 4 is thus 0.09375 or 9%.

Similarly there no 4 digit xenodromes in base 3 so we need to switch to three digits. For base 3 and 3 digits:

There are 4 permutations of 3 digits in base 3 so that all are distinct with no leading zero. There are a total of 18 possible permutations of 3 digits in base 3 with no leading zero. Probability of a 3 digit xenodrome in base 3 is thus 0.2222 or 22%.

Finally, for base 2 and 2 digits we have:

There is 1 permutation of 2 digits in base 2 so that all are distinct with no leading zero. There are a total of 2 possible permutations of 2 digits in base 2 with no leading zero. Probability of a 2 digit xenodrome in base 2 is thus 0.5000 or 50%.

Here is the permalink again to carry out these calculations. Just adjust for the number of digits and the base. 

Consolidating Fibonacci-like Numbers

In my post titled Additive Fibonacci-like Numbers I was dealing with additive digital roots to generate additional digits after the starting two digits were in place. For example, let's start with 78:$$ \begin{align} 78 \rightarrow 7 + 8 =15 \rightarrow 1+5=6 &\rightarrow 786 \\786 \rightarrow 8+6=14 \rightarrow 1+4=5 &\rightarrow 7865 \\7865 \rightarrow 6+5 =11 \rightarrow 1+1=2 &\rightarrow 78652 \end{align} $$We could keep going forever. The advantage of this approach is that the sum of the two previous digits reduces to a single digit between 1 and 9. Let's call these types of numbers Additive Fibonacci-like Numbers of the First Type. Between 100 and 1,000,000 these numbers are:

101, 112, 123, 134, 145, 156, 167, 178, 189, 191, 202, 213, 224, 235, 246, 257, 268, 279, 281, 292, 303, 314, 325, 336, 347, 358, 369, 371, 382, 393, 404, 415, 426, 437, 448, 459, 461, 472, 483, 494, 505, 516, 527, 538, 549, 551, 562, 573, 584, 595, 606, 617, 628, 639, 641, 652, 663, 674, 685, 696, 707, 718, 729, 731, 742, 753, 764, 775, 786, 797, 808, 819, 821, 832, 843, 854, 865, 876, 887, 898, 909, 911, 922, 933, 944, 955, 966, 977, 988, 999, 1011, 1123, 1235, 1347, 1459, 1562, 1674, 1786, 1898, 1911, 2022, 2134, 2246, 2358, 2461, 2573, 2685, 2797, 2819, 2922, 3033, 3145, 3257, 3369, 3472, 3584, 3696, 3718, 3821, 3933, 4044, 4156, 4268, 4371, 4483, 4595, 4617, 4729, 4832, 4944, 5055, 5167, 5279, 5382, 5494, 5516, 5628, 5731, 5843, 5955, 6066, 6178, 6281, 6393, 6415, 6527, 6639, 6742, 6854, 6966, 7077, 7189, 7292, 7314, 7426, 7538, 7641, 7753, 7865, 7977, 8088, 8191, 8213, 8325, 8437, 8549, 8652, 8764, 8876, 8988, 9099, 9112, 9224, 9336, 9448, 9551, 9663, 9775, 9887, 9999, 10112, 11235, 12358, 13472, 14595, 15628, 16742, 17865, 18988, 19112, 20224, 21347, 22461, 23584, 24617, 25731, 26854, 27977, 28191, 29224, 30336, 31459, 32573, 33696, 34729, 35843, 36966, 37189, 38213, 39336, 40448, 41562, 42685, 43718, 44832, 45955, 46178, 47292, 48325, 49448, 50551, 51674, 52797, 53821, 54944, 55167, 56281, 57314, 58437, 59551, 60663, 61786, 62819, 63933, 64156, 65279, 66393, 67426, 68549, 69663, 70775, 71898, 72922, 73145, 74268, 75382, 76415, 77538, 78652, 79775, 80887, 81911, 82134, 83257, 84371, 85494, 86527, 87641, 88764, 89887, 90999, 91123, 92246, 93369, 94483, 95516, 96639, 97753, 98876, 99999

However, in my previous post, Variations on the Taxi Cab Number, I was not working with the digital roots and this is a severe limitation. The early digits need to be small if the digits are to progress in a Fibonacci-like manner. That's why, in the range of numbers, up to one million, the largest number is 303369. This number is constructed as follows beginning with the first two digits 3 and 0:$$ \begin{align} 30 \rightarrow 3 + 0 &= 3 \rightarrow 303 \\ 303 \rightarrow 0+3 &=3 \rightarrow 3033\\3033 \rightarrow 3 + 3 &= 6 \rightarrow 30336\\30336 \rightarrow 3+6 &= 9 \rightarrow 303369 \end{align}$$We can't go any further because of the final two digits: 6 + 9 = 15. Let's call these types of numbers Additive Fibonacci-like Numbers of the Second Type. Between 100 and 1,000,000 these numbers are:

101, 112, 123, 134, 145, 156, 167, 178, 189, 202, 213, 224, 235, 246, 257, 268, 279, 303, 314, 325, 336, 347, 358, 369, 404, 415, 426, 437, 448, 459, 505, 516, 527, 538, 549, 606, 617, 628, 639, 707, 718, 729, 808, 819, 909, 1011, 1123, 1235, 1347, 1459, 2022, 2134, 2246, 2358, 3033, 3145, 3257, 3369, 4044, 4156, 4268, 5055, 5167, 5279, 6066, 6178, 7077, 7189, 8088, 9099, 10112, 11235, 12358, 20224, 21347, 30336, 31459, 40448, 101123, 112358, 202246, 303369

With bases higher than 10, the 1 to 9 digit limitation can be exceeded. For example in base 16, if we start as before with an initial 78 then a third digit is possible:$$78 \rightarrow 7 + 8 = 15 = F \rightarrow 78F$$Thus we have:$$ \begin{align} 78F_{16} &= 7 \times 16^2 + 8 \times 16 + 15 \\ &=1935_{10} \end{align} $$This means that 1935 is an Additive Fibonacci-like Number of the Second Type in base 16. Here is a list of numbers greater than 27000 and less than 40000 that are "additive Fibonacci-like" and of the "second type" in base 16 (permalink):

• 28791 --> 7077
• 29065 --> 7189
• 29339 --> 729b
• 29613 --> 73ad
• 29887 --> 74bf
• 32904 --> 8088
• 33178 --> 819a
• 33452 --> 82ac
• 33726 --> 83be
• 37017 --> 9099
• 37291 --> 91ab
• 37565 --> 92bd
• 37839 --> 93cf

Additive Fibonacci-like Numbers of the Second Type in base 16 are thus:

28791, 29065, 29339, 29613, 29887, 32904, 33178, 33452, 33726, 37017, 37291, 37565, 37839

*****************************

Here the numbers greater than 27000 and less than 40000 for base 15:

• 27128 --> 8088
• 27370 --> 819a
• 27612 --> 82ac
• 27854 --> 83be
• 30519 --> 9099
• 30761 --> 91ab
• 31003 --> 92bd
• 33910 --> a0aa
• 34152 --> a1bc
• 34394 --> a2ce
• 37301 --> b0bb
• 37543 --> b1cd

Additive Fibonacci-like Numbers of the Second Type in base 15 are (permalink):

27128, 27370, 27612, 27854, 30519, 30761, 31003, 33910, 34152, 34394, 37301, 37543

*****************************

Here are the numbers greater than 27000 and less than 40000 for base 14 (permalink): 

• 27590 --> a0aa
• 27802 --> a1bc
• 30349 --> b0bb
• 30561 --> b1cd
• 33108 --> c0cc
• 35867 --> d0dd
• 38628 --> 10112

Additive Fibonacci-like Numbers of the Second Type in base 14 are thus:

27590, 27802, 30349, 30561, 33108, 35867, 38628

*****************************

Here are the numbers greater than 27000 and less than 40000 for base 13 (permalink):

• 28745 --> 10112
• 31140 --> 11235
• 33535 --> 12358
• 35930 --> 1347b

Additive Fibonacci-like Numbers of the Second Type in base 13 are thus:

28745, 31140, 33535, 35930

*****************************

For base 12, there are none between 27000 and 40000 but for base 11 we have (permalink):

• 29550 --> 20224
• 31027 --> 21347
• 32504 --> 2246a

Additive Fibonacci-like Numbers of the Second Type in base 11 are thus:

29550, 31027, 32504

*****************************

Just for completeness I'll now look at bases 10 and lower. For base 10, we have (permalink):

• 30336 --> 30336
• 31459 --> 31459

Additive Fibonacci-like Numbers of the Second Type in base 10 are thus:

30336, 31459

*****************************

For base 9 there are none but for base 8 there is one (permalink):

• 33363 --> 101123

Additive Fibonacci-like Numbers of the Second Type in base 8 are thus :

33363

*****************************

For base 7, we have 

• 34432 --> 202246

Additive Fibonacci-like Numbers of the Second Type in base 7 are thus:

34432

There are no suitable numbers in the range 27000 to 40000 for bases 2, 3, 4, 5 and 6. I've added this determination of whether a number is additive Fibonacci-like of the second type to my multipurpose algorithm.

RIGHT TO LEFT INSTEAD OF LEFT TO RIGHT

There's no compulsion to proceed from left to right when working with digits and so a new set of numbers can be generated by simply reversing the order of the digits. Thus Additive Fibonacci-like Numbers of the First Type are shown below where digit progression is from right to left:

101, 119, 128, 137, 146, 155, 164, 173, 182, 191, 202, 211, 229, 238, 247, 256, 265, 274, 283, 292, 303, 312, 321, 339, 348, 357, 366, 375, 384, 393, 404, 413, 422, 431, 449, 458, 467, 476, 485, 494, 505, 514, 523, 532, 541, 559, 568, 577, 586, 595, 606, 615, 624, 633, 642, 651, 669, 678, 687, 696, 707, 716, 725, 734, 743, 752, 761, 779, 788, 797, 808, 817, 826, 835, 844, 853, 862, 871, 889, 898, 909, 918, 927, 936, 945, 954, 963, 972, 981, 999, 1101, 1191, 1283, 1375, 1467, 1559, 1642, 1734, 1826, 1918, 2119, 2202, 2292, 2384, 2476, 2568, 2651, 2743, 2835, 2927, 3128, 3211, 3303, 3393, 3485, 3577, 3669, 3752, 3844, 3936, 4137, 4229, 4312, 4404, 4494, 4586, 4678, 4761, 4853, 4945, 5146, 5238, 5321, 5413, 5505, 5595, 5687, 5779, 5862, 5954, 6155, 6247, 6339, 6422, 6514, 6606, 6696, 6788, 6871, 6963, 7164, 7256, 7348, 7431, 7523, 7615, 7707, 7797, 7889, 7972, 8173, 8265, 8357, 8449, 8532, 8624, 8716, 8808, 8898, 8981, 9182, 9274, 9366, 9458, 9541, 9633, 9725, 9817, 9909, 9999, 11918, 12835, 13752, 14678, 15505, 15595, 16422, 17348, 18265, 19182, 21101, 21191, 22927, 23844, 24761, 25687, 26514, 27431, 28357, 29274, 31283, 32119, 33936, 34853, 35779, 36606, 36696, 37523, 38449, 39366, 41375, 42202, 42292, 43128, 44945, 45862, 46788, 47615, 48532, 49458, 51467, 52384, 53211, 54137, 55954, 56871, 57707, 57797, 58624, 59541, 61559, 62476, 63303, 63393, 64229, 65146, 66963, 67889, 68716, 69633, 71642, 72568, 73485, 74312, 75238, 76155, 77972, 78808, 78898, 79725, 81734, 82651, 83577, 84404, 84494, 85321, 86247, 87164, 88981, 89817, 91826, 92743, 93669, 94586, 95413, 96339, 97256, 98173, 99909, 99999

Similarly Additive Fibonacci-like Numbers of the Second Type are shown below where digit progression is from right to left:

101, 202, 211, 303, 312, 321, 404, 413, 422, 431, 505, 514, 523, 532, 541, 606, 615, 624, 633, 642, 651, 707, 716, 725, 734, 743, 752, 761, 808, 817, 826, 835, 844, 853, 862, 871, 909, 918, 927, 936, 945, 954, 963, 972, 981, 1101, 2202, 3211, 3303, 4312, 4404, 5321, 5413, 5505, 6422, 6514, 6606, 7431, 7523, 7615, 7707, 8532, 8624, 8716, 8808, 9541, 9633, 9725, 9817, 9909, 21101, 42202, 53211, 63303, 74312, 84404, 85321, 95413, 321101, 642202, 853211, 963303

If we proceed from left to right, then the third digit is the difference between the first and second digits and so on (this is the subtraction sequence mentioned in my previous post).

A Variation on the Taxi Cab Number

The number 1729 is famous as the so-called "taxi cab number" in memory of the interchange between the mathematicians Hardy and Ramanujan in which the latter observed that the number of the taxi cab in which the former had arrived at the hospital was far from boring (as Hardy had thought). Instead 1729 is the first positive integer that is the sum of two positive cubes in two different ways:$$ \begin{align} 1729 &= 1^3+12^3\\ &= 9^3+10^3 \end{align} $$Today I observed a taxi with the number plate T 3257 and noted that there is a touch of Fibonacci about its digits because:$$ 3 +2 = 5 \text{ and } 2 + 5 = 7$$The digits thus form a Fibonacci-type sequence:$$ 3, 2, 5, 7$$This got me thinking about what numbers with three or more digit have this Fibonacci-like property. Well, up to one million, there are only 82 such numbers so they form a rather exclusive club. 

Here they are (permalink) in a sequence that we'll called the ADDITION SEQUENCE:

101, 112, 123, 134, 145, 156, 167, 178, 189, 202, 213, 224, 235, 246, 257, 268, 279, 303, 314, 325, 336, 347, 358, 369, 404, 415, 426, 437, 448, 459, 505, 516, 527, 538, 549, 606, 617, 628, 639, 707, 718, 729, 808, 819, 909, 1011, 1123, 1235, 1347, 1459, 2022, 2134, 2246, 2358, 3033, 3145, 3257, 3369, 4044, 4156, 4268, 5055, 5167, 5279, 6066, 6178, 7077, 7189, 8088, 9099, 10112, 11235, 12358, 20224, 21347, 30336, 31459, 40448, 101123, 112358, 202246, 303369

I have made a related post titled Additive Fibonacci-like Numbers on the 7th August 2024 but this involved finding the digital roots of numbers unlike what I've done here. So for me 3257 will remain my personal taxi cab number.

Another sequence will emerge if, instead of adding the second number to the first and so on, we SUBTRACT the second from the first and so on. In this scenario, 3211 would satisfy because:$$  3 - 2 = 1 \text{ and } 2 -1 =1$$Up to one million, there are 99 such numbers of three digits or more. Here they are in a sequence we'll call the SUBTRACTION SEQUENCE (permalink):

101, 110, 202, 211, 220, 303, 312, 321, 330, 404, 413, 422, 431, 440, 505, 514, 523, 532, 541, 550, 606, 615, 624, 633, 642, 651, 660, 707, 716, 725, 734, 743, 752, 761, 770, 808, 817, 826, 835, 844, 853, 862, 871, 880, 909, 918, 927, 936, 945, 954, 963, 972, 981, 990, 1101, 2110, 2202, 3211, 3303, 4220, 4312, 4404, 5321, 5413, 5505, 6330, 6422, 6514, 6606, 7431, 7523, 7615, 7707, 8440, 8532, 8624, 8716, 8808, 9541, 9633, 9725, 9817, 9909, 21101, 32110, 42202, 53211, 63303, 64220, 74312, 84404, 85321, 95413, 96330, 321101, 532110, 642202, 853211, 963303

It can be noted that some numbers containing zero feature in both sequences. These numbers are 101, 202, 303, 404, 505, 606, 707, 808 and 909.

While we're at it why not consider multiplication in which the first two digits multiply together to give the third digit and so on. In the range up to one million, there are 78 such numbers and here they are in a sequence we'll call the MULTIPLICATION SEQUENCE (permalink):

100, 111, 122, 133, 144, 155, 166, 177, 188, 199, 200, 212, 224, 236, 248, 300, 313, 326, 339, 400, 414, 428, 500, 515, 600, 616, 700, 717, 800, 818, 900, 919, 1000, 1111, 1224, 1339, 2000, 2122, 2248, 3000, 3133, 4000, 4144, 5000, 5155, 6000, 6166, 7000, 7177, 8000, 8188, 9000, 9199, 10000, 11111, 12248, 20000, 21224, 30000, 31339, 40000, 50000, 60000, 70000, 80000, 90000, 100000, 111111, 200000, 212248, 300000, 400000, 500000, 600000, 700000, 800000, 900000, 1000000

An example is 212248 where we have:$$2 \times 1 = 2, \, 2 \times 1 = 2, \, 2 \times 2 = 4 \text{ and } 4 \times 2 = 8$$The zeros of course make some of these numbers a little trivial and so with the digit 0 excluded we have 41 suitable numbers (permalink) in a sequence we'll call the MULTIPLICATION WITHOUT ZERO SEQUENCE:

111, 122, 133, 144, 155, 166, 177, 188, 199, 212, 224, 236, 248, 313, 326, 339, 414, 428, 515, 616, 717, 818, 919, 1111, 1224, 1339, 2122, 2248, 3133, 4144, 5155, 6166, 7177, 8188, 9199, 11111, 12248, 21224, 31339, 111111, 212248

If we consider dividing the second digit into the first to give the third digit and so on then, excluding numbers with zero, we have the following numbers (permalink) in what we'll call the DIVISION WITHOUT ZERO SEQUENCE:

111, 212, 221, 313, 331, 414, 422, 441, 515, 551, 616, 623, 632, 661, 717, 771, 818, 824, 842, 881, 919, 933, 991, 1111, 2212, 3313, 4221, 4414, 5515, 6616, 7717, 8422, 8818, 9331, 9919, 11111, 42212, 84221, 93313, 111111, 842212

Many of the numbers in the division sequence are not surprisingly the reverse of numbers in the multiplication sequence e.g. 842212 in the division sequence is the reverse of 212248 in the multiplication sequence.

An Interesting Sequence

Here's an interesting question that was posed by PUZZLE A DAY on 30th October 2024: What number comes next in this sequence and why? 1, 7, 11, 27, ...

Mathematically, there doesn't seem to be any pattern and I needed the clue provided to proceed any further. Here is the clue:

Clue: Say the number aloud.

Immediately we know that the sequence is about how the number sounds and not about anything to do with the numbers per se. When we sound the numbers out, we have:

• one ... single syllable
• seven ... two syllables
• eleven ... three syllables
• twenty seven ... four syllables

Clearly we are looking for the next number that has five syllables and that numbers happens to be 77 or seventy seven. So we now have 1, 7, 11, 27, 77, ...

That's where the puzzle ends but the question then arises as to the next number in the sequence. This will take us into the hundreds and we need to decide whether to sound out the "and" or not. For example, 101 can be read as "one hundred and one" or "one hundred one". Let's go with the latter and search for the next number. 

As far as I can determine this must be 107 or one hundred seven that has five syllables. After that it would be 111 or one hundred eleven with six syllables. So now we have: 1, 7, 11, 27, 77, 107, 111. 

After this the next number with seven syllables is 127 or one hundred twenty seven. For eight syllables, it would 177 or one hundred seventy seven. Let's stop there and then list what we have so far is 1, 7, 11, 27, 77, 107, 111, 127, 177, ...

The puzzle's focus on the sound of the number rather than the number itself reminds of aban, eban, iban, oban and uban numbers. See my blog post Iban Numbers. It's also reminscent of the Look and Say Sequence. Finally, the Mathematics section of PUZZLE A DAY can be found here.

Multipurpose Algorithm using SageMath

Given the recent problems with Numbers Aplenty (see blog post Trouble with Numbers Aplenty), I've continued developing a SageMath program that will generate information about a number in a manner similar to Numbers Aplenty. At the moment, for any number that is input, the following output is generated:

• factorisation of the number or declaration that it is prime
• number of divisors
• list of divisors
• sum of divisors and factorisation of the sum
• sum of proper divisors and factorisation of the sum
• determination of whether a number is deficient, perfect or abundant
• if abundant, determination of whether the number is pseudoperfect or weird
• if abundant, determination of whether the number is a Zumkeller number
• if abundant, determination of whether the number is practical
• determination of whether number is Duffinian
• totient and its factorisation
• cototient and its factorisation
• determination of whether number is cyclic
• sum of squares of digits and factorisation of the sum
• sum of cubes of digits and factorisation of the sum
• determination of whether the number is admirable
• determination of whether the number is an interprime number
• determination of whether the number is a Bogota number
• determination of whether the number is a Curzon number
• determination of whether the number is a plaindrome in any base from 2 to 16
• determination of whether the number is metadrome in any base from 2 to 16
• determination of whether the number is a nialpdrome in any base from 2 to 16
• determination of whether the number is a katadrome in any base from 2 to 16
• determination of whether the number is a xenodrome in any base from 2 to 16
• determination of whether the number is an additive Fibonacci-like number of the second type in any base from 2 to 16
• Gray Code equivalent of the number and its factorisation
• absolute difference between the number and its Gray Code and factorisation of this difference
• determination of whether the number is an Ulam number and, if so, display the two numbers that add together to form it
• binary complement of the number and its factorisation
• absolute difference between the number and its binary complement and factorisation of this difference
• determination of whether the number is energetic
• number expressed as a sum of two squares if possible
• number expressed as a sum of two cubes if possible
• sum of digits (SOD) and product of digits (POD)
• number + SOD and factorisation of this sum
• number - SOD and factorisation of this sum
• number + POD and factorisation of this sum
• number - POD and factorisation of this sum
• sequence generated by repetition of Number + SOD - POD
• sequence generated by repetition of Number + SOD - POD but with zeros ignored
• Collatz trajectory of the number
• aliquot sequence determined up to a maximum of 101 steps
• antidivisors of the number
• arithmetic derivative of the number
• determination as to whether number is a D-number
• determination as to whether number is a de Polignac number
• sequence produced by Maximum-Minimum recursive algorithm
• Minimum Goldbach Decomposition
• sequence generated by Reverse and Add 
• Home Prime calculated to a maximum of 50 steps
• Additive Persistence
• Multiplicative Persistence taking zeros into account
• Multiplicative Persistence NOT taking zeros into account
• sequence generated by repetition of Number + Sum of Odd Digits - Sum of Even Digits
• Circulant Matrix
• Determinant of Circulant Matrix and its factorisation
• absolute difference between the numbers and the determinant and factorisation of this difference

I'll keep adding to this program over time using my Jupyter notebook. The program will run in SageMathCell although the creation of a Permalink in not possible. Here's a link to a Google Doc bookmark however:

https://docs.google.com/document/d/1xwJ2A0dtWN_i3e3ns2YJzmOGyeCaQ-COIwBBS74LFFQ/edit?tab=t.0#bookmark=id.o38tcw5ek6w7

The code located at this bookmark can be copied into SageMathCell and run.

POSTSCRIPT: 31st October 2024

The day after I made this post the Numbers A Plenty website was down again so all the more reason to keep refining my SageMath algorithm.

Number + SOD - POD Sequences

In this post, I will be using SOD as meaning the Sum Of Digits of a number and POD as meaning the Product Of Digits of a number. Consider a number like 27602 that is the number associated with my diurnal age today. What happens if I repeatedly add the sum of its digits and subtract the product of its digits? Because it contains the digit zero, its POD is initially 0 but this soon changes as I repeat the process. Here is the 27602's trajectory that consists of 26 steps (permalink):

276002, 276019, 276044, 276067, 276095, 276124, 275474, 267663, 258621, 257685, 240918, 240942, 240963, 240987, 241017, 241032, 241044, 241059, 241080, 241095, 241116, 241083, 241101, 241110, 241119, 241065, 241083

As can be seen, a loop is reached. What about 27610? Its trajectory consists of 27 steps (permalink):

27610, 27626, 26641, 26372, 25888, 20799, 20826, 20844, 20862, 20880, 20898, 20925, 20943, 20961, 20979, 21006, 21015, 21024, 21033, 21042, 21051, 21060, 21069, 21087, 21105, 21114, 21115

There is no loop here so why does the sequence terminate with 21115. The answer is that this number has a SOD (2 + 1 + 1 + 1 + 5 = 10) equal to its POD (2 x 1 x 1 x 1 x 5 = 10). So for every number, the corresponding sequence will have a finite number of steps ending in either a loop or a number with SOD = POD. 

What's of interest now are the progressive record lengths of the sequences as we consider larger and larger numbers. Table 1 shows the record lengths up to 110,000. 

Around 10,000 it can be seen that there is a big jump from 39 (9541) to 85 (9980) and again aroud 100,000 there is an even bigger jump from 98 (98907) to 631 (99970). If I were to continue there would be another big jump around one million. The sequence of numbers with record lengths is:

1, 10, 400, 417, 432, 482, 730, 2200, 2217, 2232, 2282, 2800, 2903, 3610, 4601, 5177, 6821, 7248, 9380, 9541, 9980, 9990, 10002, 98907, 99970, 99980, 99990, 100008

This sequence is not listed in the OEIS and I certainly won't be proposing it for inclusion. Variations on this theme are of course possible. For example, the product of digits could exclude the digit 0. This digit makes no difference to the sum of a number's digits but it has the effect of always sending the product of its digits to zero.

This makes a big difference to the sequence of record lengths and we no longer get those big jumps around 10,000, 100,000 and 1,000,000. Firstly though, let's go back to the original number 27601 and see how its trajectory is affected. The new trajectory is:

27601, 27533, 26923, 26297, 24811, 24763, 23777, 21745, 21484, 21247, 21151

The initial product of digits is now not zero but 2 x 7 x 6 x 1 = 84. The sum of its digits is still 16 but the SOD - POD now becomes 16 - 84 = -68 and so the next number in the sequence is 27533. The sequence ends with 21151 which has its SOD = POD.

The record lengths are shown in Table 2.

Table 2

The sequence of numbers with record lengths with zero not counting in the POD is now as follows:

1, 11, 100, 810, 1918, 1931, 2614, 2831, 2905, 3149, 3226, 3638, 3943, 4116, 4228, 4290, 4543, 6242, 7504, 7600, 7730, 8152, 8405, 9714, 19911, 23191, 23318, 23470, 25364, 27001, 29270, 44573, 45552, 46680, 47163, 47730, 49434, 54181, 54641, 55418, 56100, 57135, 71620, 73191, 73302, 74620, 75210, 78543, 81200, 82452, 83292, 87294, 88803, 95900, 97630, 98180, 100000, 100002

On the subject of numbers from 10 up to 100,000 in which SOD = POD, it is only these combination of digits that satisfy (permalink):

[2, 2], [1, 2, 3], [1, 1, 2, 4], [1, 1, 1, 2, 5], [1, 1, 1, 3, 3], [1, 1, 2, 2, 2]

There are 68 numbers in the range from 1 to 100,000 that satisfy SOD = POD and they are:

1, 2, 3, 4, 5, 6, 7, 8, 9, 22, 123, 132, 213, 231, 312, 321, 1124, 1142, 1214, 1241, 1412, 1421, 2114, 2141, 2411, 4112, 4121, 4211, 11125, 11133, 11152, 11215, 11222, 11251, 11313, 11331, 11512, 11521, 12115, 12122, 12151, 12212, 12221, 12511, 13113, 13131, 13311, 15112, 15121, 15211, 21115, 21122, 21151, 21212, 21221, 21511, 22112, 22121, 22211, 25111, 31113, 31131, 31311, 33111, 51112, 51121, 51211, 52111

However, if we don't include zero in the product of digits then there are 164 numbers that qualify because we can insert zeros into all the above numbers with impunity. Of course this means that some numbers will exceed 100,000 and so will not be included in the list below (permalink).

1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 20, 22, 30, 40, 50, 60, 70, 80, 90, 123, 132, 202, 213, 220, 231, 312, 321, 1023, 1032, 1124, 1142, 1203, 1214, 1230, 1241, 1302, 1320, 1412, 1421, 2013, 2020, 2031, 2103, 2114, 2130, 2141, 2301, 2310, 2411, 3012, 3021, 3102, 3120, 3201, 3210, 4112, 4121, 4211, 10124, 10142, 10203, 10214, 10230, 10241, 10302, 10320, 10412, 10421, 11024, 11042, 11125, 11133, 11152, 11204, 11215, 11222, 11240, 11251, 11313, 11331, 11402, 11420, 11512, 11521, 12014, 12030, 12041, 12104, 12115, 12122, 12140, 12151, 12212, 12221, 12401, 12410, 12511, 13020, 13113, 13131, 13311, 14012, 14021, 14102, 14120, 14201, 14210, 15112, 15121, 15211, 20103, 20114, 20130, 20141, 20301, 20310, 20411, 21014, 21030, 21041, 21104, 21115, 21122, 21140, 21151, 21212, 21221, 21401, 21410, 21511, 22112, 22121, 22211, 23010, 24011, 24101, 24110, 25111, 30102, 30120, 30201, 30210, 31020, 31113, 31131, 31311, 32010, 33111, 40112, 40121, 40211, 41012, 41021, 41102, 41120, 41201, 41210, 42011, 42101, 42110, 51112, 51121, 51211, 52111

If we try reversing the order of the subtraction so that we have number + POD - SOD, the POD soon causes the sequences for many numbers to increase without bound. In the case of POD with zeros counted, the number 23 was the first to exceed a sequence length of 700. When zeros were not counted, 516 was the first number to exceed 700.