Conjectures on Sums of Consecutive Primes
Edwige Tolla · 2026
Read Conjectures on Sums of Consecutive Primes by Edwige Tolla (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2601-15346.txt.
Introduction
Edwige Tolla states conjectures about what happens when consecutive primes are added up. Her 2026 arXiv paper deals with the additive properties of the prime numbers. Start at any prime and add the next ones in a row, for an odd count of terms.
Those block sums have their own patterns. Tolla studies when the sums are themselves prime. She also studies the broader additive behavior of such blocks. The paper sets out conjectures, claims that fit the computed cases but await proof. Number theorists can test and attack them in turn.
Catalog record
| CREATOR | Edwige Tolla |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 21 KB |
| VERSION | 1.0 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2601-15346
Browse the full library · About the Tower of Babel
Full text
The opening of the work, from the library's hosted text:
Conjectures on Sums of Consecutive Primes Edwige Tolla Abstract We study additive properties of consecutive prime numbers and the primality of the sums they generate. For a given prime number p n p_{n} , we consider the sums S k ( p n ) = p n + p n + 1 + ⋯ + p n + k − 1 , S_{k}(p_{n})=p_{n}+p_{n+1}+\cdots+p_{n+k-1}, where k ≥ 3 k\geq 3 is an odd integer. We first formulate an existence conjecture asserting that, for every prime number p n p_{n} , there exists at least one odd length k ≥ 3 k\geq 3 such that S k ( p n ) S_{k}(p_{n}) is itself a prime number. An exhaustive computational verification covering the first one million prime numbers revealed no counterexamples. We then propose a strengthened conjecture according to which, for every prime number p n p_{n} , there exist infinitely many odd lengths k k such that S k ( p n ) S_{k}(p_{n}) is prime. This strong version is supported by a probabilistic heuristic showing that the series of the corresponding primality probabilities diverges, suggesting that the phenomenon is not exceptional but recurrent. We also analyze the possible modular obstructions, showing that they are local in nature and cannot persist when the length k k varies among odd integers. A Diophantine interpretation of the problem is proposed, together with a conceptual comparison with the generalized Goldbach conjecture. Finally, we discuss the role of the Generalized Riemann Hypothesis (GRH) in controlling the distribution of the sums under consideration. These structural, modular, Diophantine, and probabilistic (heuristic) arguments support both conjectures and formalize heuristic theorems of Cramér, GRH, and Hardy–Littlewood type explaining the expected absence of counterexamples. 1 Existence Conjecture 1.1 Introduction Additive properties of prime numbers constitute a central theme in number theory, with classical results such as the work of Hardy and Littlewood on additive problems [ 6 , 7 ] , as well as advances related to Goldbach’s conjecture [ 8 , 9 ] and arithmetic progressions in the primes [ 10 ] . In this context, we focus on a specific additive structure: sums of consecutive primes that are themselves prime. Let p_{n}\in{\mathscr{P}} denote the n -th prime, and let (p_{n})_{n\geq 1} be the increasing sequence of consecutive primes. For an integer k\geq 1 , consider S_{k}(p_{n})=\sum_{i=0}^{k-1}p_{n+i}. Our numerical and heuristic investigations suggest the following phenomenon: for every prime p_{n} , there exists at least one odd length k\geq 3 such that S_{k}(p_{n}) is prime. This observation aligns with probabilistic and heuristic studies on prime numbers, such as Cramér’s model [ 11 , 12 ] or Hardy-Littlewood conjectures [ 7 ] . This study aims to formalize this conjecture, present the computational results up to the one-millionth prime, and provide heuristic, modular, and Diophantine arguments supporting its plausibility. We also discuss how this problem conceptually relates to classical additive problems while introducing the constraints of consecutivity and odd length sums. 1.2 Statement of the Conjecture Conjecture 1 (Consecutive Prime Sums) . For every integer n>1 , there exists an odd integer k\geq 3 such that S_{k}(p_{n})=p_{n}+p_{n+1}+\cdots+p_{n+k-1}\in{\mathscr{P}}. The conjecture is existential: it does not assert the uniqueness of k nor the existence of a universal bound. 1.3 Computational Results An algorithm was implemented to test, for each prime p_{n} with n\leq 10^{6} ( p_{10^{6}}=15,485,863 ), the existence of an odd k\geq 3 such that S_{k}(p_{n}) is prime. Algorithm 1 Searching for sums of consecutive primes that are prime 1: List of N=10^{6} primes P=[p_{1},\dots,p_{N}] 2: For each p_{n} , find the minimal odd k with S_{k}(p_{n}) prime 3: for n=1 to 10^{6} do 4: found\leftarrow false 5: k\leftarrow 3 \triangleright Start with the smallest odd length 6: while not found do 7: Compute S_{k}=\sum_{i=0}^{k-1}P[n+i] 8: if S_{k} is prime then 9: Record (n,k,S_{k}) 10: found\leftarrow true 11: else 12: k\leftarrow k+2 \triangleright Move to the next odd length 13: end if 14: end while 15: end for This procedure ensures that for each starting prime p_{n} , the first odd length k giving a prime sum is found. Verification was performed for N=10^{6} . Examples: • The last prime tested: S_{3}(p_{10^{6}})=15485863+15485867+15485917=46457647\in{\mathscr{P}}. • The longest chain found among the first million primes: S_{349}(p_{651511})=9788183+\cdots+9793573=3417024811\in{\mathscr{P}}. • The next longest chain: S_{299}(p_{448696})=6561461+\cdots+6566299=1962589843\in{\mathscr{P}}. Result. — No counterexample was found among the first one million primes. 2 Conditional Discussion: Cramér Model and GRH The conjecture can be analyzed within standard conditional frameworks. 2.1 Cramér Model Cramér’s probabilistic model provides a standard heuristic framework to study the distribution of primes, predicting gaps between consecutive primes of order O((\ln p_{n})^{2}) . Assuming heuristic independence, the probability that S_{k}(p_{n}) is prime increases with k , as the sum spans a larger interval. Under this model, the probability of a counterexample is extremely low, providing a heuristic justification for the universality of the conjecture. 2.2 Application to the Conjecture In this model, each integer n\geq 3 is declared prime independently with probability 1/\ln n . Although this independence assumption is false in a strict sense, it reproduces many known global statistics of primes. Fix a prime p_{n} and consider S_{k}(p_{n})=\sum_{i=0}^{k-1}p_{n+i}. By the prime number theorem, p_{n+i}\sim p_{n}+i\ln p_{n}, which gives, for k small relative to p_{n} , S_{k}(p_{n})\sim kp_{n}+\frac{k(k-1)}{2}\ln p_{n}. The heuristic probability that S_{k}(p_{n}) is prime is \mathbb{P}(S_{k}(p_{n})\ \text{prime})\sim\frac{1}{\ln(kp_{n})}. Considering odd lengths k=3,5,7,\dots,2m+1 and assuming the events are heuristically independent E_{k}=\{S_{k}(p_{n})\ \text{is prime}\}, the probability that none of these sums is prime is \mathbb{P}_{\text{Cram\'{e}r}}(\text{no solution})=\prod_{j=1}^{m}\left(1-\frac{1}{\ln((2j+1)p_{n})}\right) One has \sum_{j=1}^{m}\frac{1}{\ln((2j+1)p_{n})}\sim\frac{m}{\ln p_{n}}, then \prod_{j=1}^{m}\left(1-\frac{1}{\ln((2j+1)p_{n})}\right)\approx\exp\!\left(-\sum_{j=1}^{m}\frac{1}{\ln((2j+1)p_{n})}\right)\longrightarrow 0. Theorem 1 (Cramér-type Heuristic) . Under Cramér’s probabilistic model, the probability that there exists a prime p_{n} for which no odd-length sum S_{k}(p_{n}) is prime is zero. Remark. The events E_{k} are not strictly independent due to local arithmetic constraints. Cramér’s model should be interpreted as describing average behavior rather than providing a formal proof. 2.3 Generalized Riemann Hypothesis (GRH) Consider \pi(x;q,a)=\left|\{p\leq x\mid p\in{\mathscr{P}},\,p\equiv a\pmod{q}\}\right|, \varphi(q) is Euler’s totient function: \varphi(q)=\left|\left\{1\leq k\leq q|\text{ gcd}(k,q)=1\right\}\right|, and \text{Li}(x)=\int_{0}^{x}\frac{dt}{\ln t}, the logarithmic integral function, which provides the best known approximation to \pi(x)\sim\text{Li}(x) . Under the GRH, one has optimal bounds for the error term in the prime number theorem for arithmetic progressions. More precisely, for any q\geq 1 and any a coprime to q , one has \pi(x;q,a)=\frac{\mathrm{Li}(x)}{\varphi(q)}+O\!\left(x^{1/2}\ln(xq)\right), uniformly in a and q where \pi(x;q,a) denotes the number of primes \leq x that are congruent to a\pmod{q} . For consecutive sums S_{k}(p_{n})=p_{n}+\cdots+p_{n+k-1}, For a fixed modulus q , the residue class of S_{k}(p_{n}) modulo q depends on the residue classes of the p_{n+i} . Under GRH, the near-uniform distribution of primes in residue classes modulo q implies that, as k varies, the values of S_{k}(p_{n})\bmod q are asymptotically uniformly distributed, in the absence of any explicit structural obstruction. A systematic failure of the conjecture would contradict this distribution. In particular, for any prime \ell , the probability that \ell divides S_{k}(p_{n}) is approximately \frac{1}{\ell} , except for well-identified local configurations for small values of k . Thus, under GRH, no systematic divisibility by a small prime can persist for infinitely many odd lengths k . Moreover, under GRH, the gaps between consecutive primes satisfy p_{n+1}-p_{n}=O(\sqrt{p_{n}}\ln p_{n}), which allows us to control the growth of S_{k}(p_{n}) and to