On certain classes of $Sp(2,R)$ symmetric $G_2$ structures

Pawe{\l} Nurowski

REVIEW 3 major objections 4 minor 50 references

On certain classes of $Sp(2,R)$ symmetric $G_2$ structures

T0 review · 3 major / 4 minor · reviewed 2026-08-14 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗ Review JSON

Pith's one-line read The paper constructs two families of $\mathrm{Sp}(2,\mathbb{R})$-invariant split $G_2$ structures in seven dimensions, one integrable ($\tau_2=0$) and one coclosed ($\tau_1=\tau_2=0$).

desk verdict Explicit split G2 examples that are genuinely new; torsion claims are unshown but checkable, and the genericity worry is not real. read the letter →

arxiv 1908.04544 v1 pith:VARBUYXC submitted 2019-08-13 math.DG

Pawe{l} Nurowski This is my paper · ORCID

classification math.DG MSC 53C2953C3053C50

keywords G2structuressplitrealformSp(2R)homogeneousspacestorsionintegrablecoclosedrootdiagram

verification ladder T0 review T1 audit T2 compute T3 formal

The pith

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

The reading

The paper constructs explicit $G_2$ structures on two seven-dimensional homogeneous spaces obtained by quotienting $\mathrm{Sp}(2,\mathbb{R})$ by two different $\mathrm{SL}(2,\mathbb{R})$ subgroups, one attached to the long roots and one to the short roots of $\mathfrak{sp}(2,\mathbb{R})$. The metrics are pseudo-Riemannian of signature $(3,4)$, corresponding to the split real form of $G_2$. On the long-root quotient the most general compatible pair is a three-parameter family with vanishing $\tau_2$, so every member is integrable; on the short-root quotient the most general compatible pair is a one-parameter family with $\tau_1=\tau_2=0$, so the structure is coclosed. The paper notes that the two homogeneous spaces are geometrically distinct because their invariant rank-three distributions have different growth.

What carries the argument

The central mechanism is the compatibility condition $(X\lrcorner\varphi)\wedge(Y\lrcorner\varphi)\wedge\varphi=3g(X,Y)\,\mathrm{vol}(g)$, which pairs a metric and a 3-form into a $G_2$ structure, together with the torsion equations $d\varphi = \tau_0\star\varphi + 3\tau_1\wedge\varphi + \star\tau_3$ and $d\star\varphi = 4\tau_1\wedge\star\varphi + \tau_2\wedge\varphi$. These convert the search into an algebraic classification: find all invariant 3-forms on the quotient that satisfy the compatibility equation with the Killing-form metric, then compute their exterior derivatives in the invariant coframing. The two choices of subgroup arise from the distinction between long and short roots in the root diagram of $\mathfrak{sp}(2,\mathbb{R})$; this root-geometry difference is what makes the two quotient spaces and their $G_2$ structures different.

What would settle it

Take the 3-form from Corollary 3.4 with fixed generic parameters $(a,p,q)$, compute its exterior derivative in the coframing (3.1), and substitute into the torsion equations; if the resulting $\tau_2$ is not identically zero, the theorem is false. Similarly for Theorem 4.5, compute $d\star\varphi$ and check whether $\tau_1$ and $\tau_2$ vanish.

Watch

Extended reading notes

Core claim

For $M_l = \mathrm{Sp}(2,\mathbb{R})/\mathrm{SL}(2,\mathbb{R})_l$, starting from the invariant metric $g_K$ obtained by restricting the Killing form, the compatibility condition $(X\lrcorner\varphi)\wedge(Y\lrcorner\varphi)\wedge\varphi = 3g_K(X,Y)\,\mathrm{vol}(g_K)$ reduces the general invariant 3-form to a three-parameter family with parameters $(a,p,q)$. Solving the torsion equations for this family yields $\tau_0$, $\tau_1$, $\tau_3$ as explicit functions of $a,p,q$ and $\tau_2=0$ identically, so all these $G_2$ structures are integrable, meaning they admit a totally skew-symmetric torsion. For $M_s = \mathrm{Sp}(2,\mathbb{R})/\mathrm{SL}(2,\mathbb{R})_s$, the same procedure gives a one-parameter family with parameter $q$ and torsion $\tau_0=-18/7$, $\tau_3=\frac{2}{7}(4f_{147}+f_{246}+2f_{345})-\frac{3}{7}(qf_{136}+q^{-1}f_{257})$, while $\tau_1=\tau_2=0$, so the structure is coclosed. The author remarks that a subfamily of the $M_l$ structures obtained by $p=2a$ is also coclosed, but believes the two geometries are genuinely nonequivalent because the invariant rank-three distributions on $M_l$ and $M_s$ have different growth (constant $(2,3)$ versus integrable).

Load-bearing premise

The torsion classification rests on a lengthy algebraic computation that is asserted without derivation; if that computation contains an error, the claimed torsion types would not follow.

Editorial extensions

If this is right

If the theorems are correct, explicit examples of split $G_2$ structures with $\tau_2=0$ exist on $M_l$, giving a three-parameter family of integrable structures with totally skew-symmetric torsion.
On $M_s$, the one-parameter family with $\tau_1=\tau_2=0$ provides explicit coclosed $G_2$ structures in signature $(3,4)$, the torsion type relevant to certain physical compactifications.
Setting $p=2a$ in the $M_l$ family produces coclosed structures on $M_l$, so coclosed split $G_2$ structures also exist on the long-root quotient.
The explicit torsion components give a complete description of the intrinsic torsion for these homogeneous geometries, going beyond existence statements.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

The compatibility equation with a nondegenerate metric automatically guarantees that the 3-form lies in an open orbit of $\mathrm{GL}(7,\mathbb{R})$, so the genericity assumption stated in the paper may be redundant for these explicit solutions.
The unshown torsion computations could be verified by a direct computer-algebra substitution, which would quickly settle the correctness of the two theorems.
Because the parameter $q$ in the $M_s$ family enters as $q f_{136}+q^{-1}f_{257}$, different $q$ values may give non-isometric $G_2$ structures; one could test this by comparing curvature invariants.
The existence of a three-parameter integrable family on $M_l$ suggests that deforming the structure within this family preserves integrability, which is not typical for general $G_2$ geometries and could be explored further.

Share X Bluesky LinkedIn Reddit HN

Signed reviews

No signed human review yet.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Desk Editor's Note

read the letter

Two things you should know. First, this is a genuinely new set of examples: explicit split G2 structures on Sp(2,R)/SL(2,R)_l with tau2=0 and on Sp(2,R)/SL(2,R)_s with tau1=tau2=0, neither of which appears in the cited literature. Second, the central torsion claims are asserted without derivation, so the paper stands or falls on algebra you cannot see.

What the paper does well: it takes the standard invariant-geometry framework, solves the compatibility equations explicitly, and produces clean 3-parameter and 1-parameter families. The exposition is straightforward, and the author is clear that this is an example note rather than a new theory. The self-citation to the car paper is motivational only—no circularity.

Soft spots, in order of importance. The torsion components in Theorems 3.5 and 4.5 are stated, not derived. No computer algebra is supplied, and no structural argument shows e.g. that tau3 lies in the 27-dimensional component or that the splitting of d*phi is correct. A single sign error would change the torsion type and invalidate the advertised classification. This is the load-bearing unverified part. Also, the 'most general' claims divide by a, p, q−1, q without discussing exceptional cases; that affects completeness but not existence of the families. The final speculation that the two families are nonequivalent is plausible but not proven—fine as a remark, but it should be labeled as such.

One concern in the reader's report does not hold up: the worry that phi might not be a generic/stable 3-form. Equation (3.5) forces the bilinear form B_phi to equal 3 g vol, and since g is nondegenerate, B_phi is nondegenerate, which is exactly the stability condition for a 3-form in seven dimensions. That gap is not real.

Who this is for: people working in G2 geometry, especially split signature examples and their torsion strata. It is a useful testbed. It deserves a serious referee: a referee can check the algebra (or ask for the script) and verify the classifications. The paper is not a paradigm shift, but it is a solid, citable source of examples. I'd send it to review.

Referee Report

3 major / 4 minor

Summary. The paper studies invariant G2 structures on the seven-dimensional homogeneous spaces M_l = Sp(2,R)/SL(2,R)_l and M_s = Sp(2,R)/SL(2,R)_s, where the two subgroups correspond to the long-root and short-root sl(2,R) subalgebras. Working with the split real form of G2, the author restricts attention to the metric g_K obtained from the Killing form and solves the compatibility equations for invariant 3-forms. On M_l (Section 3) the compatible pairs form a 3-parameter family, and Theorem 3.5 states that all of them have torsion τ2=0. On M_s (Section 4) the compatible pairs form a 1-parameter family, and Theorem 4.5 states that τ1=τ2=0, equivalently d*φ=0. The paper concludes that these are explicit families of integrable, respectively coclosed, split G2 structures.

Significance. If the torsion formulas are correct, the paper supplies new explicit examples of split G2 structures in two torsion classes, with the metric fixed by the Killing form and the 3-form varying in a low-dimensional family. The construction is direct and is not circular: the torsion components are obtained from Bryant's equations rather than fitted to data, and the self-cited car paper is only motivational. The main mathematical content, however, is concentrated in the unshown algebraic computations behind Theorems 3.5 and 4.5, and the manuscript as written does not allow the reader to check them. The explicit formulas are a strength: they are concrete and, in principle, machine-checkable.

major comments (3)

[Section 3.3, Theorem 3.5] The classification τ2=0 for the whole 3-parameter family is the paper's central claim, but the computation of dφ and d*φ and their decomposition into irreducible G2 components is not shown. In particular, the assertion that τ3 lies in the 27-dimensional irreducible component Λ^3_27, equivalently τ3∧φ=0 and τ3∧*φ=0, is stated without verification. Since a single coefficient or sign error would change the torsion type, the manuscript should include the derivation, a computer-algebra script, or at least a detailed outline of the calculation.
[Section 4.1, Theorem 4.5] The same issue arises for the coclosed claim d*φ=0, equivalently τ1=τ2=0. The equality is asserted with no computation, and this is the entire new content for M_s. The derivation of the reductions in Proposition 4.3, which leads to the 1-parameter family, is also not supplied. The theorem cannot be considered established unless the computation is made verifiable.
[Corollaries 3.4 and 4.4] The phrase 'most general' is not fully justified because the derivation divides by a, p, and q−1 in the M_l case and by q in the M_s case, and the excluded values are not discussed. If compatible pairs exist at those parameter values, the uniqueness claims in the corollaries and theorems are false; if such pairs do not exist, the exclusion should be stated and proved. This does not affect the existence of the advertised families, but it affects the completeness of the classification.

minor comments (4)

[Section 3.2] The genericity condition for φ is not verified explicitly. It would be helpful to state that (3.5), together with nondegeneracy of g_K, forces the bilinear form defined by the left-hand side to be nondegenerate, so φ automatically lies in an open GL(7,R) orbit; as written the reader must infer this.
[Section 2] The text refers to the 'Killing form for sl(2,R)' but the displayed formula is the Killing form of sp(2,R); this should be clarified to avoid confusion.
[Final paragraph] The assertion that the G2 geometries on M_l and M_s are 'really nonequivalent' is based on a comparison of invariant distributions; this is suggestive but not a proof. If this is intended as a theorem, an argument should be supplied; otherwise it should be phrased as a conjecture.
[Throughout] There are several typos and small inconsistencies, including 'homogoneous', 'restirict', 'diﬀerent', 'cooresponding', and the use of 'SL(2,R)_l/SL(2,R)_s' notation in the abstract. These should be corrected in a revision.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity; the construction is an explicit invariant algebraic derivation.

full rationale

The paper derives Sp(2,R)-invariant G2 structures by (i) writing Maurer-Cartan equations for invariant forms, (ii) classifying invariant metrics and 3-forms via Lie derivative conditions (Propositions 3.1, 3.2, 4.1, 4.2), (iii) solving the compatibility condition (3.5) for constant coefficients (Propositions 3.3, 4.3), and (iv) stating torsion forms obtained by substituting phi into Bryant's equations (3.6). No parameter is fitted to the claimed outputs; the torsion components in Theorems 3.5 and 4.5 are asserted consequences of direct differentiation, and the 'most general' claims follow from solving algebraic equations for coefficients. The only self-citation, [5], appears as motivation from a car problem and is not load-bearing for the torsion classification. Bryant's framework [1,2] is standard external background, not an input that secretly encodes the result. The absence of displayed derivations for the torsion algebra is a verification gap, not circularity: nothing in the paper defines tau1=tau2=0 into existence or renames an input as a prediction. The genericity of phi is also not a circular step: the compatibility equation defines a nondegenerate bilinear form 3g_K, which places phi in an open GL(7,R) orbit. Hence the central claims have independent computational content and no circular reduction is exhibited.

Assumptions & free parameters 0 free parameters · 5 assumptions · 0 invented entities

No new particles, forces, or auxiliary structures are postulated. The paper relies on the standard sp(2,R) Lie algebra structure, the standard compatibility and torsion framework, and a deliberate restriction to the Killing-form metric. The unverified genericity of phi is the main hidden assumption.

assumptions (5)

standard math The Maurer-Cartan equations (3.1) and (4.1) correctly encode the Lie algebra sp(2,R) in the chosen bases.
All later invariant metric and 3-form computations use these equations as input; they are derived from the displayed commutation relations.
standard math A compatible pair (g, phi) satisfying (3.5), with phi generic, defines a split G2 structure, and Bryant's torsion equations (3.6) define the torsion components.
This is the standard framework cited to [1,2], and the paper's conclusions about tau0,...,tau3 depend on it.
standard math Invariance of g and phi under the leaf distribution is sufficient to descend them to the quotient homogeneous spaces.
Used when passing from Sp(2,R) to M_l and M_s.
ad hoc to paper The restriction to the single Killing-form induced metric g_K is an intentional scope limitation, stated at Sections 3.1 and 4.
The classification claims are only for g=g_K, not for the full 7-parameter and 4-parameter invariant metric families found in Propositions 3.1 and 4.1.
domain assumption The 3-forms phi are generic, meaning they lie in an open GL(7,R) orbit, though this is not verified in the paper.
Section 3.2 requires genericity for a G2 structure; the paper does not check it for Corollaries 3.4 and 4.4.

how reviews work

0 comments

Cite this review

Pith. "Pith review of On certain classes of $Sp(2,R)$ symmetric $G_2$ structures." pith.science (2026). https://pith.science/paper/VARBUYXC

@misc{pith2026190804544,
  author       = {Pith},
  title        = {Pith review of: On certain classes of $Sp(2,R)$ symmetric $G_2$ structures},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/VARBUYXC}},
  note         = {Machine review of arXiv:1908.04544}
}

abstract

We find two different families of $Sp(2,R)$ symmetric $G_2$ structures in seven dimensions. These are $G_2$ structures with $G_2$ being the split real form of the simple exceptional complex Lie group $G_2$. The first family has $\tau_2\equiv 0$, while the second family has $\tau_1\equiv\tau_2\equiv 0$. The families are different in the sense that the first one lives on a homogoneous space $Sp(2,R)/SL(2,R)_l$, and the second one lives on a homogeneous space $Sp(2,R)/Sl(2,R)_s$. Here $SL(2,R)_l$ is an $SL(2,R)$ corresponding to the $\mathfrak{sl}(2,R)$ related to the long roots in the root diagram of $\mathfrak{sp}(2,R)$, and $SL(2,R)_s$ is an $SL(2,R)$ corresponding to the $\mathfrak{sl}(2,R)$ related to the short roots in the root diagram of $\mathfrak{sp}(2,R)$.

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

50 extracted references · 48 canonical work pages

[1]

D. An, P. Nurowski, Twistor space for rolling bodies, Commun. Math. Phys., 326 , 393 - 414 (2014)

work page 2014
[2]

A. A. Agrachev, ``Rolling balls and Octonions'', Proceedings of the Steklov Institute of Mathematics 08/2007; 258(1):13-22, doi:10.1134/S0081543807030030

work page doi:10.1134/s0081543807030030 2007
[3]

Agrachev and Yu.L

A.A. Agrachev and Yu.L. Sachkov, ``An Intrinsic Approach to the Control of Rolling Bodies'', Proceedings of the 38-th IEEE Conference on Decision and Control , vol. 1, Phoenix, Arizona, USA, December 7-10, 1999, 431-435

work page 1999
[4]

M. F. Atiyah, N. J. Hitchin and I. M. Singer, ``Self-duality in four dimensional Riemannian geometry'' Proc. R. Soc. A 362, 425--61 (1978)

work page 1978
[5]

R. L. Bryant, \'Elie Cartan and geometric duality, in Journees Elie Cartan 1998 et 1999, Institut Elie Cartan 16 (2000), 5--20, available at: https://services.math.duke.edu/\ bryant/Cartan.pdf

work page 2000
[6]

Bryant and L

R. Bryant and L. Hsu, ``Rigidity of integral curves of rank two distributions'', Invent. Math. 114 (1993), 435--461; http://citeseerx.ist.psu.edu/viewdoc/summary?doi=10.1.1.46.7262

work page 1993
[7]

Bryant, R. L. (1987) Metrics with exceptional holonomy, Ann. of Math. (2) 126 , no. 3, 525--576

work page 1987
[8]

Akbulut, T Onder, and R.J

Bryant, R, L (2006) Some remarks on G_2 structures, Proceeding of Gokova Geometry-Topology Conference 2005, edited by S. Akbulut, T Onder, and R.J. Stern (2006), International Press, 75--109, https://arxiv.org/abs/math/0305124

arXiv 2006

Show all 50 references

[9]

Bor and R

G. Bor and R. Montgomery, `` G_2 and the rolling distribution'', L'Enseignement Mathematique 55 (2009), 157--196; arXiv:math/0612469

2009 arXiv
[10]

R. L. Bryant, ``Two exotic holonomies in dimension four, path geometries, and twistor theory'', in Complex geometry and Lie theory (Sundance, UT, 1989), 33 -- 88, Proc. Sympos. Pure Math. 53, Amer. Math. Soc., Providence, RI, 1991

1989
[11]

R. L. Bryant, ``Elie Cartan and geometric duality'', Journees Elie Cartan 1998 et 1999, Institut Elie Cartan 16 (2000), \\5 -- 20, available at: https://services.math.duke.edu/ \,\,\, bryant/Cartan.pdf

2000
[12]

Cartan, ``Sur la structure des groupes simples finis et continus'', C

E. Cartan, ``Sur la structure des groupes simples finis et continus'', C. R. Acad. Sci. 116 (1893), 784--786
[13]

Cartan E, ``Les systemes de Pfaff a cinq variables et les equations aux derivees partielles du seconde ordre'' Ann. Sc. Norm. Sup. 27, 109-192 (1910)

1910
[14]

A. Cap, J. Slov\'ak, (2009) Parabolic Geometries I, Background and General Theory, Math. Surveys Monogr., vol. 154, Amer. Math. Soc

2009
[15]

Cartan, (1893) Sur la structure des groupes simples finis et continus, C

\'E. Cartan, (1893) Sur la structure des groupes simples finis et continus, C. R. Acad. Sci. 116 , 784--786
[16]

Cartan, (1924) Sur les varietes a connection projective, Bull

\'E. Cartan, (1924) Sur les varietes a connection projective, Bull. Soc. Math. France 52 205--241

1924
[17]

Cartan (1941) La geometria de las ecuaciones diferenciales de tercer orden Rev

\'E. Cartan (1941) La geometria de las ecuaciones diferenciales de tercer orden Rev. Mat. Hispano-Amer. 4 1--31

1941
[18]

S. S. Chern, The geometry of differential equation y'''=F(x,y,y',y'') , Sci. Rep. Nat. Tsing Hua Univ., 4 (1940), 97-111

1940
[19]

Eastwood, P

M. Eastwood, P. Nurowski (2019) Aerobatics of flying saucers, https://arxiv.org/abs/1810.04852

2019 arXiv
[20]

Eastwood, P

M. Eastwood, P. Nurowski (2019) Aerodynamics of flying saucers, https://arxiv.org/abs/1810.04855

2019 arXiv
[21]

Engel, (1893) Sur un groupe simple a quatorze parametres, C

F. Engel, (1893) Sur un groupe simple a quatorze parametres, C. R. Acad. Sci. 116 , 786--788
[22]

S. S. Chern, (1966), The geometry of G-structures, Bull. Amer. Math. Soc. 72 , 167-219

1966
[23]

Fox, Contact projective structures, Indiana University Mathematics Journal 54 (2005) 1547, arXiv:math/0402332

D. Fox, Contact projective structures, Indiana University Mathematics Journal 54 (2005) 1547, arXiv:math/0402332

2005 arXiv
[24]

Friedrich Th, Ivanov S (2002) Parallel spinors and connections with skew-symmetric torsion in string theory, Asian J. of Math. 6 , 303--336. arXiv:math.DG/0102142

2002
[25]

Friedrich Th, Ivanov S (2003) Killing spinor equations in dimension 7 and geometry of integrable G_2 -manifolds, J. Geom. Phys. 48 , 1-11, arXiv:math.DG/0112201

2003
[26]

Godli\'nski, (2008) Geometry of third order ordinary differential equations and its applications in General Relativity, PhD thesis, Faculty of Physics, University of Warsaw

M. Godli\'nski, (2008) Geometry of third order ordinary differential equations and its applications in General Relativity, PhD thesis, Faculty of Physics, University of Warsaw

2008
[27]

Godli\'nski, P

M. Godli\'nski, P. Nurowski, Geometry of third order ODEs, (2009) arXiv: 0902.4129,

2009 arXiv
[28]

Gutt, Projective geometry of simple nonholonomic models, Colloquium at CFT PAS, 24 January 2014, https://www.youtube.com/watch?v=4LV3J\_mPUAI

J. Gutt, Projective geometry of simple nonholonomic models, Colloquium at CFT PAS, 24 January 2014, https://www.youtube.com/watch?v=4LV3J\_mPUAI

2014
[29]

Helgason, Sophus Lie, the mathematician, In Proceedings Sophus Lie Memorial Conference, Oslo 1992, O

S. Helgason, Sophus Lie, the mathematician, In Proceedings Sophus Lie Memorial Conference, Oslo 1992, O. A. Laudal, B. Jahren, eds. Oslo, (1994), pp. 3-21

1994
[30]

C. D. Hill, P. Nurowski, Differential equations and para-CR structures, Bollettino dell'Unione Matematica Italiana, (9) III (2010) 25-91

2010
[31]

C. D. Hill, P. Nurowski (2019) A car as parabolic geometry, https://arxiv.org/abs/1908.01169

2019 arXiv
[32]

C. D. Hill, P. Nurowski, (2019) Nonholonomic mechanics and the exceptional simple Lie group G_2 , in preparation

2019
[33]

Pansu, (1989), M\'etriques de Carnot-Carath\'eodory et quasiisom\'etries des espaces sym\'etriques de rang un, Annals of Mathematics, Second Series, 129 (1): 1 -- 60

P. Pansu, (1989), M\'etriques de Carnot-Carath\'eodory et quasiisom\'etries des espaces sym\'etriques de rang un, Annals of Mathematics, Second Series, 129 (1): 1 -- 60

1989
[34]

Penrose, (1967), Twistor algebra, J

R. Penrose, (1967), Twistor algebra, J. Math. Phys. 8 , 345 -- 66

1967
[35]

Ward, R., Wells, Jr, R. (1990). Twistor Geometry and Field Theory (Cambridge Monographs on Mathematical Physics). Cambridge: Cambridge University Press. doi:10.1017/CBO9780511524493

1990 doi
[36]

Eastwood, P

M. Eastwood, P. Nurowski, ``Aerodynamics of flying saucers'', (2018)

2018
[37]

Engel, ``Sur un groupe simple a quatorze parametres'', C

F. Engel, ``Sur un groupe simple a quatorze parametres'', C. R. Acad. Sci. 116 (1893), 786--788
[38]

C. R. Graham, T. Willse, ``Parallel tractor extension and ambient metrics of holonomy split G_2 '' J. Differential Geom. , to appear, arXiv:1109.3504v1

arXiv
[39]

Hilbert Lecture notes
[40]

(2010) ``Stabilizers of null k -planes in split real form of the exceptional Lie group G_2 '', mimeographic notes

Leistner Th., Nurowski P. (2010) ``Stabilizers of null k -planes in split real form of the exceptional Lie group G_2 '', mimeographic notes

2010
[41]

Montgomery, A Tour of Subriemannian Geometries, Their Geodesics and Applications,

R. Montgomery, A Tour of Subriemannian Geometries, Their Geodesics and Applications,
[42]

Mano G., Nurowski P., Sagerschnig K (2018) ``The geometry of marked contact twisted cubic structures''

2018
[43]

Nurowski P, ``Twistor bundles, Einstein equations and real structures'' Class. Q. Grav. , 14, (1997), A261--A290

1997
[44]

Differential equations and conformal structures

P. Nurowski, "Differential equations and conformal structures" Journ. Geom. Phys. 55, 19-49 (2005)

2005
[45]

Nurowski, ``Comment on (2, ) geometry of 4th order ODEs'', J

P. Nurowski, ``Comment on (2, ) geometry of 4th order ODEs'', J. Geom. Phys., 59 (2009) 267-278

2009
[46]

P. Nurowski, ``On exceptional contact geometries'', talk at the Australian National University, 25th October 2013, notes avaliable at http://www.fuw.edu.pl/ \,\,\, nurowski/prace/talk\_canberra.pdf

2013
[47]

Penrose, ``Twistor algebra'', J

R. Penrose, ``Twistor algebra'', J. Math. Phys. 8 345--66 (1967)

1967
[48]

Strazzullo, Symmetry Analysis of General Rank--3 Pfaffian Systems in Five Variables , (2009), PhD Thesis, Utah State University, Logan, Utah

F. Strazzullo, Symmetry Analysis of General Rank--3 Pfaffian Systems in Five Variables , (2009), PhD Thesis, Utah State University, Logan, Utah

2009
[49]

Szereszewski, private communications

A. Szereszewski, private communications
[50]

Zelenko, ``On variational approach to differential invariants of rank two distributions'', Diff

I. Zelenko, ``On variational approach to differential invariants of rank two distributions'', Diff. Geom. Appl. 24 (2006), 235--259; http://www.sciencedirect.com/science/article/pii/S0926224505000902

2006

Pith tools

Reviewed August 14, 2026 · model on record in the stance chip above.

[1] [1]

D. An, P. Nurowski, Twistor space for rolling bodies, Commun. Math. Phys., 326 , 393 - 414 (2014)

work page 2014

[2] [2]

A. A. Agrachev, ``Rolling balls and Octonions'', Proceedings of the Steklov Institute of Mathematics 08/2007; 258(1):13-22, doi:10.1134/S0081543807030030

work page doi:10.1134/s0081543807030030 2007

[3] [3]

Agrachev and Yu.L

A.A. Agrachev and Yu.L. Sachkov, ``An Intrinsic Approach to the Control of Rolling Bodies'', Proceedings of the 38-th IEEE Conference on Decision and Control , vol. 1, Phoenix, Arizona, USA, December 7-10, 1999, 431-435

work page 1999

[4] [4]

M. F. Atiyah, N. J. Hitchin and I. M. Singer, ``Self-duality in four dimensional Riemannian geometry'' Proc. R. Soc. A 362, 425--61 (1978)

work page 1978

[5] [5]

R. L. Bryant, \'Elie Cartan and geometric duality, in Journees Elie Cartan 1998 et 1999, Institut Elie Cartan 16 (2000), 5--20, available at: https://services.math.duke.edu/\ bryant/Cartan.pdf

work page 2000

[6] [6]

Bryant and L

R. Bryant and L. Hsu, ``Rigidity of integral curves of rank two distributions'', Invent. Math. 114 (1993), 435--461; http://citeseerx.ist.psu.edu/viewdoc/summary?doi=10.1.1.46.7262

work page 1993

[7] [7]

Bryant, R. L. (1987) Metrics with exceptional holonomy, Ann. of Math. (2) 126 , no. 3, 525--576

work page 1987

[8] [8]

Akbulut, T Onder, and R.J

Bryant, R, L (2006) Some remarks on G_2 structures, Proceeding of Gokova Geometry-Topology Conference 2005, edited by S. Akbulut, T Onder, and R.J. Stern (2006), International Press, 75--109, https://arxiv.org/abs/math/0305124

arXiv 2006