There are also several minor changes involving clarifications and corrections of misprints.
The Abstract of the prior report remains the same as follows: The galaxies of the nonstandard enlargements of connected, conventionally infinite graphs as well as of connected transfinite graphs are defined, analyzed, and illustrated by some examples. It is then shown that any such enlargement either has exactly one galaxy, its principal one, or it has infinitely many galaxies. In the latter case, the galaxies are partially ordered by their “closeness” to the principal galaxy. If an enlargement has a galaxy different from its principal galaxy, then it has a twoway infinite sequence of galaxies that are totally ordered according to that “closeness” property. There may be many such totally ordered seqences.
Key Words: Nonstandard graphs, enlargements of graphs, transfinite graphs, galaxies in nonstandard graphs, graphical galaxies.
1 Introduction
In this work we extend the idea of galaxies in the hyperreal line
${}^{*}IR$
to nonstandard enlargements of conventionally infinite graphs and also of transfinite graphs. We stipulate henceforth that every graph considered herein is connected. Since graphs have structures much different from that of the real line
$IR$
, the enlargements of graphs have properties not possessed by
${}^{*}IR$
. The graphical galaxies of those enlargements comprise one aspect of that distinctive complexity. We will show that that any such enlargement has either one galaxy or infinitely many of them. Moreover, just as
${}^{*}IR$
contains images of the real numbers, called the standard hyperreals, as well as hyperreals that are nonstandard, so too may the enlargement
${}^{*}G$
of a graph
$G$
contain “hypernodes,” some of which are images of nodes of
$G$
and others of which are nonstandard hypernodes. In addition, there are “hyperbranches” incident to pairs of hypernodes; some of these hyperbranches are images of branches of
$G$
, but there may be others that are not.
The galaxies graphically partition
${}^{*}G$
in the sense the every hypernode belongs to exactly one galaxy, and so too does every hyperbranch. There is a unique galaxy, which we refer to as the “principal galaxy,” that contains the standard hypernodes and possibly nonstandard hypernodes as well. In the event that there are infinitely many galaxies, those galaxies are partially ordered according to how “close” they are to the principal galaxy. In fact, if there is a galaxy different from the principal galaxy, then there is a twoway infinite sequence of galaxies that are totally ordered according to their “closeness” to the principal galaxy.
There may be many such totally ordered sequences, but a galaxy in one such sequence may not be comparable to a galaxy in another sequence according to that “closeness” property.
We speak of “conventionally infinite” graphs to distinguish them from transfinite graphs of ranks 1 or higher [
5,Chapter2]
, [
6,Chapter2]
. Sections 2 through 4 herein are devoted to the enlargements of conventionally infinite graphs. The results for such enlargements extend to enlargements of transfinite graphs, but in more complicated ways. We show this in Sections 5 through 11, but only for transfinite graphs of rank 1. Results for transfinite graphs of still higher ranks are obtained similarly but in still more complicated ways and with additional complexity in the symbols. For the sake of brevity, the latter results are not included herein, but they may be found in [
7]
as well as in the archive www.arxiv.org in the category “mathematics” under “Zemanian.” Our notations and terminology follow the usual conventions of nonstandard analysis.
$IN=\{0,1,2,\dots \}$
is the set of natural numbers, and
${}^{*}IN$
is the set of hypernaturals. The standard hypernaturals are (i.e., can be identified with) the natural numbers. Also,
$\langle {a}_{n}\rangle $
or
$\langle {a}_{n}:n\in IN\rangle $
or
$\langle {a}_{0},{a}_{1},{a}_{2},\dots \rangle $
denotes a sequence whose elements can be members of any set, such as the set
$X$
of nodes in a conventional graph
$G=\{X,B\}$
, where
$B$
is the set of branches, a branch being a twoelement set of nodes. On the other hand,
$\left[{a}_{n}\right]$
denotes an equivalence class of sequences, where two sequences
$\langle {a}_{n}\rangle $
and
$\langle {b}_{n}\rangle $
are taken to be equivalent if
$\{n:{a}_{n}={b}_{n}\}\in \mathcal{\mathcal{F}}$
, where
$\mathcal{\mathcal{F}}$
is any chosen and fixed free ultrafilter.^{
$1$
}
$\mathcal{\mathcal{F}}$
will be so fixed throughout this work. The
${a}_{n}$
appearing in
$\left[{a}_{n}\right]$
are understood to be the elements of any one of the sequences in the equivalence class. At times, we will use the more specific notation
$[\langle {a}_{0},{a}_{1},{a}_{2},\dots \rangle ]$
. More generally, we adhere to the notations and terminology appearing in [
3]
.
The ordinals are denoted in the usual way:
$\omega $
is the first transfinite ordinal. With
$\tau \in IN$
, the product
$\omega \cdot \tau $
is the sum of
$\tau $
terms, each being
$\omega $
.
2 The Nonstandard Enlargement of a Graph
Throughout Sections 2 to 4, we assume that the conventionally infinite graph
$G$
is connected and has infinitely many nodes. The definition of a nonstandard graph that we use herein is given in [
6,Section8.1]
, a special case of which is the “enlargement” of a graph
$G$
.
Let us define the
enlargement
${}^{*}G$
of
$G$
here as well in order to remove any need for referring to [
6]
.
$G=\{X,B\}$
is now taken to be a conventional connected graph having an infinite set
$X$
of nodes and therefore an infinite set of branches as well, each branch being a twoelement set of nodes. Thus, there are no parallel branches (i.e., multiple branches).
$\mathcal{\mathcal{F}}$
will denote a chosen and fixed free ultrafilter.
$\mathbf{x}=\left[{x}_{n}\right]$
denotes an equivalence class of sequences of nodes as stated in the Introduction. Specifically,
$\left[{x}_{n}\right]$
and
$\left[{x}_{n}^{\prime}\right]$
are in the same equivalence class
$\mathbf{x}$
if
$\{n:{x}_{n}={x}_{n}^{\prime}\}\in \mathcal{\mathcal{F}}$
.
$\mathbf{x}$
will be called a hypernode.^{
$2$
}
Thus, the set of all sequences of nodes from
$G$
is partitioned into hypernodes.
${}^{*}X$
denotes the set of hypernodes. If all the elements of one of the representative sequences
$\langle {x}_{n}\rangle $
for a hypernode
$\mathbf{x}=\left[{x}_{n}\right]$
are the same node (i.e.,
${x}_{n}=x$
for all
$n$
), then
$\mathbf{x}=\left[x\right]$
can be identified with
$x$
; in this case,
$\mathbf{x}$
is called a standard hypernode. Otherwise,
$\mathbf{x}=\left[{x}_{n}\right]$
is called a nonstandard hypernode.
We turn now to the definition of a “hyperbranch.” Let
$\mathbf{x}=\left[{x}_{n}\right]$
and
$\mathbf{y}=\left[{y}_{n}\right]$
be two hypernodes. Also, let
$\mathbf{b}=\left[\right\{{x}_{n},{y}_{n}\left\}\right]$
, where
$\langle \{{x}_{n},{y}_{n}\}\rangle $
is a sequence of pairs of nodes from
$G$
such that, for almost all
$n$
,
$\{{x}_{n},{y}_{n}\}$
is a branch in
$G$
; that is,
$\{n:\{{x}_{n},{y}_{n}\}\in B\}\in \mathcal{\mathcal{F}}$
. It can be shown [
6,page155]
that this definition is independent of the representative sequences
$\langle {x}_{n}\rangle $
and
$\langle {y}_{n}\rangle $
chosen of
$\mathbf{x}$
and
$\mathbf{y}$
respectively and that we truly have an equivalence relation for the set of all sequences of branches from
$G$
. We let
$\mathbf{b}=\left[\right\{{x}_{n},{y}_{n}\left\}\right]$
denote such an equivalence class and will call it a hyperbranch; we write
$\mathbf{b}=\{\mathbf{x},\mathbf{y}\}$
. Also,
${}^{*}B$
will denote the set of all hyperbranches. If
$\mathbf{x}=\left[{x}_{n}\right]$
and
$\mathbf{y}=\left[{y}_{n}\right]$
are standard hypernodes, then
$\mathbf{b}=\left[\right\{x,y\left\}\right]$
is called a standard hyperbranch. Otherwise,
$\mathbf{b}$
is called a nonstandard hyperbranch.
Finally, the pair
${}^{*}G={\{}^{*}X{,}^{*}B\}$
denotes the enlargement of
$G$
. It is a special case of a nonstandard graph, as defined in [
6,page155]
.^{
$3$
}
3 Distances and Galaxies in Enlarged Graphs
The length
$\left{P}_{x,y}\right$
of any path
${P}_{x,y}$
connecting two nodes
$x$
and
$y$
in a graph
$G$
is the number of branches in
${P}_{x,y}$
. The distance
$d(x,y)$
between
$x$
and
$y$
is
$d(x,y)=min\left\{\right{P}_{x,y}\left\right\}$
, where the minimum is taken over all paths terminating at
$x$
and
$y$
. In the trivial case,
$d(x,x)=0$
.
$d$
satisfies the triangle inequality, namely, for any three nodes
$x$
,
$y$
, and
$z$
in
$G$
,
$d(x,y)\le d(x,z)+d(z,y)$
. In fact,
$d$
satisfies the other metric axioms, too, and the set
$X$
of nodes in
$G$
along with
$d$
is a metric space.
The metric
$d$
can be extended into an internal function
$\mathbf{d}$
mapping the Cartesian product
${}^{*}X{\times}^{*}X$
into the set of hypernaturals
${}^{*}IN$
as follows: For any
$\mathbf{x}=\left[{x}_{n}\right]$
and
$\mathbf{y}=\left[{y}_{n}\right]$
in
${}^{*}X$
,
$\mathbf{d}$
is defined by
$$\mathbf{d}(\mathbf{x},\mathbf{y})=\left[d\right({x}_{n},{y}_{n}]{\in}^{*}IN.$$
By the transfer principle, we have, for any three hypernodes
$\mathbf{x}$
,
$\mathbf{y}$
, and
$\mathbf{z}$
,
$$\begin{array}{c}\mathbf{d}(\mathbf{x},\mathbf{z})\le \mathbf{d}(\mathbf{x},\mathbf{y})+\mathbf{d}\left(\right(\mathbf{y},\mathbf{z}).\end{array}$$ 
(1)

From the point of view of an ultrapower construction, this means that
$$\{n:d({x}_{n},{z}_{n})\le d({x}_{n},{y}_{n})+d({y}_{n},{z}_{n}\}\in \mathcal{\mathcal{F}}.$$
The other metric axioms, such as
$\mathbf{d}(\mathbf{x},\mathbf{x})=0$
, are obviously satisfied by
$\mathbf{d}$
.
We define the “galaxies” of
${}^{*}G$
as nonstandard subgraphs of
${}^{*}G$
by first defining the “nodal galaxies.” Two hypernodes
$\mathbf{x}=\left[{x}_{n}\right]$
and
$\mathbf{y}=\left[{y}_{n}\right]$
are taken to be in the same nodal galaxy
$\dot{\Gamma}$
of
${}^{*}G$
if
$\mathbf{d}(\mathbf{x},\mathbf{y})$
is no greater that a standard hypernatural
$\mathbf{k}$
, that is, if there exists a natural number
$k\in IN$
such that
$\{n:d({x}_{n},{y}_{n})\le k\}\in \mathcal{\mathcal{F}}$
. In this case, we say that
$\mathbf{x}$
and
$\mathbf{y}$
are limitedly distant, and we write
$\mathbf{d}(\mathbf{x},\mathbf{y})\le \mathbf{k}$
.
Let
${N}_{\mathbf{x},\mathbf{y}}$
be the set of all standard hypernaturals that are no less than
$\mathbf{d}(\mathbf{x},\mathbf{y})$
.
${N}_{\mathbf{x},\mathbf{y}}$
is a wellordered set, and therefore it has a minimum
${\mathbf{k}}_{\mathbf{x},\mathbf{y}}$
. So, we can say that
$\mathbf{x}$
and
$\mathbf{y}$
are in the same nodal galaxy
$\dot{\Gamma}$
if
$\mathbf{d}(\mathbf{x},\mathbf{y})={\mathbf{k}}_{\mathbf{x},\mathbf{y}}$
.
Lemma 3.1. The nodal galaxies partition the set
${}^{*}X$
of all hypernodes in
${}^{*}G$
.
Proof. The property of two hypernodes being limitedly distant is a binary relation on
${}^{*}X$
that is obviously reflexive and symmetric. Its transitivity follows directly from ( 1 ).
Alternatively, we can use an ultrapower argument. Assume that
$\mathbf{x}=\left[{x}_{n}\right]$
and
$\mathbf{y}=\left[{z}_{n}\right]$
are in some nodal galaxy and that
$\mathbf{y}$
and
$\mathbf{z}=\left[{z}_{n}\right]$
are in some nodal galaxy; we want to show that those galaxies are the same. There exist two standard natural numbers
${k}_{1}$
and
${k}_{2}$
such that
${N}_{\mathbf{x},\mathbf{y}}=\{n:d({x}_{n},{y}_{n})\le {k}_{1}\}\in \mathcal{\mathcal{F}}$
and
${N}_{\mathbf{y},\mathbf{z}}=\{n:d({y}_{n},{z}_{n})\le {k}_{2}\}\in \mathcal{\mathcal{F}}$
. Since
$d({x}_{n},{z}_{n})\le d({x}_{n},{y}_{n})+d({y}_{n},{z}_{n})$
,
$$\{n:d({x}_{n},{z}_{n})\le {k}_{1}+{k}_{2}\}\supseteq {N}_{\mathbf{x},\mathbf{y}}\cap {N}_{\mathbf{y},\mathbf{z}}\in \mathcal{\mathcal{F}}.$$
So, the lefthand side is a set in
$\mathcal{\mathcal{F}}$
. Thus,
$\mathbf{x}$
and
$\mathbf{z}$
are limitedly distant, too, and
$\mathbf{x}$
,
$\mathbf{y}$
, and
$\mathbf{z}$
are all in the same nodal galaxy.
$\square $
We define a galaxy
$\Gamma $
of
${}^{*}G$
as a maximal nonstandard subgraph of
${}^{*}G$
whose hypernodes are all in the same nodal galaxy
$\dot{\Gamma}$
; that is, the hyperbranches of
$\Gamma $
corresponding to
$\dot{\Gamma}$
are all those pairs
$\{\mathbf{x},\mathbf{y}\}$
such that
$\mathbf{x},\mathbf{y}\in \dot{\Gamma}$
. We will say that a hypernode
$\mathbf{x}$
is in
$\Gamma $
when
$\mathbf{x}\in \dot{\Gamma}$
and that a hyperbranch
$\{\mathbf{x},\mathbf{y}\}$
is in
$\Gamma $
when
$\mathbf{x},\mathbf{y}\in \dot{\Gamma}$
. It follows from Lemma 3.1 that the galaxies of
${}^{*}G$
partition
${}^{*}G$
in the sense of graphical partitioning (i.e., each hyperbranch is in one and only one galaxy).
The
principal galaxy
${\Gamma}_{0}$
of
${}^{*}G$
is that unique galaxy, each of whose hypernodes is limitedly distant from some standard hypernode (and therefore from all standard hypernodes). All the nodes in
$G$
will be (i.e., can be identified with) standard hypernodes in
${\Gamma}_{0}$
, but there may be nonstandard hypernodes in
${\Gamma}_{0}$
as well. The following examples illustrate this point.
Example 3.2. Consider the endless (i.e., twoway infinite) path:
$$P=\langle \dots ,{x}_{1},{b}_{1},{x}_{0},{b}_{0},{x}_{1},{b}_{1}\dots \rangle $$
with nodes
${x}_{k}$
and branches
${b}_{k}$
,
$k\in ZZ$
,
$ZZ$
being the set of integers. The enlargement
${}^{*}P$
of
$P$
has hypernodes, each being represented by
$\left[{x}_{{k}_{n}}\right]$
where
$\langle {k}_{n}\rangle $
is some sequence of integers.
Each hyperbranch is represented by
$\left[\right\{{x}_{{k}_{n}},{x}_{{k}_{n}+1}]$
. The nodal galaxies are infinitely many because they correspond bijectively with the galaxies of the enlargement
${}^{*}ZZ$
of
$ZZ$
. Moreover, the principal galaxy
${\Gamma}_{0}$
of
${}^{*}P$
has only standard hypernodes and in fact is (i.e., can be identified with)
$P$
itself. Also, every galaxy is graphically isomorphic to
${\Gamma}_{0}$
and therefore to every other galaxy.
$\square $
Example 3.3. Now, consider a oneended path:
$$T=\langle {x}_{0},{b}_{0},{x}_{1},{b}_{1},{x}_{2},{b}_{2},\dots \rangle $$
Each hypernode in the enlargement
${}^{*}T$
of
$T$
is represented by
$\left[{x}_{{k}_{n}}\right]$
, where
$\langle {k}_{n}\rangle $
is some sequence of natural numbers. Thus,
${}^{*}T$
has a hypernode set
${}^{*}X$
that can be identified with the set
${}^{*}IN$
of hypernaturals. Hence,
${}^{*}T$
has an infinitely of galaxies, too. The principal galaxy
${\Gamma}_{0}$
of
${}^{*}T$
is the oneended path
$T$
. However, any hypernode
$\mathbf{x}=\left[{x}_{{k}_{n}}\right]$
in a galaxy
$\Gamma $
different from
${\Gamma}_{0}$
will be such that, for every
$m\in IN$
,
$\{n:{k}_{n}>m\}\in \mathcal{\mathcal{F}}$
. Such a hypernode is adjacent both to
$\left[{x}_{{k}_{n}+1}\right]$
and to
$\left[{x}_{{k}_{n}1}\right]$
, where we are free to replace
${x}_{{k}_{n}1}$
by, say,
${x}_{0}$
whenever
${k}_{n}=0$
. (The set
$\{n:{k}_{n}=0\}$
will not be a member of
$\mathcal{\mathcal{F}}$
when
$\mathbf{x}=\left[{x}_{{k}_{n}}\right]$
is in
$\Gamma $
.) Thus,
$\mathbf{x}=\left[{x}_{{k}_{n}}\right]\in \Gamma $
has both a predecessor and a successor, which implies that
$\Gamma $
is graphically isomorphic to an endless path. In fact, all the galaxies other than
${\Gamma}_{0}$
are isomorphic to each other, being identifiable with an endless path.
$\square $
Example 3.4. Consider next the grounded, oneway infinite ladder
$L$
of Figure 1. Now, for every
$k\in IN$
,
$d({x}_{k},{x}_{g})=d({x}_{k},{x}_{k+1})=1$
, and, for every
$k,l\in IN$
with
$kl>1$
,
$d({x}_{k},{x}_{l})=2$
. In this case, for every two hypernodes
$\mathbf{x}$
and
$\mathbf{y}$
,
$\mathbf{d}(\mathbf{x},\mathbf{y})\le \left[2\right]=2$
. Thus, every two hypernodes are limitedly distant from each other, which means that
${}^{*}L$
has only one galaxy, its principal galaxy
${\Gamma}_{0}$
. Now,
${\Gamma}_{0}$
has both standard and nonstandard hypernodes.
$\square $
Example 3.5. Furthermore, consider the graph
$G$
obtained from
$L$
by appending a oneended path
$P$
starting at
${x}_{g}$
, but otherwise isolated from
$L$
, as shown in Figure 2.
In this case, we again have an infinity of galaxies by virtue of the isolation of
$P$
from
$L$
.
The principal galaxy
${\Gamma}_{0}$
has both standard and nonstandard hypernodes, its nonstandard hypernodes being due to
$L$
. All the other galaxies are graphically isomorphic to an endless path (as in Example 3.3) and thus to each other, but not to
$G$
and not to
${\Gamma}_{0}$
.
$\square $
A subgraph
${G}_{s}$
of
$G$
with the property that there exists a natural number
$k$
such that
$d(x,y)\le k$
for all pairs of nodes
$x,y$
in
${G}_{s}$
will be called a finitely dispersed subgraph of
$G$
.
Example 3.5 suggests that the structures of the galaxies other than
${\Gamma}_{0}$
do not depend upon any finitely dispersed subgraph of
$G$
. This is true in general because the nodes
${x}_{n}$
in any representative
$\langle {x}_{n}\rangle $
of any hypernode in a galaxy other than
${\Gamma}_{0}$
must lie outside any finitely dispersed subgraph of
$G$
for almost all
$n$
whatever be the choice of that finitely dispersed subgraph. For instance, consider Example 3.6. Let
${D}_{2}$
be the 2dimensional grid; that is, we can represent
${D}_{2}$
by having its nodes at the lattice points
$(k,l)$
of the 2dimensional plane, where
$k,l\in ZZ$
and with its branches being
$\left\{\right(k,l),(k+1,l\left)\right\}$
and
$\left\{\right(k,l),(k,l+1\left)\right\}$
. So, the hypernodes of
${}^{*}{D}_{2}$
occur at
${}^{*}ZZ{\times}^{*}ZZ$
. Under this representation, the principal nodal galaxy of
${}^{*}{D}_{2}$
will have its nodes at the lattice points of
$ZZ\times ZZ$
.
Next, let
$G$
be a connected graph obtained from
${D}_{2}$
by deleting or appending finitely many branches to
${D}_{2}$
. So, outside a finitely dispersed subgraph of
$G$
,
$G$
is identical to
${D}_{2}$
.
Then the principal galaxy
${\Gamma}_{0}$
of
${}^{*}G$
is the same as (i.e., is graphically isomorphic to)
$G$
, but every other galaxy is the same as
${D}_{2}$
.
$\square $
In view of Examples 3.3 and 3.4, the following theorem is pertinent. As always, we assume that
$G$
is connected and has an infinite node set
$X$
.
Theorem 3.7. Let
$G$
be locally finite. Then,
${}^{*}G$
has at least one hypernode not in its principal galaxy
${\Gamma}_{0}$
and thus at least one galaxy
${\Gamma}_{1}$
different from
${\Gamma}_{0}$
.
Proof. Choose any
${x}_{0}\in X$
. By connectedness and local finiteness, for each
$n\in IN$
, the set
${X}_{n}$
of nodes that are at a distance of
$n$
from
${x}_{0}$
is nonempty and finite. Also,
$\cup {X}_{n}=X$
by the connectedness of
$G$
. By König's Lemma [
4,page40]
, there is a oneended path
$P$
starting at
${x}_{0}$
.
$P$
must pass through every
${X}_{n}$
. Thus, there is a subsequence
$\langle {x}_{0},{x}_{1},{x}_{2},\dots \rangle $
of the sequence of nodes of
$P$
such that
${x}_{n}\in {X}_{n}$
; that is,
$d({x}_{n},{x}_{0})=n$
for every
$n$
. Set
$\mathbf{x}=\left[{x}_{n}\right]$
. Then,
$\mathbf{x}$
must be in a galaxy
${\Gamma}_{1}$
that is different from the principal galaxy
${\Gamma}_{0}$
.
$\square $
4 When
${}^{*}G$
Has a Hypernode Not in Its Principal Galaxy
In this section,
$G$
is connected and infinite but not necessarily locally finite. Let
${\Gamma}_{a}$
and
${\Gamma}_{b}$
be two galaxies that are different from the principal galaxy
${\Gamma}_{0}$
of
${}^{*}G$
. We shall say that
${\Gamma}_{a}$
is closer to
${\Gamma}_{0}$
than is
${\Gamma}_{b}$
and that
${\Gamma}_{b}$
is further away from
${\Gamma}_{0}$
than is
${\Gamma}_{a}$
if there are a
$\mathbf{y}=\left[{y}_{n}\right]$
in
${\Gamma}_{a}$
and a
$\mathbf{z}=\left[{z}_{n}\right]$
in
${\Gamma}_{b}$
such that, for some
$\mathbf{x}=\left[{x}_{n}\right]$
in
${\Gamma}_{0}$
and for every
$m\in IN$
, we have
$${N}_{0}\left(m\right)=\{n:d({z}_{n},{x}_{n})d({y}_{n},{x}_{n})\ge m\}\in \mathcal{\mathcal{F}}.$$
Any set of galaxies for which every two of them, say,
${\Gamma}_{a}$
and
${\Gamma}_{b}$
satisfy this condition will be said to be totally ordered according to their closeness to
${\Gamma}_{0}$
. With Lemma 3.1 in hand, the conditions for a total ordering (reflexivity, antisymmetry, transitivity, and connectedness) are readily shown. For instance, the proof of Theorem 4.3 below establishes transitivity.
Lemma 4.1. These definitions are independent of the representative sequences
$\langle {x}_{n}\rangle $
,
$\langle {y}_{n}\rangle $
, and
$\langle {z}_{n}\rangle $
chosen for
$\mathbf{x}$
,
$\mathbf{y}$
, and
$\mathbf{z}$
.
Proof. Let
$\langle {x}_{n}^{\prime}\rangle $
,
$\langle {y}_{n}^{\prime}\rangle $
, and
$\langle {z}_{n}^{\prime}\rangle $
be any other such representative sequences. Then,
$$d({z}_{n},{x}_{n})\le d({z}_{n},{z}_{n}^{\prime})+d({z}_{n}^{\prime},{x}_{n}^{\prime})+d({x}_{n}^{\prime},{x}_{n}).$$
So,
$$d({z}_{n}^{\prime},{x}_{n}^{\prime})\ge d({z}_{n},{x}_{n})d({z}_{n},{z}_{n}^{\prime})d({x}_{n}^{\prime},{x}_{n})=d({z}_{n},{x}_{n})$$
for all
$n$
in some
${N}_{1}\in \mathcal{\mathcal{F}}$
. Also,
$$d({y}_{n}^{\prime},{x}_{n}^{\prime})\le d({y}_{n}^{\prime},{y}_{n})+d({y}_{n},{x}_{n})+d({x}_{n},{x}_{n}^{\prime})=d({y}_{n},{x}_{n})$$
for all
$n$
in some
${N}_{2}\in \mathcal{\mathcal{F}}$
. Therefore,
$$d({z}_{n}^{\prime},{x}_{n}^{\prime})d({y}_{n}^{\prime},{x}_{n}^{\prime})\ge d({z}_{n},{x}_{n})d({y}_{n},{x}_{n})$$
for all
$n$
in
${N}_{1}\cap {N}_{2}\in \mathcal{\mathcal{F}}$
. So, for
${N}_{0}\left(m\right)$
as defined above and for each
$m$
no matter how large,
$$\{n:d({z}_{n}^{\prime},{x}_{n}^{\prime})d({y}_{n}^{\prime},{x}_{n}^{\prime})\ge m\}\supseteq {N}_{0}\left(m\right)\cap {N}_{1}\cap {N}_{2}\in \mathcal{\mathcal{F}},$$
which implies that the lefthand side is also a set in
$\mathcal{\mathcal{F}}$
. This proves Lemma 4.1.
$\square $
We will say that a set
$A$
is a totally ordered, twoway infinite sequence if there is a bijection from the set
$ZZ$
of integers to the set
$A$
that preserves the total ordering of
$ZZ$
.
Theorem 4.2. If
${}^{*}G$
has a hypernode
$\mathbf{v}=\left[{v}_{n}\right]$
that is not in its principal galaxy
${\Gamma}_{0}$
, then there exists a twoway infinite sequence of galaxies totally ordered according to their closeness to
${\Gamma}_{0}$
and with
$\mathbf{v}$
being in one of those galaxies.
Note. There may be many such sequences, and a galaxy in one sequence and a galaxy in another sequence may not be comparable according to their closeness to
${\Gamma}_{0}$
. Also, a somewhat different version of this theorem with a rather longer proof can be found in the archival website, www.arxiv.org, under Mathematics, Zemanian. Proof. Let
$\mathbf{x}=[\langle x,x,x,\dots \rangle ]$
be a standard hypernode in
${\Gamma}_{0}$
. Also, let
$\mathbf{v}=\left[{v}_{n}\right]$
be the asserted hypernode not in
${\Gamma}_{0}$
. Thus, for each
$m\in IN$
,
$$\begin{array}{c}\{n:d(x,{v}_{n})>m\}\in \mathcal{\mathcal{F}}.\end{array}$$ 
(2)

Between every two nodes of a connected, conventionally infinite graph there is a geodesic path whose length is equal to the distance between those nodes. For each
$n\in IN$
, choose a geodesic path
$P$
in
$G$
terminating at
$x$
and
${v}_{n}$
. If the natural number
$d(x,{v}_{n})$
is even (resp.
odd), there is a unique node
${u}_{n}$
in
$P$
such that
$d(x,{u}_{n})=d({u}_{n},{v}_{n})=d(x,{v}_{n})/2$
(resp.
$d(x,{u}_{n})=d({u}_{n},{v}_{n})1=\left(d\right(x,{v}_{n})1)/2)$
. It follows from this that, if there is a
$k\in IN$
such that
$\{n:d(x,{u}_{n})\le k\}\in \mathcal{\mathcal{F}}$
, then there is a
${k}^{\prime}\in IN$
with
$\{n:d(x,{v}_{n})\le {k}^{\prime}\}\in \mathcal{\mathcal{F}}$
, in violation of ( 2 ). Consequently, for each
$m\in IN$
,
$\{n:d(x,{u}_{n})>m\}\in \mathcal{\mathcal{F}}$
. This implies that
$\mathbf{u}=\left[{u}_{n}\right]$
is in a galaxy
${\Gamma}_{a}$
different from the principal galaxy
${\Gamma}_{0}$
.
Furthermore, with
$d(x,{v}_{n})$
being even (resp. odd) again and with
${u}_{n}$
being chosen as before,
$$d(x,{v}_{n})d(x,{u}_{n})=d(x,{v}_{n})/2$$
(resp.
$$d(x,{v}_{n})d(x,{u}_{n})=\left(d\right(x,{v}_{n})+1)/2).$$
Now, if there is a
$k\in IN$
such that
$$\{n:d(x,{v}_{n})d(x,{u}_{n})\le k\}\in \mathcal{\mathcal{F}},$$
then there is a
${k}^{\prime}\in IN$
such that
$$\{n:d(x,{v}_{n})\le {k}^{\prime}\}\in \mathcal{\mathcal{F}},$$
again in violation of ( 2 ). Thus, for each
$m\in IN$
,
$$\{n:d(x,{v}_{n})d(x,{u}_{n})>m\}\in \mathcal{\mathcal{F}}.$$
This implies that
$\mathbf{u}=\left[{u}_{n}\right]$
and
$\mathbf{v}=\left[{v}_{n}\right]$
are in different galaxies,
${\Gamma}_{a}$
and
${\Gamma}_{b}$
respectively, with
${\Gamma}_{a}$
being closer to
${\Gamma}_{0}$
than is
${\Gamma}_{b}$
.
We can now repeat this argument with
${\Gamma}_{b}$
replaced by
${\Gamma}_{a}$
and with
$\mathbf{u}=\left[{u}_{n}\right]$
playing the role that
$\mathbf{v}=\left[{v}_{n}\right]$
played to find still another galaxy
${\Gamma}_{a}^{\prime}$
different from
${\Gamma}_{0}$
and closer to
${\Gamma}_{0}$
than is
${\Gamma}_{a}$
. Continual repetitions yield an infinite sequence of galaxies indexed by, say, the negative integers and totally ordered by their closeness to
${\Gamma}_{0}$
.
The conclusion that there is an infinite sequence of galaxies progressively further away from
${\Gamma}_{0}$
than is
${\Gamma}_{b}$
is easier to prove. With
$\mathbf{v}\in {\Gamma}_{b}$
as before, we have that, for every
$m\in IN$
,
$\{n:d(x,{v}_{n})>m\}\in \mathcal{\mathcal{F}}$
. Therefore, for each
$n\in IN$
, we can choose
${w}_{n}$
as an element of
$\langle {v}_{n}\rangle $
such that
$$\begin{array}{c}d(x,{w}_{n})\ge d(x,{v}_{n})+n.\end{array}$$ 
(3)

Hence, for each
$m\in IN$
,
$$\{n:d(x,{w}_{n})>m\}\supseteq \{n:d(x,{v}_{n})>m\}.$$
Since the righthand side is a member of
$\mathcal{\mathcal{F}}$
, so too is the lefthand side. Thus,
$\mathbf{w}=\left[{w}_{n}\right]$
is in a galaxy different from
${\Gamma}_{0}$
. Moreover, from ( 3 ) we have that, for each
$m\in IN$
,
$\{n:d(x,{w}_{n})d(x,{v}_{n})>m\}$
is a cofinite set and therefore is a member of
$\mathcal{\mathcal{F}}$
. Consequently,
$\mathbf{w}$
is in a galaxy
${\Gamma}_{c}$
that is further away from
${\Gamma}_{0}$
than is
${\Gamma}_{b}$
.
We can repeat the argument of the last paragraph with
${\Gamma}_{c}$
in place of
${\Gamma}_{b}$
to find still another galaxy
${\Gamma}_{c}^{\prime}$
further away from
${\Gamma}_{0}$
than is
${\Gamma}_{c}$
. Repetitions of this argument show that there is an infinite sequence of galaxies indexed by, say, the positive integers and totally ordered by their closeness to
${\Gamma}_{0}$
.
The conjunction of the two infinite sequences along with
${\Gamma}_{b}$
yields the conclusion of the theorem.
$\square $
By virtue of Theorem 3.7, the conclusion of Theorem 4.2 holds whenever
$G$
is locally finite.
In general, the hypothesis of Theorem 4.2 may or may not hold. Thus,
${}^{*}G$
either has exactly one galaxy, its principal one
${\Gamma}_{0}$
, or has infinitely many galaxies.
Instead of the idea of “totally ordered according to closeness to
${\Gamma}_{0}$
,” we can define the idea of “partially ordered according to closeness to
${\Gamma}_{0}$
” in much the same way. Just drop the connectedness axiom for a total ordering.
Theorem 4.3. Under the hypothesis of Theorem 4.2, the set of galaxies of
${}^{*}G$
is partially ordered according to the closeness of the galaxies to the principal galaxy
${\Gamma}_{0}$
.
Proof. Reflexivity and antisymmetry are obvious. Consider transitivity: Let
${\Gamma}_{a}$
,
${\Gamma}_{b}$
, and
${\Gamma}_{c}$
be galaxies different from
${\Gamma}_{0}$
. (The case where
${\Gamma}_{a}={\Gamma}_{0}$
can be argued similarly.) Assume that
${\Gamma}_{a}$
is closer to
${\Gamma}_{0}$
than is
${\Gamma}_{b}$
and that
${\Gamma}_{b}$
is closer to
${\Gamma}_{0}$
than is
${\Gamma}_{c}$
. Thus, for any
$\mathbf{x}$
in
${\Gamma}_{0}$
,
$\mathbf{u}$
in
${\Gamma}_{a}$
,
$\mathbf{v}$
in
${\Gamma}_{b}$
, and
$\mathbf{w}$
in
${\Gamma}_{c}$
and for every
$m\in IN$
, we have
$${N}_{uv}=\{n:d({v}_{n},{x}_{n})d({u}_{n},{x}_{n})\ge m\}\in \mathcal{\mathcal{F}}$$
and
$${N}_{vw}=\{n:d({w}_{n},{x}_{n})d({v}_{n},{x}_{n})\ge m\}\in \mathcal{\mathcal{F}}.$$
We also have
$$d({w}_{n},{x}_{n})d({u}_{n},{x}_{n})=d({w}_{n},{x}_{n})d({v}_{n},{x}_{n})+d({v}_{n},{x}_{n})d({u}_{n},{x}_{n}).$$
So,
$${N}_{uw}=\{n:d({w}_{n},{x}_{n})d({u}_{n},{x}_{n})\ge 2m\}\supseteq {N}_{uv}\cap {N}_{vw}\in \mathcal{\mathcal{F}}.$$
Thus,
${N}_{uw}\in \mathcal{\mathcal{F}}$
. Since
$m$
can be chosen arbitrarily, we can conclude that
${\Gamma}_{a}$
is closer to
${\Gamma}_{0}$
than is
${\Gamma}_{c}$
.
$\square $
5 The Hyperordinals
In the following sections, we shall extend the results obtained so far to enlargements of transfinite graphs of rank 1, that is, to enlargements of 1graphs. For this purpose, we need to replace the set
${}^{*}IN$
of hypernaturals by a set of “hyperordinals”; these are defined as follows. A hyperordinal
$\underline{\alpha}$
is an equivalence class of sequences of ordinals where two such sequences
$\langle {\alpha}_{n}\rangle $
and
$\langle {\beta}_{n}\rangle $
are taken to be equivalent if
$\{n:{\alpha}_{n}={\beta}_{n}\}\in \mathcal{\mathcal{F}}$
. We denote
$\underline{\alpha}$
also by
$\left[{\alpha}_{n}\right]$
where again the
${\alpha}_{n}$
are the elements of one (any one) of the sequences in the equivalent class. Any set of hyperordinals is totally ordered by the inequality relation. That is, given any hyperordinals
$\underline{\alpha}=\left[{\alpha}_{n}\right]$
and
$\underline{\beta}=\left[{\beta}_{n}\right]$
, exactly one of the sets:
$$\{n:{\alpha}_{n}<{\beta}_{n}\},\{n:{\alpha}_{n}={\beta}_{n}\},\{n:{\alpha}_{n}>{\beta}_{n}\}$$
will be in
$\mathcal{\mathcal{F}}$
. So, exactly one of the expressions:
$$\underline{\alpha}<\underline{\beta},\underline{\alpha}=\underline{\beta},\underline{\alpha}>\underline{\beta}$$
holds.
6 Walks in 1Graphs
1graphs arise when conventionally infinite graphs are connected at their infinite extremities through 1nodes, the latter being a generalization of the idea of a node. Such 1nodes and the resulting 1graphs are defined in [
5,Section2.1]
and also in [
6,Section2.3]
. Let us restate the needed definitions concisely.
We will be dealing with two kinds of nodes and two kinds of graphs. A conventionally infinite graph
${G}^{0}$
will now be called a 0graph and the nodes in
${G}^{0}$
will be called 0nodes in order to distinguish these ideas from those pertaining to transfinite graphs of rank 1.
Similarly, what we called a “hypernode” previously will henceforth be called a 0
hypernode, and what we called a “galaxy” in the enlargement of a 0graph will now be called a 0galaxy.
An
infinite extremity of a 0graph
${G}^{0}$
is defined as an equivalence class of oneended paths in
${G}^{0}$
, where two such paths are considered to be equivalent if they are eventually identical. Such an equivalence class is called a 0tip of
${G}^{0}$
.
${G}^{0}$
may have one or more 0tips (or possibly none at all). To obtain the “1nodes,” the set of 0tips is partitioned in some fashion into subsets, and to each subset a single 0node may (or may not) be added under the proviso that, if a 0node is added to one subset, it is not added to any other subset.
Then, each subset (possibly augmented with a 0node) is called a 1
node. With
${X}^{1}$
denoting the set of 1nodes and
${X}^{0}$
the set of 0nodes of
${G}^{0}$
, the 1graph
${G}^{1}$
is defined as the triplet:
$${G}^{1}=\{{X}^{0},B,{X}^{1}\},$$
and
${G}^{0}=\{{X}^{0},B\}$
is now called the 0graph of
${G}^{1}$
. Furthermore, a path in
${G}^{0}$
is now called a 0path, and connectedness in
${G}^{0}$
is now called 0connectedness. We will consistently append the superscript 0 to the symbols and the prefix 0to the terminology for concepts from Sections 2 through 4 regarding 0graphs.
In order to define the “1galaxies,” we need the idea of distances in a 1graph
${G}^{1}$
.
But now, we must make a significant choice. The distances between two nodes (0nodes or 1nodes) can be defined as the minimum length of all paths—or, alternatively, of all
walks—connecting the two nodes. It turns out that a path need not exist between two nodes in a 1graph
${G}^{1}$
, but a walk always will exist between them. To ensure the existence of at least one path between every two nodes, additional conditions must be imposed on
${G}^{1}$
(see [
5,Conditions3.21and3.51]
or [
6,Condition3.12]
), and this leads to a more restrictive and yet more complicated theory involving distances. Such can be done, but it is more general and simpler to use walkbased distance ideas. This we now do.
A
nontrivial 0walk
${W}^{0}$
in a 0graph is the conventional concept. It is a (finite or oneway infinite or twoway infinite) alternating sequence:
$$\begin{array}{c}{W}^{0}=\langle \dots ,{x}_{1}^{0},{b}_{1},{x}_{0}^{0},{b}_{0},{x}_{1}^{0},{b}_{1},\dots \rangle \end{array}$$ 
(4)

of 0nodes
${x}_{m}^{0}$
and branches
${b}_{m}$
, where each branch
${b}_{m}$
is incident to the two 0nodes
${x}_{m}^{0}$
and
${x}_{m+1}^{0}$
adjacent to it in the sequence. If the sequence terminates at either side, it is required to terminate at a 0node. The 0walk is called twoended or finite if it terminates on both sides, oneended if it terminates on just one side, and endless if it terminates on neither side.
A
trivial 0walk is a singleton set whose sole element is a 0node.
A oneended 0walk
${W}^{0}$
will be called extended if its 0nodes are eventually distinct, that is, if it is eventually identical to a oneended path. We say that
${W}^{0}$
traverses a 0tip if it is extended and eventually identical to a representative of that 0tip. Finally,
${W}^{0}$
is said to reach a 1node
${x}^{1}$
if
${W}^{0}$
traverses a 0tip contained in
${x}^{1}$
. In the same way, an endless 0walk can reach two 1nodes (or possibly reach the same 1node) by traversing two 0tips, one toward the left and the other toward the right. When this is so, we say that the endless 0walk is extended. On the other hand, if a 0walk terminates at a 0node contained in a 1node, we again say that the 0walk reaches both of those nodes and does so through a branch incident to that 0node.
Every twoended 0walk contains a 0path that terminates at the two 0nodes at which the 0walk terminates, so there is no need to employ 0walks when defining distances in a 0graph. On the other hand, such a need arises for 1graphs. To meet this need, we first define a 0
section
${S}^{0}$
in a 1graph
${G}^{1}$
as a subgraph
${S}^{0}$
of the 0graph
${G}^{0}$
of
${G}^{1}$
induced by a maximal set of branches that are pairwise 0connected in
${G}^{0}$
. A 1node
${x}^{1}$
is said to be incident to
${S}^{0}$
if either it contains a 0node incident to a branch of
${S}^{0}$
or it contains a 0tip having a representative oneended path lying entirely within
${S}^{0}$
. In this case, we also say that that 0tip belongs to
${S}^{0}$
. Given two 1nodes
${x}^{1}$
and
${y}^{1}$
incident to
${S}^{0}$
, there will be a 0walk
${W}^{0}$
in
${S}^{0}$
that reaches each of
${x}^{1}$
and
${y}^{1}$
through a 0tip belonging to
${S}^{0}$
or through a branch in
${S}^{0}$
.^{
$4$
}
Moreover, there may also be a 0walk
${W}^{0}$
in
${S}^{0}$
that reaches the same 1node at both extremities of
${W}^{0}$
. To be more specific, let us state Lemma 6.1. Let
${S}^{0}$
be a 0section in
${G}^{1}$
, and let
${x}^{1}$
and
${y}^{1}$
be two 1nodes incident to
${S}^{0}$
. Then, there exists a 0walk in
${S}^{0}$
that reaches
${x}^{1}$
and
${y}^{1}$
.
Proof. That
${x}^{1}$
is incident to
${S}^{0}$
means that there is a 0path
${P}_{x}^{0}$
in
${S}^{0}$
that either reaches
${x}^{1}$
through a 0tip of
${x}^{1}$
or reaches
${x}^{1}$
through a branch. Similarly, there is such a 0path
${P}_{y}^{0}$
reaching
${y}^{1}$
. Let
${u}^{0}$
be a node of
${P}_{x}^{0}$
, and let
${v}^{0}$
be a node of
${P}_{y}^{0}$
. Since
${S}^{0}$
is 0connected, there is a 0path
${P}_{uv}^{0}$
in
${S}^{0}$
terminating at
${u}^{0}$
and
${v}^{0}$
(possibly a trivial 0path if
${u}^{0}={v}^{0}$
). Then,
${P}_{x}^{0}\cup {P}_{uv}^{0}\cup {P}_{y}^{0}$
as a 0walk in
${S}^{0}$
as asserted.
$\square $
A nontrivial, twoended 1walk
${W}^{1}$
is a finite sequence:
$$\begin{array}{c}{W}^{1}=\langle {x}_{0},{W}_{0}^{0},{x}_{1}^{1},{W}_{1}^{0},\dots ,{x}_{m1}^{1},{W}_{m1}^{0},{x}_{m}\rangle \end{array}$$ 
(5)

with
$m\ge 1$
that satisfies the following conditions.

1.
${x}_{1}^{1},\dots ,{x}_{m1}^{1}$
are 1nodes, while
${x}_{0}$
and
${x}_{m}$
may be either 0nodes or 1nodes.

2. For each
$k=0,\dots ,m1$
,
${W}_{k}^{0}$
is a nontrivial 0walk that reaches the two nodes adjacent to it in the sequence.

3. For each
$k=1,\dots ,m1$
, at least one of
${W}_{k1}^{0}$
and
${W}_{k}^{0}$
reaches
${x}_{k}^{1}$
through a 0tip, not through a branch. Also, if
$m=1$
,
${W}_{0}^{0}$
reaches at least one of
${x}_{0}$
and
${x}_{1}$
through a 0tip.
A oneended 1walk is a sequence like ( 5 ) except that it extends infinitely to the right. An endless 1walk extends infinitely on both sides. A trivial 1walk is a singleton set whose sole element is either a 0node or a 1node.
We now define a more general kind of connectedness (called “1wconnectedness” to distinguish it from pathbased 1connectedness). Two branches (resp. two nodes—either 0nodes or 1nodes) will be said to be 1
wconnected if there exists a 0walk or 1walk that terminates at a 0node of each branch (resp. that terminates at those two nodes). If a terminal node of a walk is the same as, or contains, or is contained in the terminal node of another walk, the two walks taken together form another walk. We call this the conjunction of the two walks. It follows that 1wconnectedness is a transitive binary relation for the branch set
$B$
of the 1graph
${G}^{1}$
and is in fact an equivalence relation. If every two branches of
${G}^{1}$
are 1wconnected, we will say that
${G}^{1}$
is 1wconnected.
7 WalkBased Distances in a 1Graph
The length
$\left{W}^{0}\right$
of a 0walk
${W}^{0}$
is defined as follows: If
${W}^{0}$
is twoended,
$\left{W}^{0}\right$
is the number
${\tau}_{0}$
of branch traversals in it; that is, each branch is counted as many times as it appears in
${W}^{0}$
. If
${W}^{0}$
is oneended and extended, we set
$\left{W}^{0}\right=\omega $
, the first transfinite ordinal. If
${W}^{0}$
is endless and extended in both directions, we set
$\left{W}^{0}\right=\omega \cdot 2$
.
As for a nontrivial twoended 1walk
${W}^{1}$
, its length
$\left{W}^{1}\right$
is taken to be
$\left{W}^{1}\right={\sum}_{k=0}^{m}\left{W}_{k}^{0}\right$
, where the sum is over the finitely many 0walks
${W}_{k}^{0}$
in ( 5 ). Thus,
$$\begin{array}{c}\left{W}^{1}\right=\omega \cdot {\tau}_{1}+{\tau}_{0}\end{array}$$ 
(6)

where
${\tau}_{1}$
is the number of traversals of 0tips performed by
${W}^{1}$
and
${\tau}_{0}$
is the number of traversals of branches in all the twoended (i.e., finite) 0walks appearing as terms in ( 5 ).
We take
${\sum}_{k=0}^{m}\left{W}_{k}^{0}\right$
to be the natural sum of ordinals; this yields a normal expansion of an ordinal [
1,pages354355]
.
${\tau}_{1}$
is not 0 because
${W}^{1}$
is a nontrivial, twosided 1walk. However,
${\tau}_{0}$
may be 0, this occurring when every
${W}_{k}^{0}$
in ( 5 ) is oneended or endless.
A 0node is called
maximal if it is not contained in a 1node, and nonmaximal otherwise.
A distance measured from a nonmaximal 0node is the same as that measured from the 1node containing it. Given two nodes
$x$
and
$y$
(of ranks 0 or 1), we define the wdistance^{
$5$
}
$d(x,y)$
between them as
$$\begin{array}{c}d(x,y)=min\left{W}_{x,y}\right\end{array}$$ 
(7)

where the minimum is taken over all twoended walks (0walks or 1walks) terminating at
$x$
and
$y$
. That minimum exists because any set of ordinals is a wellordered set. In view of ( 6 ),
$d(x,y)<{\omega}^{2}$
. If
$x=y$
, we set
$d(x,x)=0$
.
Clearly, if
$x\ne y$
,
$d(x,y)>0$
and
$d(x,y)=d(y,x)$
. Furthermore, the conjunction of two twoended walks is again a twoended walk, whose length is the natural sum of the ordinal lengths of the two walks. So, by taking minimums appropriately, we obtain the triangle inequality:
$$\begin{array}{c}d(x,z)\le d(x,y)+d(y,z)\end{array}$$ 
(8)

where again the natural sum of ordinals is understood. Altogether then, we have Lemma 7.1. The ordinalvalued wdistances between the maximal nodes of a 1graph satisfy the metric axioms.
8 Enlargements of 1Graphs and Hyperdistances in Them
In [
6,pages163164]
, a nonstandard 1node was defined as an equivalence class of sequences of sets of tips shorted together, with the tips taken from sequences of possibly differing 1graphs. But, since each set of tips shorted together is a 1node, that definition of a nonstandard 1node can also be stated as an equivalence class of sequences of 1nodes.
Specializing to the case where all the 1graphs are the same, we have the following definition of a nonstandard 1node, which we now call a “1hypernode.” Consider a given 1graph along with a chosen free ultrafilter
$\mathcal{\mathcal{F}}$
. Two sequences
$\langle {x}_{n}^{1}\rangle $
and
$\langle {y}_{n}^{1}\rangle $
of 1nodes in
${G}^{1}$
are taken to be equivalent if
$\{n:{x}_{n}^{1}={y}_{n}^{1}\}\in \mathcal{\mathcal{F}}$
. It is easy to show that this is truly an equivalence relation. Then,
${\mathbf{x}}^{1}=\left[{x}_{n}^{1}\right]$
denotes one such equivalence class, where the
${x}_{n}^{1}$
are the elements of any one of the sequences in that class.
${\mathbf{x}}^{1}$
will be called a 1hypernode.
The
enlargement of the 1graph
${G}^{1}=\{{X}^{0},B,{X}^{1}\}$
is the nonstandard 1graph
$${}^{*}{G}^{1}={\{}^{*}{X}^{0}{,}^{*}B{,}^{*}{X}^{1}\}$$
where
${}^{*}{X}^{0}$
and
${}^{*}B$
are respectively the set of 0hypernodes and branches in the enlargement of the 0graph
${G}^{0}=\{{X}^{0},B\}$
of
${G}^{1}$
and
${}^{*}{X}^{1}$
is the set of 1hypernodes defined above, that is, the set of all equivalence classes of sequences of 1nodes taken from
${X}^{1}$
.
We define the
hyperdistance
$\mathbf{d}$
between any two hypernodes
$\mathbf{x}$
and
$\mathbf{y}$
of
${}^{*}{G}^{1}$
(of ranks 0 and/or 1) to be the internal function
$$\begin{array}{c}\mathbf{d}(\mathbf{x},\mathbf{y})=\left[d\right({x}_{n},{y}_{n}\left)\right].\end{array}$$ 
(9)

Since distances in
${G}^{1}$
are less than
${\omega}^{2}$
,
$\mathbf{d}(\mathbf{x},\mathbf{y})$
is a hyperordinal less than
${\underline{\omega}}^{2}$
. We say that a 0hypernode
${\mathbf{x}}^{0}=\left[{x}_{n}^{0}\right]$
is maximal if the set of
$n$
for which
${x}_{n}^{0}$
is not contained in a 1node is a member of
$\mathcal{\mathcal{F}}$
. All the 1nodes in this work are perforce maximal because there are no nodes of higher rank.
$\mathbf{d}$
, when restricted to the maximal hypernodes, also satisfies the metric axioms, in particular, the triangle inequality:
$$\begin{array}{c}\mathbf{d}(\mathbf{x},\mathbf{z})\le \mathbf{d}(\mathbf{x},\mathbf{y})+\mathbf{d}(\mathbf{y},\mathbf{z})\end{array}$$ 
(10)

But, now
$\mathbf{d}$
is hyperordinalvalued.
9 The Galaxies of
${}^{*}{G}^{1}$
The 0galaxies of
${}^{*}{G}^{1}$
are defined just as they are for the enlargement
${}^{*}{G}^{0}$
of a 0graph; see Section 3. However, we henceforth write “0galaxy” in place of “galaxy” and “0limitedly distant” in place of “limitedly distant.” As was mentioned above, each 0section of
${G}^{1}$
is the subgraph of the 0graph
${G}^{0}\{{X}^{0},B\}$
of
${G}^{1}$
induced by a maximal set of branches that are 0connected. A 0section is a 0graph by itself. So, within the enlargement
${}^{*}{G}^{1}$
, each 0section
${S}^{0}$
enlarges into
${}^{*}{S}^{0}$
as defined in Section 2. Within each enlarged 0section there may be one or more 0galaxies. As a special case, a particular 0section may have only finitely many 0nodes, and so its enlargement is itself—all its 0hypernodes are standard. On the other hand, there may be infinitely many 0galaxies in some enlarged 0section. Moreover, the enlarged 0sections do not, in general, comprise all of the enlarged 0graph
${}^{*}{G}^{0}={\{}^{*}{X}^{0}{,}^{*}B\}$
of
${}^{*}{G}^{1}$
. Indeed, there can be a 0hypernode
${\mathbf{x}}^{0}=\left[{x}_{n}^{0}\right]$
where each
${x}_{n}^{0}$
resides in a different 0section; in this case
${\mathbf{x}}^{0}$
will reside in a 0galaxy that is not in an enlargement of a 0section.
Something more can happen with regard to the 0galaxies in
${}^{*}{G}^{1}$
. 0galaxies can now contain 1hypernodes. For example, this occurs when a 1node
${x}^{1}$
is incident to a 0section
${S}^{0}$
through a branch. Then, the standard 1hypernode
${\mathbf{x}}^{1}$
corresponding to
${x}^{1}$
is 0limitedly distant from the standard 0hypernodes in
${}^{*}{S}^{0}$
. So, there is a 0galaxy containing not only
${}^{*}{S}^{0}$
but
${\mathbf{x}}^{1}$
as well. See Example 9.3 below in this regard. In general, the nodal 0galaxies partition the set
${}^{*}{X}^{0}{\cup}^{*}{X}^{1}$
of all the hypernodes in
${}^{*}{G}^{1}$
. As we shall see in Examples 9.1 and 9.2 below, there may be a singleton 0galaxy containing a 1hypernode only.
Let us now turn to the “1galaxies” of
${}^{*}{G}^{1}$
Two hypernodes
$\mathbf{x}=\left[{x}_{n}\right]$
and
$\mathbf{y}=\left[{y}_{n}\right]$
(of ranks 0 and/or 1) in
${}^{*}{G}^{1}$
will be said to be in the same nodal 1galaxy
${\dot{\Gamma}}^{1}$
if there exists a natural number
$k\in IN$
depending on the choices of
$\mathbf{x}$
and
$\mathbf{y}$
such that
$\{n:d({x}_{n},{y}_{n})\le \omega \cdot k\}\in \mathcal{\mathcal{F}}$
. In this case, we say that
$\mathbf{x}$
and
$\mathbf{y}$
are 1limitedly distant, and we write
$\mathbf{d}(\mathbf{x},\mathbf{y})\le [\omega \cdot k]$
where
$[\omega \cdot k]$
denotes the standard hyperordinal corresponding to
$\omega \cdot k$
. This defines an equivalence relation on the set
${}^{*}{X}^{0}{\cup}^{*}{X}^{1}$
of all the hypernodes in
${}^{*}{G}^{1}$
. Indeed, reflexivity and symmetry are obvious. For transitivity, assume that
$\mathbf{x}$
and
$\mathbf{y}$
are 1limitedly distant and that
$\mathbf{y}$
and
$\mathbf{z}$
are 1limitedly distant, too. Then, there are natural numbers
${k}_{1}$
and
${k}_{2}$
such that
$${N}_{xy}=\{n:d({x}_{n},{y}_{n})\le \omega \cdot {k}_{1}\}\in \mathcal{\mathcal{F}}$$
and
$${N}_{yz}=\{n:d({y}_{n},{z}_{n})\le \omega \cdot {k}_{2}\}\in \mathcal{\mathcal{F}}.$$
By the triangle inequality ( 8 ),
$${N}_{xz}=\{n:d({x}_{n},{z}_{n})\le \omega \cdot ({k}_{1}+{k}_{2}\left)\right\}\supseteq {N}_{xy}\cap {N}_{yz}\in \mathcal{\mathcal{F}}.$$
So,
${N}_{xz}\in \mathcal{\mathcal{F}}$
and therefore
$\mathbf{x}$
and
$\mathbf{z}$
are 1limitedly distant. We can conclude that the set
${}^{*}{X}^{0}{\cup}^{*}{X}^{1}$
of all hypernodes in
${}^{*}{G}^{1}$
is partitioned into nodal 1galaxies by this equivalence relation.
Corresponding to each nodal 1galaxy
${\dot{\Gamma}}^{1}$
, we define a 1galaxy
${\Gamma}^{1}$
as a nonstandard subgraph of
${}^{*}{G}^{1}$
consisting of all the hypernodes in
${\dot{\Gamma}}^{1}$
along with all the hyperbranches both of whose 0hypernodes are in
${\dot{\Gamma}}^{1}$
.
No hyperbranch can have its two incident 0hypernodes in two different 0galaxies or two different 1galaxies because the distance between their 0hypernodes is 1. Thus, the hyperbranch set
${}^{*}B$
is also partitioned by the 0galaxies and more coarsely by the 1galaxies.
The
principal 1galaxy
${\Gamma}_{0}^{1}$
of
${}^{*}{G}^{1}$
is the 1galaxy whose hypernodes are 1limitedly distant from a standard hypernode in
${}^{*}{G}^{1}$
(i.e., from a node of
${G}^{1}$
).
Note that the enlargement
${}^{*}{S}^{0}$
of each 0section
${S}^{0}$
of
${G}^{1}$
has its own principal 0galaxy
${\Gamma}_{0}^{0}\left({S}^{0}\right)$
. Moreover, every
${}^{*}{S}^{0}$
lies within the principal 1galaxy
${\Gamma}_{0}^{1}$
. Indeed, any standard hypernode
$\mathbf{x}$
by which
${\Gamma}_{0}^{1}$
may be defined and any standard 0hypernode
${\mathbf{y}}^{0}$
by which
${\Gamma}_{0}^{0}\left({S}^{0}\right)$
may be defined are 1limitedly distant. Also, the hyperdistance
$\mathbf{d}({\mathbf{y}}^{0},{\mathbf{z}}^{0})$
between any two 0hypernodes
${\mathbf{y}}^{0}$
and
${\mathbf{z}}^{0}$
of
${}^{*}{S}^{0}$
is no larger than a hypernatural
$\mathbf{k}$
. So, by the triangle inequality ( 10 ), every 0hypernode of
${}^{*}{S}^{0}$
is 1limitedly distant from
$\mathbf{x}$
. Whence our assertion.
Example 9.1. Consider an endless 1path
${P}^{1}$
having an endless 0path between every consecutive pair of 1nodes in
${P}^{1}$
. The 0sections of
${P}^{1}$
are those endless 0paths, and each of their enlargements have an infinity of 0galaxies in
${}^{*}{P}^{1}$
. However, there are other 0galaxies in
${}^{*}{P}^{1}$
, infinitely many of them. Indeed, consider a 0hypernode
${\mathbf{x}}^{0}=\left[{x}_{n}^{0}\right]$
, where each 0node
${x}_{n}^{0}$
lies in a different 0section of
${P}^{1}$
;
${\mathbf{x}}^{0}$
will lie in a 0galaxy
${\Gamma}_{1}^{0}$
different from all the 0galaxies in any enlargement of a 0section of
${P}^{1}$
. The 0hypernodes of
${\Gamma}_{1}^{0}$
will be the 0hypernodes that are 0limitedly distant from
${\mathbf{x}}^{0}$
. Furthermore, there are still other 0galaxies now. Each 1hypernode
${\mathbf{x}}^{1}=\left[{x}_{n}^{1}\right]$
is the sole member of a 0galaxy. In fact, the nodal 0galaxies partition the set of all the 0hypernodes and 1hypernodes.
On the other hand, the principal 1galaxy of
${}^{*}{P}^{1}$
consists of all the standard 1hypernodes corresponding to the 1nodes of
${P}^{1}$
along with the enlargements of the 0sections of
${P}^{1}$
. Also, there will be infinitely many 1galaxies, each of which contains infinitely many 0galaxies along with 1hypernodes. In this particular case, each of the 1galaxies is graphically isomorphic to the principal 1galaxy, but this is not true in general.
$\square $
Example 9.2. An example of a nonstandard 1graph
${}^{*}{G}^{1}$
having exactly one 1galaxy (its principal one) and infinitely many 0galaxies is provided by the enlargement of the 1graph
${G}^{1}$
obtained from the 0graph of Figure 1 by replacing each branch by an endless 0path, thereby converting each 0node into a 1node. Again each endless path of that 1graph
${G}^{1}$
is a 0section, and its enlargement is like that of Example 3.2. There are infinitely many such 0galaxies in each enlargement of a 0section. Also, there are infinitely many 0galaxies, each consisting of a single 1hypernode. With regard to the 1galaxies, the enlargement
${}^{*}{G}^{1}$
of
${G}^{1}$
mimics that of Example 3.4, except that now the rank 0 is replaced by the rank 1. The hyperdistance between every two 1hypernodes (resp. 0hypernodes) is no larger than
$\omega \cdot 4$
(resp.
$\omega \cdot 6$
). Hence,
${}^{*}{G}^{1}$
has only one 1galaxy, its principal one.
$\square $
Example 9.3. Here is an example where each of the 1hypernodes is not isolated as the sole member within a 0galaxy. Replace each of the horizontal branches in Figure 1 by an endless 0path, but do not alter the branches incident to
${x}_{g}$
. Now, the nodes
${x}_{k}$
$(k=0,1,2,\dots )$
become 1nodes
${x}_{k}^{1}$
, each containing a 0node of the branch incident to
${x}_{k}^{1}$
and
${x}_{g}$
. Each 1hypernode is 0limitedly distant from the standard 0hypernode
${\mathbf{x}}_{g}$
and thus is not so isolated. The 1hypernodes (whether standard or nonstandard) along with the standard 0hypernode for
${x}_{g}$
and the (standard or nonstandard) hyperbranches incident to
${\mathbf{x}}_{g}$
all comprise a single 0galaxy. Moreover, there will be other 0galaxies obtained through equivalence classes of sequences of these nodes and branches. In fact, each of the endless paths that replace the horizontal branches yield infinitely many 0galaxies. Again, the nodal 0galaxies partition the set of all the hypernodes in
${}^{*}{G}^{1}$
.
On the other hand, there is again only one 1galaxy for
${}^{*}{G}^{1}$
.
$\square $
Example 9.4. The distances in the three preceding examples can be fully defined by paths. So, let us now present an example where walks are needed. The 1graph
${G}^{1}$
of Figure 3 illustrates one such case. It consists of an infinite sequence of 0subgraphs, each of which is an infinite series connections of fourbranch subgraphs, each in a diamond configuration, as shown. To save words, we shall refer to such an infinite series connection as a “chain.” The chain starting at the 0node
${x}_{k}^{0}$
will be denoted by
${C}_{k}$
$(k=0,1,2,\dots )$
. Each
${C}_{k}$
is a 0graph; it does not contain any 1node. Each
${C}_{k}$
has uncountably many 0tips. One 0tip has a representative 0path starting at
${x}_{k}^{0}$
, proceeding along the lefthand sides of the diamond configurations, and reaching the 1node
${x}_{k}^{1}$
. Another 0tip has a representative 0path that proceeds along the righthand sides and reaches the 1node
${x}_{k+1}^{1}$
. Still other 0tips of
${C}_{k}$
(uncountably many of them) have representatives that pass back and forth between the two sides infinitely often to reach singleton 1nodes; these are not shown in that figure. The chain
${C}_{k}$
is connected to
${C}_{k+1}$
through the 1node
${x}_{k+1}^{1}$
, as shown. Note that there is no path connecting, say,
${x}_{k}^{0}$
to
${x}_{m}^{0}$
when
$mk\ge 2$
, but there is such a walk.
Each
${C}_{k}$
is a 0section, and its enlargement
${}^{*}{C}_{k}$
has infinitely many 0galaxies. Also, the 1nodes
${x}_{k}^{1}$
together produce infinitely many 0galaxies, each being a single 1hypernode.
As before, the nodal 0galaxies comprise a partition of
${}^{*}{X}^{0}{\cup}^{*}{X}^{1}$
.
On the other hand, the enlargement
${}^{*}{G}^{1}$
of the 1graph
${G}^{1}$
of Figure 3 has infinitely many 1galaxies. Its principal one is a copy of
${G}^{1}$
. Each of the other 1galaxies is also a copy of
${G}^{1}$
except that it extends infinitely in both directions—infinitely to the left and infinitely to the right. Here, too, the nodal 1galaxies comprise a partitioning of
${}^{*}{X}^{0}{\cup}^{*}{X}^{1}$
, but a coarser one.
$\square $
These examples indicate that the enlargements of 1graphs can have rather complicated structures.
10 Locally 1Finite 1Graphs and a Property of Their Enlargements
In general,
${}^{*}{G}^{1}$
has 1galaxies other than its principal 1galaxy. One circumstance where this occurs is when
${}^{*}{G}^{1}$
is locally finite in certain way, which we will explicate below.
We need some more definitions. Two 1nodes of
${G}^{1}$
are said to be 1wadjacent if they are incident to the same 0section. A 1node will be called a boundary 1node if it is incident to two or more 0sections.
${G}^{1}$
will be called locally 1finite if each of its 0sections has only finitely many incident boundary 1nodes.^{
$6$
}
Lemma 10.1. Let
${x}^{1}$
be a boundary 1node. Then, any 1walk that passes through
${x}^{1}$
from any 0section
${S}_{1}^{0}$
incident to
${x}^{1}$
to any other 0section
${S}_{2}^{0}$
incident to
${x}^{1}$
must have a length no less than
$\omega $
.
Proof. The only way such a walk can have a length less than
$\omega $
(i.e., a length equal to a natural number) is if it avoids traversing a 0tip in
${x}^{1}$
. But, this means that it passes through two branches incident to a 0node in
${x}^{1}$
. But, that in turn means that
${S}_{1}^{0}$
and
${S}_{2}^{0}$
cannot be different 0sections.
$\square $
Remember that
${G}^{1}$
is called 1wconnected if, for every two nodes of
${G}^{1}$
, there is a 0walk or 1walk that reaches those two nodes.
Lemma 10.2. Any two 1nodes
${x}^{1}$
and
${y}^{1}$
that are 1wconnected but are not 1wadjacent must satisfy
$d({x}^{1},{y}^{1})\ge \omega $
.
Proof. Any walk 1wconnecting
${x}^{1}$
and
${y}^{1}$
must pass through at least one boundary 1node different from
${x}^{1}$
and
${y}^{1}$
while passing from one 0section to another 0section.
Therefore, that walk must be a 1walk. By Lemma 10.1, its length is no less than
$\omega $
. Since this is true for every such walk, our conclusion follows.
$\square $
The next theorem mimics Theorem 3.7 but at the rank 1.
Theorem 10.3. Let
${G}^{1}$
be locally 1finite and 1wconnected and have infinitely many boundary 1nodes. Then, given any 1node
${x}_{0}^{1}$
of
${G}^{1}$
, there is a oneended 1walk
${W}^{1}$
starting at
${x}_{0}^{1}$
:
$${W}^{1}=\langle {x}_{0}^{1},{W}_{0}^{0},{x}_{1}^{1},{W}_{1}^{0},\dots ,{x}_{m}^{1},{W}_{m}^{0},\dots \rangle $$
such that there is a subsequence of 1nodes
${x}_{{m}_{k}}^{1}$
,
$k=1,2,3,\dots $
, satisfying
$d({x}_{0}^{1},{x}_{{m}_{k}}^{1})\ge \omega \cdot k$
.
Proof.
${x}_{0}^{1}$
need not be a boundary 1node, but it will be 1wadjacent to only finitely many boundary 1nodes because of local 1finiteness and 1wconnectedness. Let
${X}_{0}$
be the nonempty finite set of those boundary 1nodes. For the same reasons, there is a nonempty finite set
${X}_{1}$
of boundary 1nodes, each being 1wadjacent to some 1node in
${X}_{0}$
but not 1wadjacent to
${x}_{0}^{1}$
. By Lemma 10.2, for each
${x}^{1}\in {X}_{2}$
, we have
$d({x}_{0}^{1},{x}^{1})\ge \omega $
. In general, for each
$k\in IN$
,
$k\ge 2$
, there is a nonempty finite set
${X}_{k}$
of boundary 1nodes, each being 1wadjacent to some 1node in
${X}_{k1}$
but not 1wadjacent to any of the 1nodes in
${\cup}_{l=0}^{k2}{X}_{l}$
.
By Lemma 10.2 again, for any such
${x}^{1}\in {X}_{k}$
, we have
$d({x}_{0}^{1},{x}^{1})\ge \omega \cdot k$
.
We now adapt the proof of König's lemma: From each of the infinitely any boundary 1nodes in
${G}^{1}$
, there is a 1walk reaching that boundary 1node and also reaching
${x}_{0}^{1}$
. Thus, there are infinitely many 1walks starting at
${x}_{0}^{1}$
and passing through one of the 1nodes in
${X}_{0}$
, say,
${x}_{{m}_{0}}^{1}$
. Among those 1walks, there are again infinitely many 1walks passing through one of the 1nodes in
${X}_{1}$
, say,
${x}_{{m}_{1}}^{1}$
. Continuing in this say, we find an infinite sequence
$\langle {x}_{{m}_{1}}^{1},{x}_{{m}_{2}}^{1},{x}_{{m}_{3}}^{1},\dots \rangle $
of 1nodes occurring in a oneended 1walk starting at
${x}_{0}^{1}$
and such that
$d({x}_{0}^{1},{x}_{{m}_{k}}^{1})\ge \omega \cdot k$
.
$\square $
Corollary 10.4. Under the hypothesis of Theorem 10.3, the enlargement
${}^{*}{G}^{1}$
of
${G}^{1}$
has at least one 1hypernode not in its principal galaxy
${\Gamma}_{0}^{1}$
and thus at least one 1galaxy
${\Gamma}^{1}$
different from its principal 1galaxy
${\Gamma}_{0}^{1}$
.
Proof. Set
${\mathbf{x}}^{1}=[\langle {x}_{0}^{1},{x}_{{m}_{0}}^{1},{x}_{{m}_{1}}^{1},\dots \rangle ]=\left[{x}_{{m}_{n}}^{1}\right]$
, where the
${x}_{{m}_{n}}^{1}$
are the 1nodes specified in the preceding proof (replace
$k$
by
$n$
). With
${\mathbf{x}}_{0}^{1}$
being the standard 1hypernode corresponding to
${x}_{0}^{1}$
, we have by Theorem 10.3 that
$\mathbf{d}({\mathbf{x}}_{0}^{1},{\mathbf{x}}^{1})\ge [\omega \cdot n]$
. Hence,
${\mathbf{x}}^{1}$
is not 1limitedly distant from
${\mathbf{x}}_{0}^{1}$
and thus must reside in a 1galaxy
${\Gamma}^{1}$
different from
${\Gamma}_{0}^{1}$
.
$\square $
11 When
${}^{*}{G}^{1}$
Has a 1Hypernode Not in Its Principal Galaxy
We are at last ready to extend the results of Section 4 to the rank 1 of transfiniteness.
The ideas are the much same as those of Section 4 except for the fact that the proof of Theorem 4.2 cannot be extended to the present case. This is because transfinite ordinals cannot be identified as being even or odd. We need another proof.
In this section
${G}^{1}$
is 1wconnected and has an infinity of boundary 1nodes, but
${G}^{1}$
need not be locally finite. Let
${\Gamma}_{a}^{1}$
and
${\Gamma}_{b}^{1}$
be two 1galaxies of
${}^{*}{G}^{1}$
that are different from the principal 1galaxy
${\Gamma}_{0}^{1}$
. We say that
${\Gamma}_{a}^{1}$
is closer to
${\Gamma}_{0}^{1}$
than is
${\Gamma}_{b}^{1}$
and that
${\Gamma}_{b}^{1}$
is further away from
${\Gamma}_{0}^{1}$
than is
${\Gamma}_{a}^{1}$
if there are a
$\mathbf{y}=\left[{y}_{n}\right]$
in
${\Gamma}_{a}^{1}$
and a
$\mathbf{z}=\left[{z}_{n}\right]$
in
${\Gamma}_{b}^{1}$
such that, for some
$\mathbf{x}=\left[{x}_{n}\right]$
in
${\Gamma}_{0}^{1}$
and for every
$m\in IN$
,
$$\{n:d({z}_{n},{x}_{n})d({y}_{n},{x}_{n})\ge \omega \cdot m\}\in \mathcal{\mathcal{F}}.$$
(The ranks of
$\mathbf{x}$
,
$\mathbf{y}$
, and
$\mathbf{z}$
may now be either 0 or 1.) Any set of 1galaxies for which every two of them, say,
${\Gamma}_{a}^{1}$
and
${\Gamma}_{b}^{1}$
satisfy these conditions will be said to be totally ordered according to their closeness to
${\Gamma}_{0}^{1}$
. Here, too, the conditions for a total ordering are readily shown.
Lemma 11.1. These definitions are independent of the representative sequences
$\langle {x}_{n}\rangle $
,
$\langle {y}_{n}\rangle $
, and
$\langle {z}_{n}\rangle $
chosen for
$\mathbf{x}$
,
$\mathbf{y}$
, and
$\mathbf{z}$
.
The proof of this lemma is the same as that of Lemma 4.1 except that the rank 0 is replaced by the transfinite rank 1. For instance, the natural number
$m$
is now replaced by the ordinal
$\omega \cdot m$
.
Theorem 11.2. If
${}^{*}{G}^{1}$
has a hypernode
$\mathbf{v}=\left[{v}_{n}\right]$
(of either rank 0 or rank 1) that is not in its principal 1galaxy
${\Gamma}_{0}^{1}$
, then there exists a twoway infinite sequence of 1galaxies totally ordered according to their closeness to
${\Gamma}_{0}^{1}$
and with
$\mathbf{v}$
being in one of those galaxies.
Proof. In this proof, we use the fact that between any two nodes in a 1graph there exists a geodesic walk terminating at those nodes; that is, the length of the walk is equal to the wdistance between those nodes. This is a consequence of the facts that the walks terminating at those nodes have ordinal lengths and that any set of ordinals is wellordered and thus has a least ordinal. That least ordinal must be the length of at least one walk terminating at those nodes, for otherwise the minimum of the walklengths would be larger.
As before, let
$\mathbf{x}=[\langle x,x,x,\dots \rangle ]$
be a standard hypernode in
${\Gamma}_{0}^{1}$
.
$\mathbf{x}$
can be of either rank 0 or 1. Since
$\mathbf{v}$
is not in
${\Gamma}_{0}^{1}$
, we have that for each
$m\in IN$
$$\begin{array}{c}\{n:d(x,{v}_{n})>\omega \cdot m\}\in \mathcal{\mathcal{F}}\end{array}$$ 
(11)

For each
$n\in IN$
, if
$d(x,{v}_{n})<\omega \cdot 6$
, set
${u}_{n}=x$
, but, if
$d(x,{v}_{n})\ge \omega \cdot 6$
, choose
${u}_{n}$
such that
$$\begin{array}{c}d(x,{v}_{n})\le d(x,{u}_{n})\cdot 3\le d(x,{v}_{n})\cdot 2\end{array}$$ 
(12)

That the latter can be done can be seen as follows.
Choose a geodesic 1walk
${W}^{1}$
terminating at
$x$
and
${v}_{n}$
. Remember that
${W}^{1}$
, as given by ( 5 ), is incident to each of its nonterminal 1nodes through at least one 0tip, as was asserted by Condition 3 of the definition of
${W}^{1}$
. Moreover, the transition through each 0tip contributes
$\omega $
to the length of
${W}^{1}$
. Upon tracing
${W}^{1}$
from
$x$
toward
${v}_{n}$
, we must encounter at least two 1nodes, both of which are neither closer to
$x$
by onethird of the number of 0tips traversed by
${W}^{1}$
nor further away from
$x$
by twothirds of the number of 0tips traversed by
${W}^{1}$
. A node on
${W}^{1}$
between those two 1nodes can be chosen as
${u}_{n}$
.
Suppose there is a
$k\in IN$
such that
$\{n:d(x,{u}_{n})\le \omega \cdot k\}\in \mathcal{\mathcal{F}}$
. By the lefthand inequality of ( 12 ),
$$\{n:d(x,{v}_{n})\le (\omega \cdot k)\cdot 3\}\supseteq \{n:d(x,{u}_{n})\le \omega \cdot k\}\in \mathcal{\mathcal{F}}.$$
Hence, the lefthand set is a member of
$\mathcal{\mathcal{F}}$
, in contradiction to ( 11 ). (These sets cannot both be in the ultrafilter
$\mathcal{\mathcal{F}}$
.) Therefore,
$\mathbf{u}=\left[{u}_{n}\right]$
satisfies ( 11 ) for every
$m\in IN$
when
${v}_{n}$
is replaced by
${u}_{n}$
; that is,
$\mathbf{u}$
is in a galaxy different from the principal 1galaxy
${\Gamma}_{0}^{1}$
.
Furthermore, by the righthand inequality of (
12 ),
$$d(x,{u}_{n})\le \left(d\right(x,{v}_{n})d(x,{u}_{n}\left)\right)\cdot 2.$$
Suppose there exists a
$j\in IN$
such that
$$\{n:d(x,{v}_{n})d(x,{u}_{n})\le \omega \cdot j\}\in \mathcal{\mathcal{F}}.$$
Then,
$$\{n:d(x,{u}_{n})\le (\omega \cdot j)\cdot 2\}\supseteq \{n:d(x,{v}_{n})d(x,{u}_{n})\le \omega \cdot j\}\in \mathcal{\mathcal{F}}$$
So, the lefthand set is in
$\mathcal{\mathcal{F}}$
, in contradiction to our previous conclusion that
$\mathbf{u}$
satisfies ( 11 ) with
${v}_{n}$
replaced by
${u}_{n}$
. We can conclude that
$\mathbf{u}$
and
$\mathbf{v}$
are in different 1galaxies
${\Gamma}_{a}^{1}$
and
${\Gamma}_{b}^{1}$
respectively, with
${\Gamma}_{a}^{1}$
closer to
${\Gamma}_{0}^{1}$
than is
${\Gamma}_{b}^{1}$
.
We can now repeat this argument with
${\Gamma}_{b}^{1}$
replaced by
${\Gamma}_{a}^{1}$
and with
$\mathbf{u}=\left[{u}_{n}\right]$
playing the role that
$\mathbf{v}=\left[{v}_{n}\right]$
played to find still another galaxy
${\Gamma}_{a}^{1}\prime $
different from
${\Gamma}_{0}^{1}$
and closer to
${\Gamma}_{0}^{1}$
than is
${\Gamma}_{a}^{1}$
. Continual repetitions yield an infinite sequence of galaxies indexed by, say, the negative integers and totally ordered by their closeness to
${\Gamma}_{0}^{1}$
.
The rest of the proof continues just like the argument for Theorem 4.2 that establishes a sequence of 1galaxies progressively further away from
${\Gamma}_{0}^{1}$
than is
${\Gamma}_{b}^{1}$
. In this case, the natural number
$m$
is replaced by the ordinal
$\omega \cdot m$
; also, the last
$n$
in ( 3 ) is replaced by
$\omega \cdot n$
.
$\square $
Finally, by mimicking the proof of Theorem 4.3, we can prove Theorem 11.3. Under the hypothesis of Theorem 11.2, the set of 1galaxies of
${}^{*}{G}^{1}$
is partially ordered according to the closeness of the 1galaxies to
${\Gamma}_{0}^{1}$
.
12 Extensions to Higher Ranks of Transfiniteness
The extension of these results to the enlargements of transfinite graphs of any naturalnumber rank is quite similar to what we have presented. The ideas are the same, but the notations and the details of the arguments are somewhat more complicated. Moreover, further complications arise with the extension to the arrow rank
$\stackrel{\u20d7}{\omega}$
of transfiniteness. Extensions to still higher ranks then proceed in much the same way. All this is explicated in the technical report [
7]
, which can also be found in the archival web site, www.arxiv.org, under Mathematics, Zemanian. References

A. Abian, The Theory of Sets and Transfinite Arithmetic, W.B. Saunders Company, Philadelphia, Pennsylvania, 1965.

C. Berge, Graphs and Hypergraphs, North Holland Publishing Co., Amsterdam, 1973.

R. Goldblatt, Lectures on the Hyperreals, Springer, New York, 1998.

R.J. Wilson, Introduction to Graph Theory, Academic Press, New York, 1972.

A.H. Zemanian, Transfiniteness for Graphs, Electrical Networks, and Random Walks, BirkhauserBoston, Cambridge, Massachusetts, 1996.

A.H. Zemanian, Graphs and Networks: Transfinite and Nonstandard, BirkhauserBoston, Cambridge, Massachusetts, 2004.

A.H. Zemanian, The Galaxies of Nonstandard Enlargements of Transfinite Graphs of Higher Ranks: II, CEAS Technical Report 820, University at Stony Brook, Stony Brook, NY 11794, April 2005.